SVX Engine cooling "Again & Again"

[Dessertrunner]

Okay guys I am going to be a pain in the A__ and stat this all over again. My reason is that at this time I don't know anywere in the world that someone has an SVX engine that stays cool at high revs. Right now we have blown 1 engine due to heat and my friend Steve with the buggy has blown 2. If we asked around I am sure the total will be over 10 engine that have either been blown completly or just head gaskets.

What I have found recently is the following:-
- Multiple pass radiator "DID NOT" slove the problem.
- A bigger Radiator "DID NOT" slove the problem (buggy radiator is 3 times bigger then my PBR).
- During the recent desert crossing I notice on my 3 tempreture guages that the water going into the top of the radiator was at 90C the water coming out of the radiator was at 70C but the dash temp was overheating, above 115C. Didn't check at the time but after noticed cool had spillied from the overflow tank.

So the big question is what do we now know.
The only way my over heating issue can happen is that the left head was overheating as that is were dash & ECU temp guage is. It is also possiable that is the water boiled in the left head the pressure would go up and would cause the coolent to overflow. This can still happen when the radiator is cool because tempreture and pressure behave differently.

Now we are cutting the cross pipe and will post photos of the job in the next couple of days.

Will keep you posted.
Tony

[michael]

This may sound stupid but have you tried water wetter?

[Dessertrunner]

Not sure that it will help with cooling but if you can explain how I would appricate it.
Tony

[SilverSpear]

Water Wetter (http://www.batterystuff.com/engine-cooling/RLWW.html)? Is that it?

I guess the overheating EG33 problem under an extreme condition outweighs by far the issue of adding a coolant.

Something has to be made concerning the issue of boiling water in the block.

[michael]

This one http://www.redlineoil.com/product.aspx?pid=74&pcid=10
Only reason I would ever suggest it is because of the quality and experience I have had with their other products. If it can lower the temps 20 degrees that is nothing to sneeze at.

[SilverSpear]

Michael, this product might be great and all, but I don't think anyone doing boost or high profile N/A would bet on the reliability of his +$6000 setup over a $10 bottle of coolant.

That might be added to tapping or whatever enhancement is done in the cooling process.

We will be waiting on Tony and Tom's project developments to see what they would come up with.

[michael]

I agree, just thought it might be something to experiment with. I'm not going to pretend I know what I'm talking about on this subject.

[Wikedjuggalo]

Wasn't their discussion somewhere of a issue with the cooling channels with in the block. Specifically the back of the block boiling the coolant almost? IT had to do with a forced injection but could possible be an issue for you?

[Trevor]

My reason is that at this time I don't know anywere in the world that someone has an SVX engine that stays cool at high revs.
What I have found recently is the following:-
- Multiple pass radiator "DID NOT" slove the problem.
- A bigger Radiator "DID NOT" slove the problem (buggy radiator is 3 times bigger then my PBR).

So the big question is what do we now know.
The only way my over heating issue can happen is that the left head was overheating as that is were dash & ECU temp guage is.

Will keep you posted. Tony

Tony, trying to apply some crude logic to your previous results.

Engine and therefore pump speed, is a deciding factor according to recorded fact. I have previously mentioned speed of water flow and others have suggested possible impeller cavitation. As far as I am aware this factor has not been properly tested, whereas there is every reason to suspect this as an issue.

Reducing the pump efficiency is not difficult and involves at the worst, only the cost of a used component. The time honoured method is to drill substantial holes through the rotors.

You appear to have concluded that only the left head is overheating, but it can not be concluded that this is confined to the particular area of the engine associated with the sensors. :confused:

Speak to some old timers who have used the flat head Ford V8 for racing.;)

You have the fortitude and will find the answer, Trevor.

[SomethingElse]

Reducing the pump efficiency is not difficult and involves at the worst, only the cost of a used component. The time honoured method is to drill substantial holes through the rotors.


Would you suggest drilling every impeller? or every other one?

[TomsSVX]

I have been granted permission to run a test on Shotgunslade's engine. I have modified the X-over manifold for the water outlet. I am going to run this modification alone w/o the aux. cooling ports in the heads to find out if this is helping the overheating issues by itself. If not, I will hook up the aux. cooling ports and work from there. After seeing the minuscule volume of the inside of the manifold I can only imagine that the flow through it is limited and I would imagine the limit is right around 6500 engine RPMS and thus the high RPM engines that are experiencing the issues. Lets see what we find and work from there.

I can honestly say that impeding the flow is not going to help anything except allowing it to overheat on both sides of the engine rather than just the one restricted at this point

Tom

[Dessertrunner]

Trevor,
I am to dumb to have concluded or come up with the idea its the left bank I have only comfirmed it as highly likly. The info that supports this argument are that the newer 6 cylinder Subarus have 2 return pipes to the radiator making me think they found the left bank wasn't getting enough cooling. Second a UK rally guy sent my buggy friend a photo of how they have moded the top pipe on the 4 cylinder so the water flows more evenly from the heads because they had been killing engines on the left bank.
Tony

[Dessertrunner]

Cooling pipe that we are modify, Tom suggestion.
Tony

[TomsSVX]

http://i65.photobucket.com/albums/h225/TomsSVX/0906091503a.jpg

Tom

[Dessertrunner]

Nice job Tom, have you blocked the heater pipe etc that return to the pump?
Tony

[TomsSVX]

yes that will be part of the system. We will be running a small electric pump for the heater core alone

Tomq

[Dessertrunner]

That pretty smart were did you find it.
Also I have this crazy idea which is a bit complicated but involves puting a second radiator infront of the first. You send the hot water from the top of the engine into the radiator closest to the engine out the bottom and into the front one then back to the inlet of the water pump. Its means you can reduce the temp of the coolent down lower with on the return then you can the same amount of water. Works on the princpal that coolest air hits cooler coolent and hotter air hits hottest coolent.
Its know in industry as counter current cooling. In our factory we send water into a sweet corn cooler at 5 c and product at 100c at the other end we get out water at 100c and corn at about 10c. This is done with equal volumes of both.
If we don't run AC condensor the other radiator can go there.
Tony

[TomsSVX]

I am going to have to say no to the twin radiator idea. Its not realistic to add it to the system. It is not the problem at hand. Lets address the current issues we know, then worry about if the radiator is sufficient which technically speaking should be good for twice the power levels anyone using them is reaching

Tom

[Dessertrunner]

The difficulity is the current radiator is okay at "What coolent flow rate". My suggestion only applies if we can't get the flow rate up that's going through the engine. X water will carry Y BTU of heat minius the inlet temp.
Tony

[TomsSVX]

I have to ask, there is not that great of a difference in the cooling of this engine with many others of similar hp/liter designs. Why would we need excessive cooling while others do not. We just need proper cooling. Address the internal flaws and I can almost guarantee that it will operate very well

Tom

[Dessertrunner]

You may be right Tom I have not compared the amount of flow to other engines. Just the other 6clyinder Subaru which has a heap more coolent flow per min. What engine have you looked at the flow of.
Tony

[TomsSVX]

haven't payed much mind to the flow. I will be chosing a 55gpm pump for Dan's engine. With the supporting X-over modification I see no reason it should not be able to use most of the pump's capacity. lets see what we get before we make any more decisions

Tom

[Dessertrunner]

Is the new water pump with the existing pump in or out?

[TomsSVX]

the pump housing will be there to direct flow into the block but it will have no impeller and a single idler will take the place of the drive pulley

Tom

[SilverSpear]

YT, it might be a dumb question from my part, you mentioned about tapping the heads to eliminate the boiling water issue in the block. Why not tapping the water passage directly in the block instead where that boiling water is?

The only answer I could find is water flow? Not to disrupt the water flow? :confused:

[shotgunslade]

Two comments:

1. I have been runing Water Wetter for years and it may help, but doesn't cure the problem. I have also been running 80% water, 20% anti-freeze to increase the specific heat of the coolant, and that hasn't cured the problem.

2. Jack Laverty has been running an electric fuel pump for some time and still has overheating problems, so this likely rules out impellor cavitation at high rpm's as the only cause of internal boiling.

[TomsSVX]

YT, it might be a dumb question from my part, you mentioned about tapping the heads to eliminate the boiling water issue in the block. Why not tapping the water passage directly in the block instead where that boiling water is?

The only answer I could find is water flow? Not to disrupt the water flow? :confused:

Because if I pump it directly into the block, it will not make much of a difference because that is where the coolant is being sent in the first place. Sending it to the head allows coolant that has been cooled by the rad to be sent into the hot stream to help it be more efficient in its job of thermal absorbtion before it hits the hot spot.

Like I said though, I doubt it will this considering how small the X-over manifold really is inside. I truly believe it is simply too small to move enough water to be efficient for the LH side of the motor

Tom

[SilverSpear]

Because if I pump it directly into the block, it will not make much of a difference because that is where the coolant is being sent in the first place. Sending it to the head allows coolant that has been cooled by the rad to be sent into the hot stream to help it be more efficient in its job of thermal absorbtion before it hits the hot spot.

Like I said though, I doubt it will this considering how small the X-over manifold really is inside. I truly believe it is simply too small to move enough water to be efficient for the LH side of the motor

Tom

Hope this X-over method works as you expect Tom, but just in case it doesn't... will it make sense to tap the left block (not head), connect it to an electric pump which works automatically at a certain temperature then to a small outside rad (bike rad?) then again to the right block?

This way you will be transferring the boiling water to the rad for cooling then inject it again to the right side.. am I making sense?

[Dessertrunner]

Dan I agree with Tom the first approach needs to be the simplest first. We need to exhaust all possiabilies before we go for things like taping in the back of the head or a second radiator such as I suggested. The big drama about tapping into flows a different positons is it might just move the problem to another spot.
I think we all agree that the problem is coming from the left back from lack of water.
Tony

[SilverSpear]

Definitely Tony, I am with you guys concerning the idea of X-over mod. But this is plan A.

I was suggesting we should also set plans B & C just in case the overheating issue is not fully eliminated with plan A.

[svxistentialist]

After seeing the minuscule volume of the inside of the manifold I can only imagine that the flow through it is limited and I would imagine the limit is right around 6500 engine RPMS and thus the high RPM engines that are experiencing the issues. Lets see what we find and work from there.

I can honestly say that impeding the flow is not going to help anything except allowing it to overheat on both sides of the engine rather than just the one restricted at this point

Tom

The info that supports this argument are that the newer 6 cylinder Subarus have 2 return pipes to the radiator making me think they found the left bank wasn't getting enough cooling. Second a UK rally guy sent my buggy friend a photo of how they have moded the top pipe on the 4 cylinder so the water flows more evenly from the heads because they had been killing engines on the left bank.
Tony

You may be right Tom I have not compared the amount of flow to other engines. Just the other 6clyinder Subaru which has a heap more coolent flow per min. What engine have you looked at the flow of.
Tony

I'm just going to make a general comment here about this problem, rather than a specific suggestion for a solution. I'll be watching for the success of what you are doing.

Do we all remember the development car picture with Twin Turbo on the rear corner? And we know that Subaru never supplied the SVX with a manual gearbox, stating that a manual gearbox would not handle the torque, which we long recognise was BS.

Based on the problems we are having here I'm inclined to believe that Fuji engineers were aware of the poor cooling, and the car was released without a manual to keep revs down, and without the turbos because they would add heat.

Note that instead of developing the EG 33, they next produce the EZ30 which in some ways is less advanced than the EG was. As Tony notes above though, the newer 6 cylinder has a better grasp of cooling the block.

Maybe this inhibiting problem was there and recognised by them all along, and they just moved on to a different block to get around it?

Just a thought

Joe

[SVXRide]

YT,
Have you checked to see if the modified pipe will fit under the intake? I haven't had the time to check to see what the clearance is with the stock pipe.
-Bill

[Dessertrunner]

Your point is interesting Joe, what I have noticed is that if I am driving the car at a constant speed but have it in 4th gear (6 speed box) when I should have been in 6 th gear the engine runs hotter. The moment I change to 6 th the tempreture of the engine drops. This is part of the reason that I agree with Tom on the top pipe. Somewere in the system when you increase flow the restriction in the left bank must get worse. Clearly there has to be a flow problem some were.
I am going to jump in at this point before I get every one coming out and saying this is a clear indication of cavitation of the pump. I AGREE THEY COULD BE RIGHT but it is best to stop this problem by reducing the restrictions in the pipe work and heads to enable the pump to handle higher speeds.
To explain at my work we have 3 by 500hp diesel pumps that each pump 300l of water per second. If we increase the speed on them to pump 400lps the suction pipe feeding the pump is to small and we get cavitation at the pump. This happens inspite of the suction pipe being 600mm in diameter. To solve the problem we changed the suction to 800mm and the problem went away.
Sorry to get off topic but I want to make the point we still have a lot of things to do such as increase the suction pipe from the radiator before we try to reduce the flow of the pump. It may turn out that we have to do that as well later but not now.

[TomsSVX]

I'm just going to make a general comment here about this problem, rather than a specific suggestion for a solution. I'll be watching for the success of what you are doing.

Do we all remember the development car picture with Twin Turbo on the rear corner? And we know that Subaru never supplied the SVX with a manual gearbox, stating that a manual gearbox would not handle the torque, which we long recognise was BS.

Based on the problems we are having here I'm inclined to believe that Fuji engineers were aware of the poor cooling, and the car was released without a manual to keep revs down, and without the turbos because they would add heat.

Note that instead of developing the EG 33, they next produce the EZ30 which in some ways is less advanced than the EG was. As Tony notes above though, the newer 6 cylinder has a better grasp of cooling the block.

Maybe this inhibiting problem was there and recognised by them all along, and they just moved on to a different block to get around it?

Just a thought

Joe


Joe I had mentioned this to Bill a while back and I think i posted it. The manual gearbox was not chosen because there was not one able to handle the torque but rather dispense it properly. Subaru did not want to put a basic 5mt into the car, they wanted an automatically controlled center differential. Since they did not have one at the time and considering their overwhelming investment in the car already, I think they planned to put it aside for a later model depending on the success/failure of sales.

The engine development was based on the GT platform which the car was being designed for. Granted I am sure they were aware of the cooling issues and this is why the RPM redline is so low. We all know now that with some better headwork the engine loves to spin

Tom

[svxistentialist]

Joe I had mentioned this to Bill a while back and I think i posted it. The manual gearbox was not chosen because there was not one able to handle the torque but rather dispense it properly. Subaru did not want to put a basic 5mt into the car, they wanted an automatically controlled center differential. Since they did not have one at the time and considering their overwhelming investment in the car already, I think they planned to put it aside for a later model depending on the success/failure of sales.

The engine development was based on the GT platform which the car was being designed for. Granted I am sure they were aware of the cooling issues and this is why the RPM redline is so low. We all know now that with some better headwork the engine loves to spin

Tom
Good point Tom and I accept that. I'm just not too convinced that all that's required is headwork. I do believe that fluid flow in the block is restrictive, and that this is an innate problem that you guys are currently tackling.

You have my respect and full attention.

:)

[TomsSVX]

no I meant valve/cam work. I think the cooling issue is restricted to the X-over manifold alone. I am looking forward to proving it in the coming weeks

Tom

[svxistentialist]

no I meant valve/cam work. I think the cooling issue is restricted to the X-over manifold alone. I am looking forward to proving it in the coming weeks

Tom

Well I am hoping that you do, and if you do I will be


:)

[Dessertrunner]

I have asked for price on a flow meter as I want to find a way to measure the flow of water through the block at different rev's. My guess is that the manual says two rates at two different rev but I don't beleive it. Clearly the higher revs is just double the flow of the lower rev. This doesn't make sense as friction would cause the flow to be non linear.
Tony

[oab_au]

This is an interesting problem.
There is nothing wrong with the cooling system when the car is used as it was intended. Hot days, full AC, slow traffic, never worries it, it obversely has the capacity to keep the engine cool under all conditions.

When it is run at higher engine rpms for an extended period, at has a heat problem. The problem seems to be with the left hand bank.

There is one pump, and one radiator inlet. If we look at the two systems, they are not equal. The outlets are different, the right bank has an outlet that is connected directly to the radiator inlet, the left bank passes through the crossover pipe to connect to where the right bank flow is exiting the head, so the left has to overcome the right flow, to pass to the radiator.

http://www.subaru-svx.net/forum/attachment.php?attachmentid=12802&d=1253158806

It would seem that at the higher engine speed, the increased pump flow causes the cooling flow to take the path of least resistance, flowing through the right bank straight to the radiator. The higher capacity from the right head prevents the left flow from overcoming its resistance so it slows down, to overheat.

Trevor touched on the problem with the older flathead V8, and Tony’s mention of the UK twin outlet 4s points to the solution, of changing the way the left blends into the right outlet, if it did not have to overcome the flow from the right side it would flow the same as the right.

Harvey.

[Trevor]

Trevor,
I am to dumb to have concluded or come up with the idea its the left bank I have only comfirmed it as highly likly. The info that supports this argument are that the newer 6 cylinder Subarus have 2 return pipes to the radiator making me think they found the left bank wasn't getting enough cooling. Second a UK rally guy sent my buggy friend a photo of how they have moded the top pipe on the 4 cylinder so the water flows more evenly from the heads because they had been killing engines on the left bank.
Tony

Tony,

Be very sure that I did not consider that you are in any way "dumb", hence my words ---
"You appear to have concluded that only the left head is overheating, but it can not be concluded that this is confined to the particular area of the engine associated with the sensors."

Thankfully you have now confirmed that there can be no doubts in this area, which was in fact my reason for my comment.

All involved are now reporting with absolute certainty, that only the left side of the engine is overheating. On this basis the problem can as a first consideration, involve the split in coolant distribution.

Under no circumstances, should it be taken for granted that the design engineers were infallible. ;)

[NiftySVX]

First let me say that I always thought the crossover pipe was a less than ideal design. I agree completely that it is possibly a restriction at higher than intended RPMs at extended periods as thought of by an engineer. We have to remember the engineering mindset. If you ask one if a cup is half full or half empty, they will tell you the cup is twice as big as it needs to be :lol: I am sure that during development testing they knew the limits of the cooling system, but since it works satisfactory with a stock setup, why make it flow more than necessary?

As for the manual transmission debate, I have always thought that the "too much torque" was a rumor, because Fuji has the capability and know how to machine stronger gears if they needed to at not much additional expense. The true reason for it that was communicated to me by a Subaru rep that was around back then told me that he remembered seeing it in writing somewhere that The SVX would not be offered in a manual transmission because it was meant to showcase the "Active" all wheel drive and they did not have a manual with an active system until the DCCD system came around.

And third, THERE IS A PICTURE OF A EG33 WITH TURBO(S) ON IT DURING DEVELOPMENT??? where??? :eek:

[dynomatt]

Tom, I assume you're using the Meziere 55gpm pump? I had found the BMW one which flows 150lpm (or about 40gpm) but wondering whether the Meziere one is better.

I notice you'll be doing what I was thinking of doing...the complete remote setup with the original mechanical pump more or less gutted...to direct flow basically.

What sort of current draw are you anticipating?

[TomsSVX]

I forget what it draws to be totally honest, I knew when I purchased it. I think it is somewhere in the realm of 15-20amps

Tom

[SilverSpear]

This is an interesting problem.
There is nothing wrong with the cooling system when the car is used as it was intended. Hot days, full AC, slow traffic, never worries it, it obversely has the capacity to keep the engine cool under all conditions.

When it is run at higher engine rpms for an extended period, at has a heat problem. The problem seems to be with the left hand bank.

There is one pump, and one radiator inlet. If we look at the two systems, they are not equal. The outlets are different, the right bank has an outlet that is connected directly to the radiator inlet, the left bank passes through the crossover pipe to connect to where the right bank flow is exiting the head, so the left has to overcome the right flow, to pass to the radiator.

It would seem that at the higher engine speed, the increased pump flow causes the cooling flow to take the path of least resistance, flowing through the right bank straight to the radiator. The higher capacity from the right head prevents the left flow from overcoming its resistance so it slows down, to overheat.

Trevor touched on the problem with the older flathead V8, and Tony’s mention of the UK twin outlet 4s points to the solution, of changing the way the left blends into the right outlet, if it did not have to overcome the flow from the right side it would flow the same as the right.

Harvey.

Harv, you are completely making sense to me. I am really convinced more by what you are saying than what is currently being done. If there is a flow restriction, that restriction might be at the level of the tube leading to the rad. If you check the red box below, you will see that it is way thinner than the rest of the X-over. Here is where the right bank flows easier and the left bank is facing resistance (what you were saying above).

http://img197.imageshack.us/img197/6707/43147896.png (http://img197.imageshack.us/i/43147896.png/)

What Tom welded was the middle part which instead of being round or squared, subaru techs "crunched" it to make it flat for clearance purposes... but it has the same flow rate. I am not generalizing this, I am just assuming it.

Again, Harv your logic is making a lot of sense especially when comparing it to V8's (same concept, two heads) and I guess in the most ideal situation we would be having independently two pumps, two flows and two rads... but we all know that this is really hard to achieve.

[Trevor]

I have asked for price on a flow meter as I want to find a way to measure the flow of water through the block at different rev's. My guess is that the manual says two rates at two different rev but I don't beleive it. Clearly the higher revs is just double the flow of the lower rev. This doesn't make sense as friction would cause the flow to be non linear.
Tony

Tony,

I agree with your thinking. It could it be that the specs. refer to the pump running free of any output constriction, a rather useless figure I agree, but likely if produced by the pump designer. Testing the flow rate left/right, as you propose, would be a bugger of a problem mechanically. All indications are such, that at this point, you can surely presume imbalance.

The placing of the sensors, would make one suspect that the designer, was aware that the left bank remained at a disadvantage, due to the hairy dual outlet arrangement. This was no doubt dictated by space restrictions, which is also a real problem involving any modification.

Any difference in restriction between the two legs of the split system will result in an imbalance in flow rate. The resistance can be anywhere within each the circuit, or at the outlets. The right has the throttle body in parallel and the left the heater. In special applications, it would be as well to close these auxiliary circuits. However the major problem to be overcome, is the unequal resistance due to the water pipe in the left leg.

The out of balance, is caused by unequal restriction which results in back pressure. The pressure difference results in a difference in flow. The unequal denominator is pressure, not as has been suggested, flow. The cart is not before the horse.

Laws dictate that an increasing common pressure, against several outputs of unequal resistance, will result in an increasing difference between the outputs. Hence the increasing problem at high RPM. Unfortunately the effect is not linear.

The easiest way to balance multiple resistive circuits, is by adding resistance rather than amplification. This is particularly so in this instance, where the problem is at the high pressure end of the system, where as a result adequate flow will exist.

It should be possible to fit a restrictive blanking plate, under the right inlet to the water pipe. A relatively soft alloy sheet and some gasket goo, a gasket or whatever could seal against the pipe, depending on the arrangement of the O-ring.

Tom has been working on distribution within the head space, and a plate could also be used to alter the flow relative to the two minor ports, if this might help.

Best of luck for all those involved in the project.

Trevor.

[Trevor]

Trevor touched on the problem with the older flathead V8, and Tony’s mention of the UK twin outlet 4s points to the solution, of changing the way the left blends into the right outlet, if it did not have to overcome the flow from the right side it would flow the same as the right.

Harvey.

The problem as suggested has no relevance with the flat head Ford V8. This had twin water pumps and complete circuits, so that pressure difference between the two banks was not a problem. In those days the stylists did not dictate available engine space. :rolleyes:

This not in fact a matter of one flow overcoming another flow, it is a matter of two unequal pressures, presented with a single outlet. ;)

[SilverSpear]

The easiest way to balance multiple resistive circuits, is by adding resistance rather than amplification. This is particularly so in this instance, where the problem is at the high pressure end of the system, where as a result adequate flow will exist.

It should be possible to fit a restrictive blanking plate, under the right inlet to the water pipe. A relatively soft alloy sheet and some gasket goo, a gasket or whatever could seal against the pipe, depending on the arrangement of the O-ring.

Tom has been working on distribution within the head space, and a plate could also be used to alter the flow relative to the two minor ports, if this might help.

Best of luck for all those involved in the project.

Trevor.

Trevor, with all due respect to your thinking and judgement, and we all know that your reasoning is without doubt logical. But wouldn't you think that for a high performance EG33 you should be shooting the other way around? Instead of pausing restriction on the "higher" flowing end, why don't you amplify the "less" flowing end? Your performant EG33 will require more cooling in that sense.

The problem as suggested has no relevance with the flat head Ford V8. This had twin water pumps and complete circuits, so that pressure difference between the two banks was not a problem. In those days the stylists did not dictate available engine space. :rolleyes:

This not in fact a matter of one flow overcoming another flow, it is a matter of two unequal pressures, presented with a single outlet. ;)

This is what I am thinking about, a system where you can have twin water pumps without any resistance or restrictions, each cooling one bank and both of them equal in performance to the OEM pump. This way you will cause cooling balance inside the engine.

Trevor, can you show us a diagram about this V8 you are talking about? Or is it general to all Ford V8's?

[Dessertrunner]

For the recod the big book says 100lpm @ 3,000rpm and 200lpm @ 6,000rpm.

I have cut my manifold as you can see from the image the next plan if that doesn't fix the problem will be to put a new pipe inthe centre were the ruler is. That pipe will be 40mm as opposed to 30mm it is now. Next change the water pump inlet pipe from 30mm to 45mm if possiable. Also both inlet and outlet on the radiator changed to suit

I feel sure the inlet and out lets of the motor are both 30mm will cause a problem. Centrifugal pumps always have a larger inlet then outlet to prevent caviation.
Great the way everyone is jumping on board I think if a few of us run tests we will beat this problem finally. Have a great day or night.
Tony

[Trevor]

Trevor, with all due respect to your thinking and judgement, and we all know that your reasoning is without doubt logical. But wouldn't you think that for a high performance EG33 you should be shooting the other way around? Instead of pausing restriction on the "higher" flowing end, why don't you amplify the "less" flowing end? Your performant EG33 will require more cooling in that sense.

Restricting the right side by a quite small amount, will result in the left side being increased. An adjustment of balance is involved, not a simple overall reduction. Evidence does not suggest a lack of flow and the opposite could be the case.

Coolant circulating too quickly, was the accepted diagnosis, in respect of the old flat head V8 Ford. This I know because in the early 1950s, I did very well in club competitions, using one of these engines in an AC sports car of 1935 vintage. Incidentally, a fan was not used and the car also served as family transport.:D

This is what I am thinking about, a system where you can have twin water pumps without any resistance or restrictions, each cooling one bank and both of them equal in performance to the OEM pump. This way you will cause cooling balance inside the engine.

Trevor, can you show us a diagram about this V8 you are talking about? Or is it general to all Ford V8's?

There is absolutely no way, anything resembling the arrangement used on the Ford could be replicated. For a start there is not room for separate outlet tubes for each side. A quick look will show why the designer had problems with space. As I pointed out, this is exact reason for the existing set up.

The Ford dual pumps were far too efficient and the cure was to reduce this quite radically. Drill all the rotor blades, or knock off every other one. Cast iron made them easy to break. Water at high pressure and full of air, is not an efficient coolant. Recesses are passed over, pockets are formed and hot spots result.

Cheers, Trevor.

[Trevor]

For the recod the big book says 100lpm @ 3,000rpm and 200lpm @ 6,000rpm.

I have cut my manifold as you can see from the image the next plan if that doesn't fix the problem will be to put a new pipe inthe centre were the ruler is.

The restriction appears to be at junction of left and right. I can not see the enlargement you have made making much difference. If you could bring out a pipe in line with the ruler, that would be another story. You will have to explain how you will create the clearance required.

P.S. The fact that a flow has been stated for 6,000 RPM, is surely a sign that cavitation within the pump is unlikely.

Keep at it. :D

[SilverSpear]

For a start there is not room for separate outlet tubes for each side.

This would have been great if otherwise... :o

[Sov13t]

Why not run a bypass in the cross over pipe so that the left hand stream enters the flow of the right post headers...

or not even mix them at all until they hit the rad with a Y like junction? Radiator Tomyx?

-Sov

[SilverSpear]

Why not run a bypass in the cross over pipe so that the left hand stream enters the flow of the right post headers...

or not even mix them at all until they hit the rad with a Y like junction? Radiator Tomyx?

-Sov


The position of the radiator inlet favors the right side bank in terms of length and position. The left side would be toooo long besides finding a way to make it reach there and clearance issues. It would be ideal though if possible. :o

[Dessertrunner]

The fact that a flow has been stated for 6,000 RPM, is surely a sign that cavitation within the pump is unlikely.

I don't mean to sound like a non believer but centrifugal's pump curves don't double the flow when you double the speed. "Subaru has got slack and cooked the books".

By the way if you look at the manifold on the car water goes from the left bank throught the throttle bodie and in to the manifold on the right bank. The only way this can happen is if there is a pressure difference from one side to the other, eg a major restriction in the centre.
Tony

[Dessertrunner]

Did a quick calc and if the pipes go from 30mm to 42ID the flow will double.
Tony

[SVXRide]

Tony,
Have you confirmed that your modified pipe will still fit under the intake? If "yes", could you provide some dimensions on the pieces of Al you welded on?
Thanks.
-Bill

[Trevor]

I don't mean to sound like a non believer but centrifugal's pump curves don't double the flow when you double the speed. "Subaru has got slack and cooked the books".

My understanding in respect of a centrifugal pump, is that the flow rate is directly proportional to speed, all extraneous contingencies being equal, so that I do not doubt the figures published by Subaru. A curve is not involved but rather a straight line. Please qualify your opinion and advise where you have found contrary information.

By the way if you look at the manifold on the car water goes from the left bank throught the throttle bodie and in to the manifold on the right bank. The only way this can happen is if there is a pressure difference from one side to the other, eg a major restriction in the centre.
Tony

This relates to a situation where there are two passages in parallel, with one more restrictive than the other. The relative rate of flow will be in proportion with the resistance and that offered by the larger passage can be near zero. Only a very small flow will be required to heat the throttle body. There is no reason to construe that a significant restriction exists within the major passage.

[Trevor]

Did a quick calc and if the pipes go from 30mm to 42ID the flow will double.
Tony

The flow will not be increased by enlarging one section of the circuit. The most restrictive point in the circuit, will continue to throttle the overall flow. ;)

[Dessertrunner]

Trevor,
Would you accept that as flow goes up pressure has to increase as the restriction beomes a bigger issue. That said then all things are not equal and flow must be reduce in proportion to revs.
Tony

[Dessertrunner]

Here is a pump curve to explain

[Dessertrunner]

This artical provides some info as well.
Tony

[Trevor]

Tony, you are trying to catch me out. :eek:

The issue I covered was your suggestion that the Subaru figures are incorrect. The figures obviously are meant to apply to a pump free of any head. ;)

If the flow rate increases against a constant head, as a result of increased RPM, there will not be an increase in pressure, unless the head is increased, or a restriction is introduced. Therefore any restriction has a direct relationship with pressure. :)

You are saying that if the flow rate is increased (as a result of increased RPM), the head/restriction becomes more relevant, i.e. a bigger issue. This can be construed but it has nothing to do with the original subject, which was pump specification. :p

You were stating that the Subaru figures covering the pump are wrong. Their figures will not refer to the pump operating against a head, i.e. as per the Reliance Electric graph 1, which in fact proves my original point. :cool:

Graph 2 covers pressure and does in a way, relate to your suggestion regarding the importance of any restriction. However the graph shows this is may very well be less than you had reasoned. The slight curve shows that pressure increases at a rate dependent on speed, presumably with a constant head creating the pressure.

The pump curve you have shown, covers flow rate as against a variable head together with power consumption and has no real significance, unless one loosely relates power to pressure. :confused:

Back to square one, left head overheating, due to a restricted junction. The restriction is shown to be of less importance at high RPM, due to steep angle of graph 2. No problems at intermediate RPM. Why is not a small degree of restriction in the right leg, a valid remedy? :)

[oab_au]

Just to expand on my last post. The problem area is in the way the two outlet pipes from the engine join to flow to the radiator. As the pump speed increases, the flow out and the suction in, both increase.
With the standard system only the right bank has the suction acting on the flow so the right bank has the highest amount of the total flow. The left bank has very little suction, so it slows and overheats.

http://www.subaru-svx.net/photopost/data/503/medium/High_speed_.JPG

If the outlet pipe is modified to allow the left bank to flow independent of the right, it will have the pumps suction acting on its flow also. Although they still have different lengths, they should still both flow the same.

http://www.subaru-svx.net/photopost/data/503/High_speed_Modified.JPG

This will remove the restriction that left bank has, in trying to overcome the flow from the right bank. How the modification is done, depends on what clearance there is under the manifold. :)

Harvey.

[Dessertrunner]

Looks cool Harvey, I totaly agree with you. One question I am thinking over, do we acheive the same thing as you suggest if the top suction is taken from the centre of the top pipe as I positioned the ruler in the other image. Not much room under there.
You diagram is exactly as the UK rally guys do it so you are on the right track.
Tony

[Trevor]

The facts. ----

A centrifugal pump does not have the ability to develop significant negative pressure, “suction.” No negative pressure “suction,” exists within the system. The complete circuit is under the same pressure, held and limited by the radiator cap.

Pressure must be and is, virtually equalised throughout the system. Only the rate of flow can vary according to resistance within the circuit. Both left and right and left legs operate at the same positive pressure and can not become independent in this respect. Experience suggests that the greater flow resistance is in the left leg.

It would appear that the resistance occurs at the junction within the cross water pipe and the left is the most affected. It is not a matter of one flow trying to overcome the other.

Provided the revised junction overcomes the present unequal resistance, the modification could work, but most certainly not along the lines suggested. However where is the pipe work to be located, and fabrication of a convoluted structure, presents real difficulties.

In the event that resistance is unequal in some other section of the circuit, and this has not been ruled out, the suggested modification will not work under any circumstance.

All manner of hypothetical arrangements can be put forward, but the practical problem is space. If shifting the alternator and whatever is an option, whoopee do, let’s have a ball.

Balancing the circuit by restricting flow slightly in one leg, is a practical answer in every respect, and as an an experiment will cost bugger all, other than time. If time/money is not a problem, a flat but wide sheet metal fabrication, of equal left/fight section could possibly be made to fit. The sky is the limit. :lol:

[SilverSpear]

Trev and Harv, both of you are right.

Harv your diagram is the better solution to the problem on hand. Here is what must be done in this respect:

http://img190.imageshack.us/img190/1918/xovermodified.png (http://img190.imageshack.us/i/xovermodified.png/)

And I think we can TRY to do that with MODELING CLAY. We can just fabricate this route with modeling clay below the manifold and by trial and error check if we have a clear path for a metal fabrication. If not then I suggest Trevor's method by restricting flow on the right side a little bit to solve that junction problem.

http://img14.imageshack.us/img14/2668/xover.png (http://img14.imageshack.us/i/xover.png/)

[Trevor]

=Dessertrunner

I have cut my manifold as you can see from the image the next plan if that doesn't fix the problem will be to put a new pipe inthe centre were the ruler is.
Tony

If credits are to be issued, they should be directed where they are due. Harvey did not suggest a centre take off. This much better idea was proposed earlier by Tony, as above. :D

[SilverSpear]

Trev, either Tony, Harv or Sov13t, I really do not care as long as we get the job done. We are all here to discuss this flaw on our engines and resolve these flaws. I do not like to argue with people on this board over stupid stuff...

I really see it childish when people put in their signatures:

First bla bla bla modification to be done
First XXX whatchamacallit used on an EG33
First BS & BS & BS....

So what? He thinks he is Einstein now? What shall I put in my signature?
First complete bodykit on an SVX? Even though people see it as ugly (and now I do myself, that was 6 years ago, I was young and stupid) it will still be the first complete bodykit... but I don't care! :barf:

There is no doubt that YT, yourself, Tony and Harv are all extremely knowledgeable people, please keep yourselves proud on a professional level and do not slip youselves on bull sh!t...

[Dessertrunner]

With the graze of god and a lot of luck we may have one of these moded pipes on a motor that is in the racing buggy tomorrow. So with any luck we should know if it is a total waste of time or an major improvment.
By the way in case I forgot to mention if your AC is off it is impossiable to run fans flat out. They will only run slow and medium.

[shotgunslade]

I think we are on the cusp of solving this problem and ought to stop throwing darts at one another about certain details and about specific authorship. We are sort of a team here, even if somewhat prickly. Here is what we know about the problem.

1. It isn't about slow steady increasing coolant temperature registered on the gauge which would suggest a simple inadequacy of the radiator to reject the heat from the engine. Instead it is characterized by a a sudden spike in temperature that suggests a steam bubble passing the sensor. This symptom suggests localized boiling somewhere in the coolant system. Note that if the problem were simply that overall water flow was inadequate or heat transfer surface of the radiator was inadequate we would get a steady slow rise in coolant temperature, rather than a quick spike that would just as quickly subside. Note also that this phenomenon is often accompanied by burping water into the overflow tank, further suggesting steam generation in the system.
2. Localized boiling could arise from two sources, a. cavitation at the entrance to the water pump, or b. some portion of the block that locally has inadequate flow so that it overheats causing local boiling.
4. I was an early proponent of cavitation, but I hadn't thought it through. The coolant system on an automobile is not a truly open system. It is a semi-closed system. The radiator cap causes it to operate at a pressure of approximately 2 bar rather than 1 bar as on a completely open system. Therefore, if the coolant temperature at the inlet to the water pump is approximately 100C, then the negative pressure at the pump inlet would have to be 1 bar relative to system pressure in order for localized boiling to occur. That is very unlikely.
5. Jack Lavety's use of an electric water pump doesn't solve the problem. Given that it would have a different inlet configuration than the stock water pump suggests that the problem isn't cavitation.
6. The experience from OZ suggests that the left side of the block is the overheating culprit. YT's observation of the unexpectedly small size of the opening in the crossover pipe exposes the likely cause of that overheating because of reduced flow across the left side of the block.
7. Trevor is correct that the easiest way to balance current flow in the 2 legs of a parallel circuit (it's easier to think about this using the electrical analogy) is to add resistance to the leg with the smaller resistance, however that will add to the overall resistance of the circuit, requiring greater voltage (pressure from the water pump) to get the same amount of current (coolant flow). A better course is to reduce the resistance on the leg with the larger resistance to the extent easily achievable and then afterwards, increase resistance on the other leg to balance current flow.
8. Remember that the current (water flow) balance doesn't have to be perfect, only that the flow through the high resistance leg (left cylinder bank) is adequate to avoid local wire meltdown (coolant boiling).

So let's cheer for those who are pursuing decreasing the resistance on the left cylinder bank (Dessertrunner and YT) and wait to see their results. In the end they may need add some flow resistance to the right bank or maybe not.

Finally, lets all congratulate ourselves for getting to the root of the problem. Remember at the end of the day, no one will remember exactly who contributed exactly what detail to the final solution. Furthermore, no idea from a single author will make it thru the process unchanged. The solution will be a group effort and the basis of congratulation should for supporting the overall effort of the group.

By the way, Trevor, we need to have a discussion about the differences between positive displacement and centrifugal pumps. This is what I do for a living and you somewhat mischaracterized the suction characteristics of centrifugals.

[RallyBob]

5. Jack Lavety's use of an electric water pump doesn't solve the problem. Given that it would have a different inlet configuration than the stock water pump suggests that the problem isn't cavitation.


Agreed. But it did help a bit. He went from overheating 4-5 laps into a race to eventually (after changing the radiator, higher psi cap, electric pump, and surge tank) overheating 17-18 laps into a 22 lap race. So progress was made, although the problem is far from solved. It probably doesn't help one bit that during the progression of cooling system changes he has continuously made power advances and rpm increases, which certainly makes isolating cause and effect difficult.

He has not run the latest configuration changes at the track, which includes the machining work at the front of the block (water pump inlet).

The best part seems to be that now the heat rises far more slowly, rather than instantaneously. And dropping the rpm shift points from 8200 to 6500 or so lets the engine cool off now, compared to before when a 6500 rpm peak was enough to boil over within 5 laps!

Anyway, I think that with all the information being thrown about here between Tony, Trevor, Harvey, Tom, etc....the cause and perhaps the solution is not too far away. I'm just sitting back and observing for now:D.

[oab_au]

For the recod the big book says 100lpm @ 3,000rpm and 200lpm @ 6,000rpm.

I have cut my manifold as you can see from the image the next plan if that doesn't fix the problem will be to put a new pipe inthe centre were the ruler is. That pipe will be 40mm as opposed to 30mm it is now. Next change the water pump inlet pipe from 30mm to 45mm if possiable. Also both inlet and outlet on the radiator changed to suit

I feel sure the inlet and out lets of the motor are both 30mm will cause a problem. Centrifugal pumps always have a larger inlet then outlet to prevent caviation.
Great the way everyone is jumping on board I think if a few of us run tests we will beat this problem finally. Have a great day or night.
Tony

Yes Tony the suggested center tap will do the same thing as the Y join, if it will fit in.

We have to keep in mind of what we are doing here. We are trying to cure a problem that is caused by another problem.
If you are stuck with using the original pump, then any of the rerouting, or restricting the right bank will help to overcome the problem, it really depends on how much work you want to do.

If in the case of a race engine, then the root cause has to be fixed. That is still too large a flow, flowing too fast. An electric pump will remove the problem of too fast a pump speed, but not if you use a BIG pump that pumps too much too fast, you are back to the same problem.

This is a case of " the bigger number, is not the best". Water is a bad conductor of heat, it takes time to heat it, and time to cool it. If the water flows too quick it will not have time to remove its maximum amount of heat from the engine block, and will not have time to loose it to the air through the radiator metal.
The boundary layer will heat and cool, but the main flow in the middle of the stream will charge straight through, unchanged.

The engine was designed to operate at 3000 to 4000 rpm, all day long. At this speed the pumps flow is about 120 lpm, this is probable the flow rate that provides the best transfer of heat. We really should be trying to replicate this flow, to restore the system to the ideal flow regardless of the sustained engine speed.

I think the best we can do is to do the Y or center tap change, and use an electric pump of about 120 lpm.

Harvey.

[Dessertrunner]

I think there are a number of "DON"T KNOWS" which is also a concerns and a number of "DON'T UNDERSTANDS". For example:-
:- What is the real water flow rate at the different revs, friction in water is the square of which in simple terms means double the flow from 100lpm to 200lpm and the amount of friction goes up by 4 times. This means the opening up of the coolent flow pasages becomes more important such as Rally Bob suggested. We need to run some trial to figure this out.
:- If there is restrictions were are they and how do we find and fix?
:- Why does Subaru on most of there higher performance engines use oil coolers, correct me if I am wrong Rally Bob but I thought you guys installed one on the engine you built.
:- How we could orderly develop a test plan to conduct orderly trials to find the soluation.

Maybe the best way forward is to develop a list of things we need to know to fix the problem so as we are not clutching at straws.

Here is my first suggestion.
1 What are the coolent flow rates at different revs.
2 What is the tempreture the oil is running at and what should it be.

Tony

[Trevor]

Tony, good thinking. ;)

Restrictions:- These can be identified by eye and using common sense.

Oil Coolers:- An advantage on any engine often running or expected to run, at or close to maximum power. The lubricating oil constitutes a direct method of reducing temperature in the bearings and the bottom end in general. Normally the sump/pan serves this purpose.

1&2. The actual figures are not important, but could br useful guide lines. The objective will be to simply make improvements, whatever.

Cheers, (Home brew in hand at this moment.:D) Trevor.

[Trevor]

Yes Tony the suggested center tap will do the same thing as the Y join, if it will fit in.

We have to keep in mind of what we are doing here. We are trying to cure a problem that is caused by another problem.
If you are stuck with using the original pump, then any of the rerouting, or restricting the right bank will help to overcome the problem, it really depends on how much work you want to do.

If in the case of a race engine, then the root cause has to be fixed. That is still too large a flow, flowing too fast. An electric pump will remove the problem of too fast a pump speed, but not if you use a BIG pump that pumps too much too fast, you are back to the same problem.

This is a case of " the bigger number, is not the best". Water is a bad conductor of heat, it takes time to heat it, and time to cool it. If the water flows too quick it will not have time to remove its maximum amount of heat from the engine block, and will not have time to loose it to the air through the radiator metal.
The boundary layer will heat and cool, but the main flow in the middle of the stream will charge straight through, unchanged.

The engine was designed to operate at 3000 to 4000 rpm, all day long. At this speed the pumps flow is about 120 lpm, this is probable the flow rate that provides the best transfer of heat. We really should be trying to replicate this flow, to restore the system to the ideal flow regardless of the sustained engine speed.

I think the best we can do is to do the Y or center tap change, and use an electric pump of about 120 lpm.

Harvey.

I have been pushing the speed of the coolant flow for quite a long time, over years in fact, with negative response. :( It is good that Harvey and I agree on this, and more so emphasis this as a being valid for early consideration. :cool:

There is no necessity for changing the pump with all the subsequent alterations. Reducing the efficiency will do the trick and is a well proven technique. It will be pointed out that this will also effect flow at all levels of engine speed, but at lower RPM the loading and heat delivery is less intense, and has not in other applications proven a problem.

Speed of flow is now identified as possible problem, therefore some restriction in the right leg will assist rather than detract. Direct cost is nil.

Drilling or reducing pump the rotor is easy and the cost can not exceed the value of a used pump.

P.S. Flow can also be reduced by fitting a restrictor within the thermostat housing. This is worth a try as a first experiment, as so little is involved. However the pump mod. provides a reduced performance somewhat in line with RPM.

As an experiment the result of an increased flow can be ascertained by doing without the thermostat. Some have done this and wondered why a negative result was the outcome, and this incidentally proves the point regarding water possibly flowing too fast.

The combined labour cost, is nowhere near any of the other suggested methods. This combination should be top of the list, even if considered as only an experiment. However it is very important that each alteration should be done and tested separately for individual evaluation, in line with correct experimental practice. ;)

[SilverSpear]

I think there are a number of "DON"T KNOWS" which is also a concerns and a number of "DON'T UNDERSTANDS". For example:-
:- What is the real water flow rate at the different revs, friction in water is the square of which in simple terms means double the flow from 100lpm to 200lpm and the amount of friction goes up by 4 times. This means the opening up of the coolent flow pasages becomes more important such as Rally Bob suggested. We need to run some trial to figure this out.
:- If there is restrictions were are they and how do we find and fix?
:- Why does Subaru on most of there higher performance engines use oil coolers, correct me if I am wrong Rally Bob but I thought you guys installed one on the engine you built.
:- How we could orderly develop a test plan to conduct orderly trials to find the soluation.

Maybe the best way forward is to develop a list of things we need to know to fix the problem so as we are not clutching at straws.

Here is my first suggestion.
1 What are the coolent flow rates at different revs.
2 What is the tempreture the oil is running at and what should it be.

Tony

Tony, I still remember a thread talking about the water pump flow at different levels. Tried to look for it, but no luck so far, I will keep trying.

But what I remember is that the maximum flow that the stock pump is capable of was 320lpm... I think someone was trying to compare it with the electric pumps or something...

I still remember clearly three figures, of which one of them was the 320.

[Dessertrunner]

Dan I think I am the one that wrote it and the 320lpm refered to the current 6 cylinder.
tony

[shotgunslade]

Trevor and Harvey:

I am sorry to say that on the one thing you agree on over these many years, you are wrong. For heat flux across any fluidic heat exchanger, the issue of variation of velocity and mass flow on either side of the heat exchanger involves three important interrelated variables. These are the residency time, that you have referenced as being critical, the other two are the thermal resistance of the boundary layer and the logarithmic mean temperature differential between the fluids.

If you look at the graph of heat transfer across the exchanger with variation of mass flow on one side, what you will always see is a drop in total heat transfer with the drop in mass flow. You will probably see a bump along the way as the fluid flow transitions from turbulent flow to laminar flow. In that bump, heat transfer will drop much faster than the rate at which water flow is decreasing. As the mass flow increases, heat transfer will approach an asymptotic value. What happens is that, as mass flow decreases, the temperature of the fluids leaving the exchanger achieve a closer approach. That approach is measured differently depending upon whether the heat exchanger is counterflow, parallel flow or cross flow. This closer approach reduces the logarithmic mean temperature differential across the exchanger, reducing total heat flux.

We often use air to water heat exchangers with very low face velocity on the air side to get a very close approach of the temperature of the leaving air to the entering water (counterflow exchanger), however, in no way does the reduction in air flow result in greater heat flux.

As you increase the water flow from a very low value, two things happen. The first is that slowly the boundary layer at the water side of the coil erodes, provding a closer approach of the exterior radiator tube temperature to that of the water inside. The second thing that happens is that the temperature drop of the water decreases as it goes through the radiator. The water has less residency time, so each unit mass of water loses less heat. But, remember that the mass flow rate is increasing, so does the mass flow rate increase faster than does the drop in heat loss per unit mass of water? The answer is yes and here's why.

Look at the radiator from the air side. The air doesn't know anything about how much water is flowing on the other side. It just knows that it is passing a hot surface and is getting warmed up. With constant air flow and air temperature, the only variable affecting heat transfer is the temperature of the radiator. Look what happens when water flow rate is reduced. The temperature of the water leaving the radiator is colder, so the logarithmitc mean temperature differential of the radiator is reduced, and the boundary layer has a greater resistance, further decreasing the wall temperature of the radiator. From the standpoint of the airside, reduction in water flow means a colder radiator, meaning less heat transfer.

So you may say this is only theory, but I can show you an unlimited number of instrumented coil tests that show that reduction in water flow for a water to air heat exchanger results in reduced total heat flux.

On the engine side, the temperature differential betwen the combustion chamber and the coolant is very high, but if you follow the same logic, you will see that increasing the mass flow rate of the water will increases the temperature differential of the water as it flows through the engine, decreasing the logarithmic mean temperature differential between the hot bits of the engine and the fluid. Thuis would tend to decrease the heat transfer to the fluid. But the heat generation of the engine can be considered to be constant at a given operating condition. In order to make up for the fact that the average temerpature of the water going through the engine is higher (due to less water flow), the engine will run hotter.

Conclusion, more coolant flow through the engine results in a cooler engine, in most circumstances, and that is a good thing.


I have been pushing the speed of the coolant flow for quite a long time, over years in fact, with negative response. :( It is good that Harvey and I agree on this, and more so emphasis this as a being valid for early consideration. :cool:

There is no necessity for changing the pump with all the subsequent alterations. Reducing the efficiency will do the trick and is a well proven technique. It will be pointed out that this will also effect flow at all levels of engine speed, but at lower RPM the loading and heat delivery is less intense, and has not in other applications proven a problem.

Speed of flow is now identified as possible problem, therefore some restriction in the right leg will assist rather than detract. Direct cost is nil.

Drilling or reducing pump the rotor is easy and the cost can not exceed the value of a used pump.

P.S. Flow can also be reduced by fitting a restrictor within the thermostat housing. This is worth a try as a first experiment, as so little is involved. However the pump mod. provides a reduced performance somewhat in line with RPM.

As an experiment the result of an increased flow can be ascertained by doing without the thermostat. Some have done this and wondered why a negative result was the outcome, and this incidentally proves the point regarding water possibly flowing too fast.

The combined labour cost, is nowhere near any of the other suggested methods. This combination should be top of the list, even if considered as only an experiment. However it is very important that each alteration should be done and tested separately for individual evaluation, in line with correct experimental practice. ;)

[dynomatt]

This has been posted before, but I'll put it up again.

EG33 pump performance from manual

@ 760rpm = 20l (5.3US gal) per minute
@ 3000rpm = 100l (26.4US gal) per minute
@ 6000rpm = 200l (52.8US gal) per minute

From 8 impeller vanes at 76mm diameter.

Exactly the same performance as the EJ22 which presumeably generates 33% less heat...that's not right surely. Also, coolant capacity is 7.0l in the EG33 and 7.2l in the EJ20 turbo, from the same pump, that doesn't have coolant problems.

EZ30 out of the 2004 spec Legacy manual

@ 5500rpm = 320l (84.5US gal) per minute

From 6 impeller vanes at 73.2mm diameter. Cooling capacity 7.8l.

I'd love to get the same specs from the 3.0 out of the RB.

From this...one could ascertain that Subaru believed a higher flow made for a better result. Clearly more capacity also plays a role.

Honda S2000...arguably a high performance NA flows 176l per minute at 6000rpm...but it's a 2.0l four cylinder

Matt

[oab_au]

Well Dan if you are OK with the increased water flow, not being a problem, it is your engine. :) We only have to fix the water outlet junction, to cure the problem.:)

Harvey.

[Trevor]

Trevor and Harvey:

I am sorry to say that on the one thing you agree on over these many years, you are wrong. ---------------- Conclusion, more coolant flow through the engine results in a cooler engine, in most circumstances, and that is a good thing.

As I see it, your theory is based solely on mass coolant flow and that the logic in respect of cooling equates exactly with that for heating. No account is taken in respect of the vast difference in relevant surface areas. The large coolant area / cooling air area incorporated in a radiator, or heat exchanger, does not equate with heating factors as exist within an engine.

What is also important is that it does not appear that a factor covering residency time, in relation contact with the heating/cooling areas is included within your explanation. Flow speed/flow rate against restriction, producing pressure within the heating cycle, but not the cooling phase, must also be considered.

At this point not able to agree with Harvey, I can not accept your explanation. ;)

By the way, I am awaiting with interest, a reply to my PM regarding centrifugal pumps.

[shotgunslade]

Actually I did refer to residency time when I said that we often use very low face velocity on a coil to achieve a very close approach of the leaving air to the entering water temeprature. The phenomenon that enables that to occur is increased residency time.

The most important thing about a heat exchanger is that there is a heat balance on each side of the exchanger. Residency time for the water enables the water to achieve a closer approach temperature to the air. The only impact that water residency time has on heat transfer to the air is that it lowers the logarithmic mean temperature difference between the air and the water.

On the airside, heat transfer can be chacterized by Q = U * A * (Delta T) where Delta T is the logarithmic mean temperature difference between the air and water. U is the heat conductance across the heat exchanger. A is the area of exchange surface. Of course, the lower the water mass flow through the heat exchanger the lower is the leaving water temperature, and thus, the lower is the average temperature of the water and the lower is average temperature differential between air and water, and thus the lower is the total heat transfer.

It is very simple physics. It is the same on the water side as it is on the airside. You can't tell me that when the car is overheating, you can ever improve the problem by reducing airflow through the radiator.

[svxistentialist]

Trevor, I think Dan's explanation on heat "flux" is more about heat transfer in an air-water interface, this seems to be quite a correct analysis to me but misses the point that is relevant to the problem here.

It's not about how fast the rad can lose heat [ I'm not talking about degrees heat here, I'm talking about mass of heat] it is more about how evenly we can ship heat from both the left and right block of cylinders.

So we have deduced the right block ships heat and gets by OK, but the left block does not. The left block would be expected to be a mirror image of the right, but we now know that heat is not shipped away from the left block as fast or as effectively as from the right. More particularly this becomes a problem at sustained higher revs, when consistently more heat is generated.

Dan I accept your analysis concerning heat flow, in particular the relevance of heat dumping over an air water interface if the mass of coolant circulating per minute is increased. What is most at issue here is the rate of heat loss is different between the two cylinder banks before the coolant gets to the rad. More cooling fluid is circulating in the right bank compared to the left bank. Trevor's suggested solution is to restrict fluid flow in the right bank in order to promote better circulation and equalise heat loss in both cylinder banks.

This is totally different from reducing the fluid flow through the rad as you seem to be arguing against. The presumption is that the same amount of heated fluid will pass through the rad, losing the same amount of heat, but restricting flow in the right cylinders [that currently "own" the path of least resistance to the pump] will allow increased circulation of fluid and corresponding increased collection of and dumping of heat from the problematic left bank of cylinders will keep the heat transfer from the two sides of the engine balanced.

The laws of physics in relation to fluid dynamics, heat transfer and laminar flow will apply here regardless. Our mission here is to ensure that all heat generated in either bank is uniformly and evenly shipped via the pump or pumps to a radiator sufficiently efficient at losing this heat to the passing air movement.

The above sounds long winded, but that's the only way I can describe the concept of even heat movement we need. Please forgive me if it sounds pedantic.

Joe

[shotgunslade]

Joe:

I absolutely agree with you. The left cylinder bank is getting shortchanged on coolant flow, possibly because of the restriction that YT found. My previous post concerning balancing current in parallel circuits directly addresses that issue.

7. Trevor is correct that the easiest way to balance current flow in the 2 legs of a parallel circuit (it's easier to think about this using the electrical analogy) is to add resistance to the leg with the smaller resistance, however that will add to the overall resistance of the circuit, requiring greater voltage (pressure from the water pump) to get the same amount of current (coolant flow). A better course is to reduce the resistance on the leg with the larger resistance to the extent easily achievable and then afterwards, increase resistance on the other leg to balance current flow.
8. Remember that the current (water flow) balance doesn't have to be perfect, only that the flow through the high resistance leg (left cylinder bank) is adequate to avoid local wire meltdown (coolant boiling).


Adding some resistance on the right bank after you have done everything you can to reduce restriction on the left bank might somewhat reduce overall flow, but it could cure our problem by shifting some flow from the right bank to the left bank.

[SVXRide]

Dan,
thanks for taking me back to Junior year Heat Transfer and Fluid Mechanics classes. Haven't been there for a while ;):lol:
-Bill (BSME, class of '83)

[Johnybeas]

damn engineers.... blah blah blah just figure it out already so us dum dums who don't live breath and eat calculus and physics can be happier ;) :p<3:drool:

[Trevor]

Trevor, I think Dan's explanation on heat "flux" is more about heat transfer in an air-water interface, this seems to be quite a correct analysis to me but misses the point that is relevant to the problem here.

It's not about how fast the rad can lose heat [ I'm not talking about degrees heat here, I'm talking about mass of heat] it is more about how evenly we can ship heat from both the left and right block of cylinders.

So we have deduced the right block ships heat and gets by OK, but the left block does not. Joe

Joe,

The radiator was only brought into the picture, as a means of comparison in order to prove a theory, but the point is still open to discussion, having not been covered. :confused:

The point at issue, was my suggestion that the speed of flow, can have a negative effect in respect of the transfer of heat within the engine. There is and has been no suggestion that the radiator is at fault. The question of balancing the flow in the two legs, is also separate issue.

We must stop this ring a ring a roses approach, and stick to each individual relevant point, as it comes up. ;)

Cheers, Trevor.

[svxistentialist]

Joe,

The radiator was only brought into the picture, as a means of comparison in order to prove a theory, but the point is still open to discussion, having not been covered. :confused:

The point at issue, was my suggestion that the speed of flow, can have a negative effect in respect of the transfer of heat within the engine. There is and has been no suggestion that the radiator is at fault. The question of balancing the flow in the two legs, is also separate issue.

We must stop this ring a ring a roses approach, and stick to each individual relevant point, as it comes up. ;)

Cheers, Trevor.

The radiator was brought into the picture in Dan's two posts #78 and #82, and I directly addressed that matter in my opening two paragraphs. Dan was theorising on how much heat would be lost if the flow rate to the rad was reduced.

I don't agree with your hypothesis that this thread is a "ring of roses". This thread is working excellently as a think tank, different people, engineers and otherwise, putting forward possible solutions.

It is educational and it is a model for owners as to the process for thinking through problems and applying solutions. It is surprising what progress can be made when people collaborate to find solutions, instead of the petty bickering about points that sometimes goes on here.

Joe :)

[SilverSpear]

Now I have read everything in this thread, understood all points of views except Dan's. I can do that though since myself I am a science/finance person but really this will take time to grasp and google for the formulas and stuff... I will do that later :rolleyes::D

I was looking at the pictures of Sov13t's engine and I see the water taps in the right and left blocks. They seem identical in diameter. Now so far, I think the weakness in our engines is also the water pump, I think it is small for that engine (accoding to Matt).

Above to what YT and Tony are doing, I think we should look for someone who can modify the water pump with more impellers and we need to PORT the left block to make way for more coolant flow.

PS: Tev, I know that you would prefer to make the right tap smaller, or put something to limit the flow though...

Am I making sense? :confused:

http://img5.imageshack.us/img5/5414/eg33.png (http://img5.imageshack.us/i/eg33.png/)

[shotgunslade]

Trevor, I did address the engine side.

On the engine side, the temperature differential betwen the combustion chamber and the coolant is very high, but if you follow the same logic, you will see that increasing the mass flow rate of the water will increases the temperature differential of the water as it flows through the engine, decreasing the logarithmic mean temperature differential between the hot bits of the engine and the fluid. Thuis would tend to decrease the heat transfer to the fluid. But the heat generation of the engine can be considered to be constant at a given operating condition. In order to make up for the fact that the average temerpature of the water going through the engine is higher (due to less water flow), the engine will run hotter.


I absolutely agree that low coolant flow through the left cylinder bank seems to be our culprit for local boiling. Overall, for nearly stock engines, the stock radiator or the PWR radiator seem to be adequate to maintain the average coolant temperature at an acceptable value. Hopefully my new engine built by YT will address the low flow issue on the left cylinder bank that we suspect as the culprit for our problems.

[Dessertrunner]

The lastest is,
we put the modified top pipe on the buggy but it didnt help the cooling, Steve, the guy with the buggy felt his problem was the Motec not doing the timing right at high revs thus causing the engine to over heat. He has riped the Motec out and changed to a different unit.

His personal opinion is that the top cross over pipe will improve the engines cooling issue when he fixs the timing. This beleive is supported by the fact that the moded pipe is now a fixure on the buggy and not coming off.

This is were I am at :-
Still trying to get a price on a water flow meter so I can put solid numbers on what rate the water pump is putting out at the different engine revs.
Next challange is to get a correct left & right head tempreture, currently I have a coolent temp gauge on the outlet of the top pipe which gives the combined tempreture on both heads & I have the standard guage on the left head. The only way I think that I can get correct data is to drill into the aluminum head on each bank and insert a probe to measue the actual head tempreture this will be proposanel to the coolent temp.
After I get all this info I will then start the mods such as change the top pipe to see if I can reduce the difference is each head.

We have a saying at work and I think this is the time to bring it up:-
"Do we know this to be a fact or is it just an assumption as every time we ASSUME we know the answer and its not support by FACTS it bites us on the arse". I am afriad that I need some facts in the form of data before I can go any further with fixing the problem.

Have a great day.
Tony

[SilverSpear]

This is were I am at :-
Still trying to get a price on a water flow meter so I can put solid numbers on what rate the water pump is putting out at the different engine revs.
Next challange is to get a correct left & right head tempreture, currently I have a coolent temp gauge on the outlet of the top pipe which gives the combined tempreture on both heads & I have the standard guage on the left head. The only way I think that I can get correct data is to drill into the aluminum head on each bank and insert a probe to measue the actual head tempreture this will be proposanel to the coolent temp.
After I get all this info I will then start the mods such as change the top pipe to see if I can reduce the difference is each head.

We have a saying at work and I think this is the time to bring it up:-
"Do we know this to be a fact or is it just an assumption as every time we ASSUME we know the answer and its not support by FACTS it bites us on the arse". I am afriad that I need some facts in the form of data before I can go any further with fixing the problem.

Have a great day.
Tony

Tony... you've just became my HERO!

[dynomatt]

Danny, I hadn't seen the block without the crossover pipe on it, so thanks. I see they are exactly the same size (visually).

Why wouldn't we go to the option of dual outlets (and inlets in the radiator) like the EZ30? To me, the combination of that and the slightly different pump and capacities points to Subaru solving this problem for us?

I had a look...admittedly real estate under the manifold on the starter side (can't say left or right, or driver or passenger due to globalness) is complicated by vacuumn and fuel lines, but easily manageable.

I'm thinking this is pointing us in the right direction.

Matt

[Nevin]

After having read several pages of information and conclusions that I entirely agree with, I just have one very small suggestion (let's call it a drop in the bucket, if even that.)

A fellow I know who does engine competitions (Got #5 in '06 In the Jeg's Engine Master's challenge in "Popular Hot Rodding" magazine) mentioned to me that a lot of the V-8 guys are now going to reverse-flow cooling systems with electric water pumps. What this does is first cools the heads instead of the block. In our cars, it seems like this could possibly help in addition to the modified X-over pipe (which is a GREAT idea.)

It seems that a reverse flow system, in our case, would keep the temps down as well as releasing the restriction of the X-over being unequal. It's unequalness it seems, would not be as bad since it would be the side first being pressurized instead of if being the side where the coolant is being "sucked" from. Pardon my lack of being able to put my thoughts into mechanical terms, and I do realize there are a few other obstacles that would need to be overcome also to do this, but... has this crossed anyone else's mind, or am I just way out of the ballpark on this one?

[TomsSVX]

That is kinda the point of running two auxiliary 1/2" ID cooling lines to the cylinder heads directly from the pump.

Tony you are saying the modified X-over on the buggy did not help at all, made it worse, or helped a little?

Tom

[Nevin]

That is kinda the point of running two auxiliary 1/2" ID cooling lines to the cylinder heads directly from the pump.

Tom

In that case, wouldn't this be easier? Instead of running auxiliary lines, if you're already using an EWP, why not just reverse it to get the same cooling effect? Not to mention the extra power you could gain with a standalone because you could run more timing (since the head is cooler.) As I was informed, that was the biggest reason they V-8 guys started doing it.

[Trevor]

It is very simple physics. It is the same on the water side as it is on the airside. You can't tell me that when the car is overheating, you can ever improve the problem by reducing airflow through the radiator.

Apples/apples. The radiator is engineered for maximum flow of both fluid and air. The block is a complex convoluted combination of many port sizes; various restrictions; tight bends; blind spots; and parallel circuits are also involved. As flow speed and pressure is increased, so does the ratio of flow difference between open and restricted spaces. Open gain at the expense of restricted.

P.S. On the practical side, it has been shown that in the case of many cars, removing the thermostat in the hope of increasing flow and therefore cooling, has had the directly opposite effect.

[Dessertrunner]

Tom the mod is benficial in Steve opinion, its just that the ECU problem didn't give it a chance to shine. We need to test it in another way when do you think you will give your pipe go.
Tony

[Trevor]

That is kinda the point of running two auxiliary 1/2" ID cooling lines to the cylinder heads directly from the pump.

Tony you are saying the modified X-over on the buggy did not help at all, made it worse, or helped a little?

Tom

Tom,

You always view things from a practical aspect and this is an absolute must. Twin auxiliary lines as you suggest could provide a simple means of balancing as well as directing flow. As an experiment, hand valves could be included for adjustment. The thermostat housing lends to this end, via the heater circuit outlet.

No one has mentioned the two parallel circuits involved, i.e. the manifold and heater. These will have an unequal effect on each leg, much so in respect of the heater. I would presume that these are blanked off in the applications being discussed. However this should be confirmed, or otherwise should be given serious consideration.

Keep your thinking cap on, Trevor.;)

[TomsSVX]

In that case, wouldn't this be easier? Instead of running auxiliary lines, if you're already using an EWP, why not just reverse it to get the same cooling effect? Not to mention the extra power you could gain with a standalone because you could run more timing (since the head is cooler.) As I was informed, that was the biggest reason they V-8 guys started doing it.

H-6 cooling path = Block-->heads -->block --> radiator. It would not really matter much

Tom

[Dessertrunner]

Update,
Have purchased two temp senders that I can fit into the back of each head were the 20 mm plugs that Tom is talking about are. Once they are in we should be able to see some numbers on the tempreture in each head.
Coolent flow is proving more problmatic as the flow meter I want to use cost $700 so I need to go back to the start and find a cheaper unit. I will keep you all posted as we collect the data.
Tony

[SilverSpear]

Tony, can we help in sourcing the units for you at least used?
If you want provide us with details...

[Myxalplyx]

Just read this whole post! Lots of information. Very informative and technical. Since I'm the anti-technical.....here it goes.

Seems most people have been focusing solely on coolant flow, with the assumption that the left bank is flowing less and causes the car to overheat at sustained high rpm. Makes sense!...

Have we broadened our view to look at other factors that may come into play, which could cause the left bank to heat up? Here's a bone I'll toss out. The side that the left bank of coolant is supposed to cool. What cylinders is it cooling? Is it 1-3-5? Have we considered whether these cylinders are getting the proper amount of fueling from the injectors? If say...injector #1, #3 or the combination is not putting out the proper amount of fuel (or too much), could this simulate the same overheating of coolant that is being experienced?

Just throwing this out there since the cleaning of my Subaru's injectors (Old school and new) always resulted in #1 and/or #3 being worse than the rest, until I had them professionally cleaned. Increasing the coolant flow of the left bank may be like taking more headache medicine for the headache instead of removing the cause of the headache. Could only be prolonging the inevitable.

If this is far fetched, besides coolant flow, try to examine what else could simulate heating up of the coolant flow of the left bank, something other than just the flow.

I'll crawl back into my reading/lurk state and continue to learn now. L8Rs...

[Dessertrunner]

Help:-

During the week I contact 3 companies to find a flow meter to measure the coolent flow but all options have been to expensive, around $700. Any suggestion or equipment to use to measure the flow would be appricated. Have thought about weighing how much coolent is pumped in 5 minutes. This seems the cheapest overall as I have all the materials. The negitive will be that the water pump wont be pressure feed.
Thinking of plugging in the engine to a 31,000 litre tank to the engine water pump then the outlet from the engine into a tank on the scale. I fully understand that the result will understate the true flow but it will be better then what we have now which is "Zero" knowledge.
Tony

[Dessertrunner]

Here is a image of the top coolent pipe clearance.
Tony

[SilverSpear]

Tony, excellent job so far. Concerning the flow meter, what is your budget in US$ for that?

The tank method might be useful... to a certain extent. As you said getting approx numbers is better than getting nothing.

As for the coolant pipe clearance, do you think it will clear a stock manifold?

[Dessertrunner]

Budget a couple of $100. This manifold is the same hight as the standard.
Tony

[TomsSVX]

Just read this whole post! Lots of information. Very informative and technical. Since I'm the anti-technical.....here it goes.

Seems most people have been focusing solely on coolant flow, with the assumption that the left bank is flowing less and causes the car to overheat at sustained high rpm. Makes sense!...

Have we broadened our view to look at other factors that may come into play, which could cause the left bank to heat up? Here's a bone I'll toss out. The side that the left bank of coolant is supposed to cool. What cylinders is it cooling? Is it 1-3-5? Have we considered whether these cylinders are getting the proper amount of fueling from the injectors? If say...injector #1, #3 or the combination is not putting out the proper amount of fuel (or too much), could this simulate the same overheating of coolant that is being experienced?

Just throwing this out there since the cleaning of my Subaru's injectors (Old school and new) always resulted in #1 and/or #3 being worse than the rest, until I had them professionally cleaned. Increasing the coolant flow of the left bank may be like taking more headache medicine for the headache instead of removing the cause of the headache. Could only be prolonging the inevitable.

If this is far fetched, besides coolant flow, try to examine what else could simulate heating up of the coolant flow of the left bank, something other than just the flow.

I'll crawl back into my reading/lurk state and continue to learn now. L8Rs...

If indeed this were the case, it could be easily seen that an engine was not burning as well on the LH bank of the engine as it were on the RH. Unfortunately this is not the case and some are running SAEM which can and will compensate for said issues. The reality of the inejctors being dirtier is that they are first in line of the fueling system. Naturally anything moving past the filter will end up.... in the LH side injectors

Tom

[SilverSpear]

Budget a couple of $100. This manifold is the same hight as the standard.
Tony

Oh ok. I thought it is lifted because of the individual throttles.

This picture surely shows that we have a lot of space to do a Y system into the rad, just in case.

As for the flow meter, I am searching and sending emails to potential suppliers. Since it is Sunday, I guess we should expect something by tomorrow... our time :o

[Dessertrunner]

My friend with the buggy has tried changing the ECU from a Motec to EMS in an effort to solve the cooling problem as he beleived that the timing was out but it didn't solve the heating problem. I think he is at a point were he will pull the engine out and change to a different type. There is nothing he can do to keep it cool and now he is about to give up on the SVX engines.
Tony

[dynomatt]

You and I should make an offer on his engines...his current price is a bit high though.

M

[oab_au]

My friend with the buggy has tried changing the ECU from a Motec to EMS in an effort to solve the cooling problem as he beleived that the timing was out but it didn't solve the heating problem. I think he is at a point were he will pull the engine out and change to a different type. There is nothing he can do to keep it cool and now he is about to give up on the SVX engines.
Tony

I reckon we should take up a collection and buy you a Fuggan camera.:D

What has he done to the cooling system on the engine.?
What pump, and flow rate does it have?
What was the modification that he did to the outlet piping?
What is the radiator that it has.?
What rev range does it run?

Harvey.

[Trevor]

I reckon we should take up a collection and buy you a Fuggan camera.:D

What has he done to the cooling system on the engine.?
What pump, and flow rate does it have?
What was the modification that he did to the outlet piping?
What is the radiator that it has.?
What rev range does it run?

Harvey.

Harvey,

Just in case criticism hits the fan, for what it is worth, I will back you on this one. I am as you are, fully aware that it is easier to advise than do, but it is becoming obvious that the train has never been on track.

The thread clearly shows a lack of any sort of logical approach in curing the problem. Basic rules covering testing have not been carried out. So far there is nothing but a confused set of trials, often two at a time. No proper conclusions have been established from work to date. Incoherent fiddling is not science.

What stands out is that the cheap and simplest options, have not been tested. Even if not a complete success, these could prove valuable in providing definitive evidence and move things forward.

There has been talk of flow testing. As I see it, this will only be of value if a figure for right and left sections can be established. From a mechanical slant, this would appear a very difficult proposition. The actual information required relates only to, --- is the flow to high or too low, on one side only, or throughout. This can be established by testing rather than measuring.

First up, blank off the heater circuit at the pump, remove the thermostat, i.e. for increased flow. Note the results before and after. A means of measuring accurately any change in temperature will be valuable. However again, actual figures are not that important.

As always, sticking neck out, Trevor. ;)

[Dessertrunner]

I understand were you guys are coming from but I am the only guy that has a method to measure the return water temp and with in 3 day able to measure the head tempreture on each bank as well as return coolent from the bottom of the radiator.
As regards the buggy we have a race track on Steve farm and we measure how far the buggy gets before overheating. At this time I have been pushing him to measure return coolent temp but only now has he agreed that it might be an issue.
What we do know from trials we have done todate.
- Take out the thermstate as it overheats to soon otherwise.
- Overheating occur from about 5,000 rpm up. Below this it isn't a problem.
- If we are using a standard cross pipe ontop of the engine we have to return coolent to the pump through the heater pipe. This suggested the top pipe was a problem.
- The orignal radiator was over 3 times bigger then ours then it was changed to a multipass unit. This change improved to distance before overheating.
- We have tried a Craig water pump with the existing turbine removed and it was no good at all.
- Of the current pump options the orginal was best.
- Its not the ECU as we have tried two types, but both are having problems reading the cam timing at 6,000 rpm with the result that the engine is breaking up at high revs.

What we don't know but will is how well the airflow and radiator are getting rid of the heat generated before returning to the engine.

If I have forgot anything ask.

What I will be doing hopfully this week end :-
Fit dual temps to the heads.
Drive down the road at 110kph on a straight flat road for 100k measure outside tempreture.
Drive for 15 minutes in 6th then measure the temps on the three guages.
Repeat in 5th, 4th and 3rd while still driving at 110kph.

Graph the results, and I am guessing that as revs go up the left bank has a harder time getting flow due to the pressure restriction. If that is confirmed I will then change the top pipe over to the moded one (i have two of them) and re conduct the trail at the same outside air tempreture.

This is long winded but I am getting somwhere, I spent two much on my race engine to assume that it won't overheat so I intend to stay on the issue till it is solved. Have a great day guys and I hope you are all healthy and well.
Tony

[TomsSVX]

sounds like a solid plan. let us know how it goes

Tom

[Boxersix]

May I jump in here and just add some food for though......

-Loss coefficients of the coolant system in case?

-LC's of the pump diffuser/case coolant inlets?

-Relieve the case of micro bubbles at bores and vent to swirl pot(highly likely due to case passage design, compare 30/30r and porsche castings to EG casting ;) )? Have had some similar issues of this manner with some engine swaps(other chassis) where normal engine position is askew, causing localized collection of micro bubbles and ultimately reduce flow in specific jacket areas. Purging the case to a swirl pot eliminated the sitiuation in all instances. 4an(1/8th NPT tap) sufficed. Not completely sure of EG33 case thickness @ the bores, may require welding on a bunk prior to drilling.

Did not have this issue with the EG33 in the 96X2 prototype chassis, but then again the entire ME-RWD drivetrain in that car sat at about a 15 degree cant(motor low).




This picture surely shows that we have a lot of space to do a Y system into the rad, just in case.


Take into account that and equalized "y-pipe" collector off the heads in place of the OE crossover will not yield a balanced left/right system as the flow path of the coolant inside the case halves from the coolant inlet(pump) is not balanced. If you were talking about some other setup for the "y pipe" then disregard this response :p



Carry on with normal discussion :)

[oab_au]

-
If we are using a standard cross pipe ontop of the engine we have to return coolent to the pump through the heater pipe. This suggested the top pipe was a problem.
- Its not the ECU as we have tried two types, but both are having problems reading the cam timing at 6,000 rpm with the result that the engine is breaking up at high revs.


Tony

Tony I take it that it is still running the standard crossover pipe?
Or is it modded with the swollen middle?
Or is it modded where they join on the right bank?

The heater will always pass water, I would not say it shows a problem. As one end is to the top pipe, the other is to the inlet side of the pump. Then you have the Bypass pipe that does the same thing from the other end of the cross-over pipe.

These types of sensors are not good for high speed operation, because of the reluctance of the unit, they can end up with a constant voltage, that can't be read. Thats why they have changed to the Hall Effect sensor for crank pick-ups, a more reliable signal at speed.

Harvey.

[Dessertrunner]

Harvey we are running the swollen cross over pipe now, when you say the hall effect type are better what cars use them and were would I get one. I was thinking of getting a sensor from a Porsche as you would think it could handle it. I think Tom made the point that the one on the crank is working at 12 times per rev and still reading okay but the cam is only read 1 time every 2 revs. Or have I got the bull by the tail instead of the horns?
Tony

[SilverSpear]

Tony, I have contacted two companies for you, I dunno if their products are applicable to car engines, pls check them out:

Liquid Controls (http://www.liquidcontrolsgroup.com/products/index.aspx)
Pathfinder Instruments (http://www.pathfinderinstruments.com/)

I am still waiting for their replies :(

[Trevor]

Take into account that and equalized "y-pipe" collector off the heads in place of the OE crossover will not yield a balanced left/right system as the flow path of the coolant inside the case halves from the coolant inlet(pump) is not balanced. If you were talking about some other setup for the "y pipe" then disregard this response :p Carry on with normal discussion :)

This is exactly correct. Difference in the flow resistance, relative to each side of the dual cooling path, will be the deciding factor, in spite of both outlets flowing equal and free. Correction must be achieved elsewhere.

Bubbles/air/turbulence as suggested, constitutes a very likely problem worthy of attention.

[Trevor]

Tony,

Thanks for the extra info. which will lead to more questions and suggestions. In this regard your continued patience is much appreciated.

“If we are using a standard cross pipe on top of the engine we have to return coolant to the pump through the heater pipe. This suggested the top pipe was a problem.”

This is a little difficult to get ones head around. The heater normally constitutes a direct first in line circuit, of limited capacity, outside of thermostat control. How do you use this as a return for coolant, and just how does this relate to the cross pipe?

Do these findings relate to tests, with or without the thermostat fitted?

“What we don't know but will is how well the airflow and radiator are getting rid of the heat generated before returning to the engine.”

As you advise experience with a radiator three times larger than standard, this would appear to provide an answer regarding the above.

Your determination to see the venture completed is great, and many are behind you. I guess this makes this thread/you a sticky. :)

Best of fortune, Trevor.

[Dessertrunner]

Danny They look great but my guess is they will be out of our price range.

Trevor I will attempt to cover your question is I miss one feel free to ask again.
Heater pipe being connected was tried with and with out thermostate. Putting the thermostate was bad news what ever else was done. The heater pipe goes from the left bank direct to the water pump inlet. I think the suggestion that this action giving a improvment would be because the total volume of coolent going through that head had increased there for the cross over pipe must have restricted the flow. This is best guess and needs to be confirmed by a different method.

As regards air flow its worth making a point. My wife noticed that every time we get stuck behind a truck on a hot day the radiator return coolent tempreture goes up. This happens even when the dash guage doesn't change. It has to be bad air flow.

I need to clear some misconceptions or confusion. Water flow in a pipe is greatly effected by friction. In simple terms the left bank and right bank can have the same friction at 3,000rpm. At 6,000rpm the friction on the Left bank could be greatly higher then the right which would mean that it is possiable that the Left bank is getting less coolent then when running at 3,000rpm. Water or coolent flow can reach a point that even if we doubled the pressure the flow would be lucky to go up by 10%. This princpal is used to make drip irrigation systems work around the world, they are know as pressure compensated drippers. If the SVX cooling system or even the cross over pipe acheive this level of friction then we have one hell of a challange in front of us.

Have a great day.
Tony

[SilverSpear]

Let's wait and see Tony, maybe they have an offer on a good meter...

This happened to me some time ago, I got something really nice for the SVX for a very very reasonable offer... :rolleyes:

[FriendlyTurkey]

Talk to Motec on a crank angle sensor. It shouldn't be hard getting a unit that will function well at an SVX's rpm range given you are simply using the CPS as a sinc signal.

Page 57:

http://www.motec.com.au/filedownload.php/MoTeC_Catalog%201.40.pdf?docid=2185

[oab_au]

Harvey we are running the swollen cross over pipe now,

OK then it still has the restriction for the junction of the two outlets, at the right hand outlet.

when you say the hall effect type are better what cars use them and were would I get one. I was thinking of getting a sensor from a Porsche as you would think it could handle it. I think Tom made the point that the one on the crank is working at 12 times per rev and still reading okay but the cam is only read 1 time every 2 revs. Or have I got the bull by the tail instead of the horns?
Tony

Most of the modern engines use the Hall Effect for the crank position sensors, usually fitted behind the crank pulley. We fitted one to the Hybrid Nissan engine, came off a later Holden engine.

Tom is right on the Cam sensor speed, but it is the type of sensor that is the trouble. Their output varies with speed, age, heat, contamination, and electrical interference (it should be using a shielded wire).

The output signal is a sine wave that is clocked on the 'positive to negative' zero crossing point. This is affected by the reduced amplitude of the voltage that will cause this point to be not as precise.

The Hall Effect has a constant square wave output, clocked on the 'positive to negative' change. It is unaffected by speed, interference, heat, or anything else, just a good clean signal. :)

Harvey.

[Trevor]

Danny They look great but my guess is they will be out of our price range.

Trevor I will attempt to cover your question is I miss one feel free to ask again.
Heater pipe being connected was tried with and with out thermostate. Putting the thermostate was bad news what ever else was done. The heater pipe goes from the left bank direct to the water pump inlet. I think the suggestion that this action giving a improvment would be because the total volume of coolent going through that head had increased there for the cross over pipe must have restricted the flow. This is best guess and needs to be confirmed by a different method. ----------------- Have a great day. Tony

Thanks Tony,

Not really sure what this means. --- “Putting the thermostat was bad news what ever else was done.” Guess it should be “pulling”.

Which was best? Thermostat installed and working, or removed for increased flow?

I do not understand the heater arrangement you describe as, --- “The heater pipe goes from the left bank direct to the water pump inlet.” Is this a modification, or as original? Very interesting in view of your left bank temperature problem.

The heater core on my JDM SVX is as per manual and forms a separate intact circuit, starting and finishing at the water pump housing. This puts it out of thermostat control and first in line to warm up at start up, which makes sense.

Research continues. ;) Trevor.

[dynomatt]

Most modern cars have the issue that if you remove the thermostat completely, water doesn't flow correctly which often leads to overheating. A better way is to drill more holes into the plate, or force it open.

[oab_au]

Thanks Tony,


I do not understand the heater arrangement you describe as, --- “The heater pipe goes from the left bank direct to the water pump inlet.” Is this a modification, or as original? Very interesting in view of your left bank temperature problem.

The heater core on my JDM SVX is as per manual and forms a separate intact circuit, starting and finishing at the water pump housing. This puts it out of thermostat control and first in line to warm up at start up, which makes sense.

Research continues. ;) Trevor.

Don't think this is right Trev. The heater is fed from the right hand side of the crossover pipe, and is run back to the inlet side of the pump. The By-pass pipe runs from the left hand side of the crossover pipe, back to the inlet side of the pump.

The Book shows the two hoses at the pump, as "heater inlet" and "heater outlet", but this is not right, as they would both be connected to the same point, and there would be no pressure difference to make it flow.
They should be labeled "heater return", and "by-pass return".

Harvey.

[Dessertrunner]

Remember guys the questions and answer refer to a buggy with a SVX engine in, so no heater core. Best with the thermostate out. Harvey the connection of the heater return pipe was beofre we put the new cross over pipe. At that time we blocked off the heater pipe so all the water went to the radiator.
Tony

[oab_au]

Remember guys the questions and answer refer to a buggy with a SVX engine in, so no heater core. Best with the thermostate out. Harvey the connection of the heater return pipe was beofre we put the new cross over pipe. At that time we blocked off the heater pipe so all the water went to the radiator.
Tony

OK, uderstood. I really would like a picture of the Buggy engine, save a lot of guessing.:D

Harvey.

[Dessertrunner]

Harvey you are right again I have expressed myself incorrectly, The pipe on the left bank goes direct back to the pump and the right bank one goes through the heater back to the pump in a normal car. What Steve did was return both these pipe to the pump which gave him an improvement. Hope that makes sense.
Tony

[Boxersix]

This is exactly correct. Difference in the flow resistance, relative to each side of the dual cooling path, will be the deciding factor, in spite of both outlets flowing equal and free. Correction must be achieved elsewhere.

Bubbles/air/turbulence as suggested, constitutes a very likely problem worthy of attention.

I brought this up because in my some 15 odd years of dealing with race motors in a host of one off chassis and configurations, I've fought with the problem more times than I can remember. Mostly due to putting motors in chassis positions that they were clearly not intended to run in. All motors run very hot and hard end up with some coolant "micro boiling" (for lack of better words) which results in tiny bubbles forming in the coolant mix. How well a motors internal cooling circuit is designed, as well as it's anchilaries(pump, exchangers, etc, etc), determine how well those bubbles are accumulated and purged off into the top of the system(ultimately the radiator cap and overflow reservoir). Poorly designed systems often have areas in the cooling jacket that accumulate but do not purge these

The EG33 coolant casting design is pulled directly from the EJ series motor design, and comparing pictures of the cases shows the similarities.

http://banilla.com/xanga/build/ej25pistonsb4.jpg

http://www.ecutune.com/images/P1020024.jpg

Comparing those to that of say a newer EZ30

http://blogs.cobbtuning.com/wp-content/uploads/2007/10/ez30r_pro1.JPG


What subaru did not change when adding the forward cylinders to the EG was the location of the coolant return ports on the top of the case halves. While I do not have a digitized format of an EG33 case to run through CFD, one can ascertain that the LC's through the cooling system are quite different between the two motors despite sharing quite similar cooling system. Subaru quite likely knew of this, but logically looked at cost of changing tooling/molds/dies for a different setup + the fact that it worked well within the limits the EG was designed for(4eat driven SVX) and went forth in the direction taken with the EG design. Within the EZ series casting, subaru took a different approach to the castings as well as overall cooling system design to likely correct the LC's of the motor and provide a more effective way of allowing the motor to purge.

(As a some what side tangent to this topic, take into account Porsches case cooling circuit design on all 986/996+ liquid cooled flat six's. Porsche took into account all LC's of the system and allow each cylinder jacket to purge directly into a main cooling channel prior to exiting the case to the heat exchangers.)

It's quite likely that the EG is suffering from a localize "vapor lock" so to say, in the cooling jacket of the forward two cylinders. This phenomenon has happened on other motor makes/models, so the possibility should be looking into. I never had this problem in the kit car chassis when the EG was powering it, however it's possible that the angle in which the drivetrain sat provided enough change in the system to allow the jacket to purge itself naturally, or at least better than it normally would. The motor sat in that car are about a 15 degree angle down, meaning that the nose of the motor was lower than the flywheel end of it. Typically in the subarus, the motor sits level to 1-2 degrees nose high pending on engine mount condition and other variables of each chassis.

My corrective fix for this issue on other motors was to tap the water jacket at the most likely place of vapor accumulation with an 1/8npt-4an fitting(in a welded bung) and direct that to a swirl pot into the cooling system. Even purging back to the exchanger inlet often worked w/o the need of a swirl pot but each application was specific in it's needs.


This is all subjective and may not even be the problem, so don't take this as gospel. While I have worked with the EG33 internally as a motor, I have not studied this cooling phenomenon you guys are running into on this motor because I simply didn't have the issue while I was testing with the powerplant( and testing was very very hard on course). However I must also state that I had twin, ground effect ducted heat exchangers for the motor alone which had a multitude more airflow than the front of the SVX could possibly achieve.


I really hope you guys figure this out. Sounds like a most displeasing scenario with your cars. Nothing worse to me in the auto field than track cars you can't run hard. My EG not resides in my 2002 WRX and will hopefully be running again soon. Being a daily driver /winter beater I'll still likely not run into this problem with the motor.

Good luck in any event! :)

[Dessertrunner]

We News:-

Fitted the tempreture guages to the back of the heads and to the inlet to the water pump. Even on idle the left hed heated up faster then the right by about 3C. The out side temp was 10c so not the ideal time to run the trial but drove at 6,500rpm then recorded what happened there was still a difference in the head temps. Will need the outside temp to be higher to test it better.
What I don't get is that at the same speed but higher revs the engine went up. Even the return water to the radiator shot up. My question is if the load is the same why the hell did all the tempreture climb, heads went up 10C and return temp up 25c.
Tony

[FriendlyTurkey]

http://www.youtube.com/watch?v=1_OXM4mr_i0&feature=related

[SilverSpear]

We News:-

Fitted the tempreture guages to the back of the heads and to the inlet to the water pump. Even on idle the left hed heated up faster then the right by about 3C. The out side temp was 10c so not the ideal time to run the trial but drove at 6,500rpm then recorded what happened there was still a difference in the head temps. Will need the outside temp to be higher to test it better.
What I don't get is that at the same speed but higher revs the engine went up. Even the return water to the radiator shot up. My question is if the load is the same why the hell did all the tempreture climb, heads went up 10C and return temp up 25c.
Tony

Well Tony it seems the EG33 needs some serious cooling scheme in order to cope with its heating. First of all I say a balance should be set between both banks in terms of internal cooling length and the upper outlet going into the rad. A Y system should be on the top of the engine with measurements equalling in sum the addition of the internal distance between both banks.

For that I think that internal flow should be measured along with the distance and the Y system should be designed with corresponding lengths accordingly.

Then YT's idea of Aux system going from the heads through a very long cooling trajectory in order for overheating water to cool properly. Boxersix might also have a valid point. I refered to tapping the block in my previous post instead of the heads since the general direction was going to the overheating issue coming from cavities in the block.
I dunno why but I am imagining using the Mercedes (92~93 S500 class) automatic gearbox external rad or some BMW one. Putting any of these in front of the condenser should definitely do some nice job.

Tell Steve not to give up on this engine... yet.

BTW, any update on the emails I forwarded to you?

[Dessertrunner]

Haven't been able to run trials yet as the outside tempreture has been to cold. The uints I got emails ore to expensive but will keep looking.

Wait till we learn some more before deciding to tap into the back of the head and installining a different radiator I feel sure that the answer is just around the corner.
Tony

[oab_au]

There is another side effect to the high pump speed increasing water output.

With the restriction at the junction of the left to right cylinder slowing the flow from the left side, it also alters the water pressure in the left side.

The water flow will take the path of least resistance and flow most of the extra flow through the right side. The left side is pushing against the high flow from the right, so not only does it slow to heat up, it also causes the pressure in the left to rise. If the right is at 14 psi the left may have risen to 16 psi. This will raise the boiling point of the water, so that bank will run at a higher temp. It may boil or it may not, but when you come off the track into the pits, the pump speed slows, the pressure drops from 16 psi to 14 psi.

The left now has water at a temperature that is over the boiling point of water at 14 psi, so it erupts into instant boiling. The temp gauge jumps between red and normal as the steam and water go by, and the overflow tank spews water out.

Sound familiar Dan?

Harvey.

[TomsSVX]

exactly my thinking at this point as well Harvey.

So, instead of piping more cool water in... How about helping it get out?

Two ports on the top of the block with adjustable ball valves to help balance the return from the engine to the radiator in combination with the modified manifold might do the trick here

Tom

[svxistentialist]

I agree with you chaps that localised boiling is part of what the circuit racers are experiencing and Harvey's pressure hypothesis has merit.

Another point touched on earlier is the fact that when bubbles are created they are not getting away or are accumulating and may be causing vapour lock in the liquid cooling circuit.

Have a look at the enclosed picture:

http://www.subaru-svx.net/photopost/data/562/Non_Horizontal_Engine.jpg

The SVX engine is tilted upward at the front by quite a few degrees. Yet we know that the coolant fluid connecting bar for connecting the two halves is at the back of the block.

Is it possible that bubbles when created are accumulating over time in the top highest point of the two blocks, particularly the left one? This might mean that any gas created is unable to vent to the expansion bottle and create a circulation problem.

What do you think?

Joe

[Trevor]

There is another side effect to the high pump speed increasing water output.

With the restriction at the junction of the left to right cylinder slowing the flow from the left side, it also alters the water pressure in the left side.

The water flow will take the path of least resistance and flow most of the extra flow through the right side. The left side is pushing against the high flow from the right, so not only does it slow to heat up, it also causes the pressure in the left to rise. If the right is at 14 psi the left may have risen to 16 psi. This will raise the boiling point of the water, so that bank will run at a higher temp. It may boil or it may not, but when you come off the track into the pits, the pump speed slows, the pressure drops from 16 psi to 14 psi.

The left now has water at a temperature that is over the boiling point of water at 14 psi, so it erupts into instant boiling. The temp gauge jumps between red and normal as the steam and water go by, and the overflow tank spews water out.

Sound familiar Dan?

Harvey.


The application involves a circuit supplied from one point, remote from the junction, and pressures within the circuit must at all time tend to equalise. Any flow between legs or branches within the circuit is dependant on resistance and time.

The low pressure side of any junction at a pressure lower than the input pressure to the circuit, must constitute an outlet.

In the event that one flow joins another and both are at the same pressure, but one is restricted by a resistance, the lesser flow will be assisted by the greater flow at a rate dependant on the resistance, rather than be obstructed, as has been suggested.

In the event that one flow joins another and are of unequal pressure, the lesser pressure flow will be overcome by the higher pressure flow, and that flow will be reversed at a rate dependant on the resistance presented, to a point whereby pressure is equalised.

[SVXRide]

So, if we were able to insert a restriction plate within the crossover pipe, shouldn't we be able to fine tune the overall system (have different plates, each with a different size opening)? I'm also curious about the use of a swirl tank and where we might be able to tap it into the system.
-Bill

[SilverSpear]

So, if we were able to insert a restriction plate within the crossover pipe, shouldn't we be able to fine tune the overall system (have different plates, each with a different size opening)? I'm also curious about the use of a swirl tank and where we might be able to tap it into the system.
-Bill

I was brainstorming this issue yesterday nightime, we definitely need a Y system there with TWO TAPS. Fine tuning could be done through these taps and with a water flow meter to get the exact same flow in both engine's banks. What do you guys think? :rolleyes:

[Dessertrunner]

More info,
On my way to work this morning I did a little trial to see what happened to the head temps. Left head was at 90 and right at about 87 traveling at 100kph outside temp 5c.
Decided to speed up and change down to 4 gear so revs went to 6,500k and speed to 140kph.
Interesting result the right head temp dropped to about 83-84 and the left head went up to 92-4. What interesting is that the outside temp was only 5c which is 5 above freezing.
I need a day were our temps get higher so I can do a proper test. I have a question.
What is the target operation temp we want the engine to run at and with the exist cooling system what temp will the coolent boil at?
Tony

[oab_au]

Boxersix and Joe raise the point of air bubbles in the system, causing trouble.

The problem of the coolant becoming aerated does happen, but not in this engine. If the system is purged when it is filled, there is no air in there and it can't get in.

If there are bubbles, they would have to be bubbles of steam or voids. Voids (bubbles of nothing) will disappear with pressure equalization. Bubbles of steam will disappear with temperature equalization.

The two outlets do come off the rear of the heads, but there appears to be a couple of bled holes forward of the outlet that would serve to remove any trapped air, when it is filled.

Harvey.

[svxistentialist]

exactly my thinking at this point as well Harvey.

So, instead of piping more cool water in... How about helping it get out?

Two ports on the top of the block with adjustable ball valves to help balance the return from the engine to the radiator in combination with the modified manifold might do the trick here

Tom

Boxersix and Joe raise the point of air bubbles in the system, causing trouble.

The problem of the coolant becoming aerated does happen, but not in this engine. If the system is purged when it is filled, there is no air in there and it can't get in.

If there are bubbles, they would have to be bubbles of steam or voids. Voids (bubbles of nothing) will disappear with pressure equalization. Bubbles of steam will disappear with temperature equalization.

The two outlets do come off the rear of the heads, but there appears to be a couple of bled holes forward of the outlet that would serve to remove any trapped air, when it is filled.

Harvey.

My point about bubbles, cavitation, steam, voids, whatever we call it, was in support of Tom's theory above and relating to your earlier hypothesis on pressure drop boiling, which Trevor dissed.

The whole system may be equalised by a relatively small bore interconnecting pipe at the top of the heads that would bleed any superheated gassy coolant back to the better circulating right side and vent any gas to the tank. With adjustable ball valves if they are a help.

Joe

[oab_au]

More info,
On my way to work this morning I did a little trial to see what happened to the head temps. Left head was at 90 and right at about 87 traveling at 100kph outside temp 5c.
Decided to speed up and change down to 4 gear so revs went to 6,500k and speed to 140kph.
Interesting result the right head temp dropped to about 83-84 and the left head went up to 92-4. What interesting is that the outside temp was only 5c which is 5 above freezing.
I need a day were our temps get higher so I can do a proper test. I have a question.
What is the target operation temp we want the engine to run at and with the exist cooling system what temp will the coolant boil at?
Tony

Tony the thermostat opens at 90*C, and the cap pressure is 14psi, so the water will boil at about 119*C.
For efficiency, the hotter the engine runs the better, as it looses less heat energy to the cooling.

Harvey.

[Trevor]

My point about bubbles, cavitation, steam, voids, whatever we call it, was in support of Tom's theory above and relating to your earlier hypothesis on pressure drop boiling, which Trevor dissed.

The whole system may be equalised by a relatively small bore interconnecting pipe at the top of the heads that would bleed any superheated gassy coolant back to the better circulating right side and vent any gas to the tank. With adjustable ball valves if they are a help.

Joe

Good idea Joe,

A small bore pipe at the latest possible/practical point in the circuit, connecting each side, would equalise pressure difference between them, without any negative effect on the main flow. Only one adjustable valve would be needed for experimental purposes. In this respect, a flexible tube could be partly closed with a simple clamp during experiments. However if say a half inch tube is used, all should be in order.

As I pointed out in my last post, any difference occurs because of unequal resistance in the circuits post inlet, and a time element is involved. Your idea of a balance pipe late in the circuit close to the outlets, would balance both. Elimination of the time element, will depend on the bore/resistance of the pipe.

Cheers, Trevor.

[Dessertrunner]

My next steps on this job, (been a bit busy building a CNC mill).
- Run a more detailed test on a warm day, above 20*C in an attempt to get left bank to boil or show a clear problem. If it doesn't warm up then I will park car pull cooling fan fuses then do the rev test. Not sure it will generate enough heat with out load but I can only try. (might think about blocking the air to the radiator.
- Following this test change the top cross over pipe to the modified version to see if there is a improvement and it fixes the problem.
- If there are still issues then weld a centre pipe on the top return as we have all discussed. Aim here is to get both banks to flow the same and both guages on each head to read the same.

If all that doesn't work then "WE" will have to work out a new plan. (note the WE bit) as we are all in this together. Well team have a great day.
Tony

[yt]

I discussed it more with Dan today. Looks like we will be reinventing the wheel on the water manifold with his engine. The main discharges from the block will be relatively simple with a pair of two bolt flanges. Will drill and tap the smaller outlets in the forward portions of the block and add a third to each cylinder bank. The three outlets will join the main discharge for each side. With this each bank will be separated until they join is a Y just before the radiator inlet. This is essentially what Harvey had spoken of before. In the name of science and possibly the KISS method, we will be using the stock water pump first to see if we have actually found the issue.

At this point there are really two issues that could be causing the localized boiling.

1. Pressure differential creating a lowered boiling point.

2. Increased resistance for one bank creating a lack of flow through one side and increased flow through the other (path of least resistance)

I really cannot see anything else causing this localized boiling that we are experiencing. So lets try and decrease resistance across both cylinder banks and try to equalize them. From here we must determine if the pump's capacity is less than we need.

Tom

[Trevor]

Boxersix and Joe raise the point of air bubbles in the system, causing trouble.

The problem of the coolant becoming aerated does happen, but not in this engine. If the system is purged when it is filled, there is no air in there and it can't get in.

Air is dissolved in water at ambient temperature, to become released at about 80 deg. C or more, depending on pressure. This air remains within the closed system, unless bled out due to the cap opening to the overflow container and then not completely.

If there are bubbles, they would have to be bubbles of steam or voids. Voids (bubbles of nothing) will disappear with pressure equalization. Bubbles of steam will disappear with temperature equalization.

Bubbles of nothing?:confused:
Bubbles of water vapour/steam will form if the pressure at any point within the system, temporally exceeds the set radiator cap pressure whereby the water will boil. Such pressure hot spots can occur due to resistance to flow or blind sections, as per my previous posts. The essence of the problem with water vapour/steam, is not "temperature equalisation," which is a result, not a cause.

The two outlets do come off the rear of the heads, but there appears to be a couple of bled holes forward of the outlet that would serve to remove any trapped air, when it is filled. Harvey.

If were are such holes, the system could not hold the designed pressure. No outlets, including any at the rear of the head can bypass the radiator cap and bleed to the atmosphere. This is the very reason for careful slow filling, being required.

It must be concluded, that the only way of stopping water vapour being generated at any specific point, is to eliminate localised pressure.
P.S. As I now see yt Tom has diligently pointed out.

[Dessertrunner]

Tried pulling the fuses on the fans and reving the engine to see if the cooling is different between the heads. Bad news it didn't work as radiator just moved up at the same rate as the engine. There was not enough cooling to work. Even with the fans on it didn't work.
Tony

[FriendlyTurkey]

Air is dissolved in water at ambient temperature, to become released at about 80 deg. C or more, depending on pressure. This air remains within the closed system, unless bled out due to the cap opening to the overflow container and then not completely.



Bubbles of nothing?:confused:
Bubbles of water vapour/steam will form if the pressure at any point within the system, temporally exceeds the set radiator cap pressure whereby the water will boil. Such pressure hot spots can occur due to resistance to flow or blind sections, as per my previous posts. The essence of the problem with water vapour/steam, is not "temperature equalisation," which is a result, not a cause.



If were are such holes, the system could not hold the designed pressure. No outlets, including any at the rear of the head can bypass the radiator cap and bleed to the atmosphere. This is the very reason for careful slow filling, being required.

It must be concluded, that the only way of stopping water vapour being generated at any specific point, is to eliminate localised pressure.
P.S. As I now see yt Tom has diligently pointed out.

I agree that a excessive pressure drop somewhere in the system may be causing cavitation to occur. I see the same problem when sizing industrial controls valves at my work. Attached is a simple graph showing what happens when I size a valve with too much pressure drop for a given system.

http://www.spiraxsarco.com/images/resources/steam-engineering-tutorials/6/3/fig_6_3_11.gif

Has anyone tried a 35psi pressure cap to give more headroom for this pressure drop? May not solve the problem, but mask it enough for some.

http://www.speedwaymotors.com/High-Pressure-Radiator-Cap-28-32-Lbs,1837.html?parentDisplayId=13000

The major trouble with running the cooling system at elevated pressure is the reduction in flow from the pump.

FT

[shotgunslade]

I had dismissed the idea of local pressure drop cavitation because this is a closed loop, but I suppose that a local restriction which would require a large decrease in static pressure to accelerate the fluid to the required velocity through the restriction, which could conceivably drop the local pressure below the flash point of the fluid.

Assuming that the radiator cap isn't blowing, then the average system pressure is going to be below 2 Bar, depending where in the radiator cap is located in the line of pressure drop around the system. Let DP be the pressure drop through the whole circulation system; the amount of pressure the pump has to generate to get the flow. Assume the radiator cap is located halfway between (from a pressure drop standpoint) the pump inlet and the pump outlet. The absolute pressure at the pump outlet is going to be

(2 Bar + 1/2* DP).

The pressure at the pump inlet is going to be

(2 Bar - 1/2 * DP)


So, if the pressure drop through the system is 1.2 Bar, then the absolute pressure at pump inlet is going to be .9 Bar. If the water is a 220 DegF, it's boiling at that point, and the pump will lose flow, until the steam bubble passes or collapses and the pump rotor is fully immersed again.

This could happen with a centrifugal water pump operating well above its design RPM, developing more flow against a system pressure drop that increases with increasing flow.

So, we could have a negative suction head problem even in a closed system, when the fluid is operating at a temperature near its flash point.

[FriendlyTurkey]

I like your ideas shotgunslade! I hadn't had my cup of coffee yet! I didn't realize I was looking at a closed loop system with the pressure cap just being a safety valve for overpressure conditions.

It would really be of great value to measure pump inlet and outlet pressure!

[Trevor]

I had dismissed the idea of local pressure drop cavitation because this is a closed loop, but I suppose that a local restriction which would require a large decrease in static pressure to accelerate the fluid to the required velocity through the restriction, which could conceivably drop the local pressure below the flash point of the fluid.

Assuming that the radiator cap isn't blowing, then the average system pressure is going to be below 2 Bar, depending where in the radiator cap is located in the line of pressure drop around the system. Let DP be the pressure drop through the whole circulation system; the amount of pressure the pump has to generate to get the flow. Assume the radiator cap is located halfway between (from a pressure drop standpoint) the pump inlet and the pump outlet. The absolute pressure at the pump outlet is going to be

(2 Bar + 1/2* DP).

The pressure at the pump inlet is going to be

(2 Bar - 1/2 * DP)


So, if the pressure drop through the system is 1.2 Bar, then the absolute pressure at pump inlet is going to be .9 Bar. If the water is a 220 DegF, it's boiling at that point, and the pump will lose flow, until the steam bubble passes or collapses and the pump rotor is fully immersed again.

This could happen with a centrifugal water pump operating well above its design RPM, developing more flow against a system pressure drop that increases with increasing flow.

So, we could have a negative suction head problem even in a closed system, when the fluid is operating at a temperature near its flash point.

Sorry but your theory does not hold water. ;)

The closed loop includes a radiator and the following does not normally apply.====

“So, if the pressure drop through the system is 1.2 Bar, then the absolute pressure at pump inlet is going to be .9 Bar. If the water is a 220 DegF, it's boiling at that point, and the pump will lose flow, until the steam bubble passes or collapses and the pump rotor is fully immersed again.”

[Trevor]

The pressure to be considered is in no way related to the atmosphere. The pressure within the circuit depends on the pump output as is opposed to the available flow path.

Increasing pump speed against a fixed resistance/head, increases the internal pressure. This will constitute a constantly reducing pressure gradient. However, because of huge variations in resistance throughout the flow path, this will not be consistent. The lowest pressure will finally occur at the pump inlet.

Flow will not increase in line with pressure. Pump speed, pressure, flow, will not increase in a linear fashion, far from it. In fact there is a final point whereby flow will choke at a maximum.

Taking the above and all of that mentioned in previous posts into account, there is a point where pressure becomes an absolute disadvantage. It is very possible that this point coincides with approximately 5,500 engine RPM.

Is there not strong motivation for a cheap and simple experiment involving reduced pump efficiency? :confused:

[shotgunslade]

The relationship with atmospheric is that the radiator cap will blow at some differential above atmospheric. The system is closed until that happens. The absolute pressure at the pump discharge may actually be above the level necessary to blow the radiator cap, but pressure drop between the pump and the cap may reduce it, so that the cap doesn't blow. If the absolute pressure at the pump inlet is below the flash point of the coolant, you will get cavitation.

If we know that the absolute pressure of the system at the radiator cap is below 2 Bar (the cap hasn't blown), we could calculate the pressure drop. If it is below the flash pressure of the coolant at whatever temperature the coolant is at that point, you will get boiling.

In general, system pressure drop follows the equation: (where DP = pressure differential, K is a consant and Q = flow)

DP = K * Q**2

So, yes, ieven though the system pressure drop increases with the square of the flow if we are running the engine way above the lnormal rev range, the pump will still have increased flow with increased rpm, even though it is only increasing at less than the square root of the rpm increase. Pressure drop across the ystem will be increasing drastically. Assuming the cap hasn't blown, the system flow pressure drop between the cap and the pump inlet could make the absolute pressure at teh pump inlet less than the flash pressure of the coolant at that point.

[dynomatt]

Remember the old cars with something called a fan clutch? They operated the mechanical radiator cooling fan on a hydraulic disc such that when it achieved a certain load (or rpm), it couldn't spin any faster...this sounds like something we should try with the water pump...above certain RPM's it becomes a fixed speed.

We can do this with electric pumps, but not with the mechanical one.

If this is the case, why have the people who have dabbled with electric pumps (I can think of a handful), not been able to achieve a solution?

I've procured a BMW water electric water pump in an effort to solve some of this.

Based on everything I've heard, my collection of ideas to solve this is:

1) Dual top outlets, separate from each other (like the current six cylinders)
2) Alloy radiator - (PWR or similar)
3) Electric pump
4) Gutted original pump to operate just like an idle pulley, but paying particular attention to flows in and around the original pump
5) Heater system removed - to avoid complications
6) Throttle body water heating removed - to avoid complications
7) Dual high air displacement fans
8) Engine oil cooler
9) Modified thermostat

When will I do this all...hopefully over our summer break...but slowly.

Cheers,
Matt

[yt]

Electric pumps have proven only to be a bandaid to the cooling issue. it makes it better but does not address the real issue at hand.

Tom

[Dessertrunner]

I agree with Tom at this point in time there is no evidence to suggest that the pump is the problem.
As regards pressure in the system, a higher pressure can only happen if part of the coolent system has a restriction that decreases the flow. Any potential pressure restriction is likly to be with in the engine which would mean that there is a potential vacum on the radiator side of the circut due to the pump suction. Having driven for the last week with the 3 temp guages there has to be a major problem with the flow in the left bank. Every time I look the two heads are showing near 10c difference. I agree with the general opinion that the next work needs to be on the top pipe.
The recent discussion leads me to think that I need to find a way to fit a pressure guage or is it correct to assume that Temp is flow. Is the current thought that the left and right heads have the same coolent pressure. I personally think they would be different due to the reduced flow between left and right but am open to correction.
Tony

[oab_au]

While I agree with Toms approach to use two pipes to a two inlet radiator, it may be beyond the possibility's of the average member. I agree it is best to keep using the std pump, till we find if removing the restriction will solve the problem, as most people would be using the std pump.

I think there is merit in Tony's center tap of the pipe, to remove the restriction and provide equal flow, as far as we can.

If the modded pipe was provided with a center divider to direct the flow from each side to the outlet pipe, each side would have the same resistance, and flow the same.
I bit like this.

http://www.subaru-svx.net/photopost/data/500/medium/Center_tap_with_divide_to_equalise_flow_.JPG

May not be too hard to do .

Harvey.

[yt]

add a third small hole to each side... Open up a block and you will see why.

Also, remember making them equal on top means they are unequal as a system. it would be best to run both pipes on the RH side of the engine. The coolant has to flow into the block farther on the RH side so the make it as equal as possbile having a shorter exit would essential try to balance the two halves out

Tom

[oab_au]

add a third small hole to each side... Open up a block and you will see why.

Also, remember making them equal on top means they are unequal as a system. it would be best to run both pipes on the RH side of the engine. The coolant has to flow into the block farther on the RH side so the make it as equal as possbile having a shorter exit would essential try to balance the two halves out

Tom

Well I can only go as far as opening the hood.:D I have to leave it to you blokes to supply the inside details.

The bottom may be uneven, but it is the top outlet that is causing the main problem. If we can get that as even as we can to eliminate the problem there, we can see if there is a need to go further.

Harvey.

[yt]

by creating one leg longer than the other you are essentially causing less resistance in the shorter leg and we are back at step 1.

Tom

[Trevor]

The relationship with atmospheric is that the radiator cap will blow at some differential above atmospheric. The system is closed until that happens. The absolute pressure at the pump discharge may actually be above the level necessary to blow the radiator cap, but pressure drop between the pump and the cap may reduce it, so that the cap doesn't blow. If the absolute pressure at the pump inlet is below the flash point of the coolant, you will get cavitation.

If we know that the absolute pressure of the system at the radiator cap is below 2 Bar (the cap hasn't blown), we could calculate the pressure drop. If it is below the flash pressure of the coolant at whatever temperature the coolant is at that point, you will get boiling.

Yes the pressure at the pump inlet will obviously be less than the output. Also the input pressure can be close to the pressure as set by the radiator cap valve dependent on restriction within the radiator.

However at the pump inlet, the coolant has been cooled by the radiator so that at that point, under normal circumstances, the coolant will not be boiling. As previously pointed out, the radiator must be taken into account.

In general, system pressure drop follows the equation: (where DP = pressure differential, K is a consant and Q = flow)

DP = K * Q**2

So, yes, ieven though the system pressure drop increases with the square of the flow if we are running the engine way above the lnormal rev range, the pump will still have increased flow with increased rpm, even though it is only increasing at less than the square root of the rpm increase. Pressure drop across the ystem will be increasing drastically. Assuming the cap hasn't blown, the system flow pressure drop between the cap and the pump inlet could make the absolute pressure at teh pump inlet less than the flash pressure of the coolant at that point.

Although it is difficult to understand the exact point being made within the post, I certainly agree agree that “Pressure drop across the system will be increasing drastically.” as result of increased RPM. This is the very point, to which I have constantly been drawing attention. ;)

However no one appears to appreciate the consequences of this issue, and it therefore remains ignored. :confused:

[oab_au]

by creating one leg longer than the other you are essentially causing less resistance in the shorter leg and we are back at step 1.

Tom

If the difference is at the block inlet, it won't affect the water pressure inside the block.
We won't be able to get both sides flowing the exact same amount, the cast water passages in the block and heads won't be the same, but does it really matter?
As long as we remove the outlet restriction so both sides can flow their own flow, that should remove that fault and allow us to come back to the high water flow, to see if that needs looking at.

If we are ever to get good track horsepower out of this engine, we need to put torque out at 6000/6500 rpm, the engine needs to run between 5000 to 8000 rpm. If we don"t, successfully cure this problem, all the engine work is wasted.

One step at a time, change one thing at a time, we will get there.

Harvey.

[Trevor]

While I agree with Toms approach to use two pipes to a two inlet radiator, it may be beyond the possibility's of the average member. I agree it is best to keep using the std pump, till we find if removing the restriction will solve the problem, as most people would be using the std pump.

I think there is merit in Tony's center tap of the pipe, to remove the restriction and provide equal flow, as far as we can.

If the modded pipe was provided with a center divider to direct the flow from each side to the outlet pipe, each side would have the same resistance, and flow the same.

May not be too hard to do .

Harvey.

No positive evidence has been established, that the small difference in resistance at this last point in each circuit, is having a significant effect on flow balance.

The difference in flow within the two sides, which appears to be indicated as a result of uneven temperatures, will continue as a result of unequal restriction and distances in each side, at any point post the water pump outlet. This is the major issue, rather than a limited restriction at the final output, where flow is at minimum pressure, and therefore least affected, to be in any event balanced before the final restriction of the radiator.

There are two circuits in parallel, pressured from a single source and ending at a single outlet. The flow in each is governed by the individual total restriction in each. The existence and affect of unequal resistance bypass routes within each circuit, is important, with these constituting the opposite of a resistance.

It should be kept in mind that a limited section balance pipe, between circuits at any critical point, will balance pressure and flow. There should be terminals where one or more could be fitted, without too much strife. It is a matter of fluid mechanics, not an airy fairy mystery. :D

[Trevor]

If the difference is at the block inlet, it won't affect the water pressure inside the block.
We won't be able to get both sides flowing the exact same amount, the cast water passages in the block and heads won't be the same, but does it really matter?
As long as we remove the outlet restriction so both sides can flow their own flow, that should remove that fault and allow us to come back to the high water flow, to see if that needs looking at.

If we are ever to get good track horsepower out of this engine, we need to put torque out at 6000/6500 rpm, the engine needs to run between 5000 to 8000 rpm. If we don"t, successfully cure this problem, all the engine work is wasted.

One step at a time, change one thing at a time, we will get there.

Harvey.

Any difference between the cast water passages within the block and heads most certainly has a critical effect, particularly as these are early in the circuit where there is maximum pressure. The suggestion that this does not really matter is beyond belief. The current paramount question, is that the flow in one head is in doubt and is causing overheating.

It is this very factor of resistance through tortuous passages which intrudes more as pressure, increases. (Due to engine RPM.) High turbulent water flow, constitutes a very poor coolant. This is scientific fact which must not be ignored.

What is more, distance in a passage comprises a definite resistance to flow/pressure. Unequal distances post water pump outlet have a definite effect on each circuit.

Yes, one step at a time, but the likely effective, the simple, the most economic, first.

[Trevor]

Remember the old cars with something called a fan clutch? They operated the mechanical radiator cooling fan on a hydraulic disc such that when it achieved a certain load (or rpm), it couldn't spin any faster...this sounds like something we should try with the water pump...above certain RPM's it becomes a fixed speed.

We can do this with electric pumps, but not with the mechanical one.

If this is the case, why have the people who have dabbled with electric pumps (I can think of a handful), not been able to achieve a solution?

I've procured a BMW water electric water pump in an effort to solve some of this.

Cheers,
Matt

Interesting Matt, best news here so far. :D

I believe that experiments with electric pumps to date have been based on and limited towards the need to increase pressure/flow, whereas the the opposite has been the problem. You will have the ideal set up to experiment in this regard. I certainly hope you proceed as planned.

What you require is a speed controller for the pump. A set up delivering pulsed DC to the motor is not hard to arrange. Try a googly ---- DC motor speed control circuit, or building a DC motor speed control circuit, with the object of obtaining some hobby electronic info. There is plenty out there. If you draw a blank, get back to me via PM with full specs on the motor you have on hand.

All the very best with your project, Trevor. ;)

[Trevor]

add a third small hole to each side... Open up a block and you will see why.

Tom, will you please expend on this point as it appears very applicable and should be given some air. :)

Also, remember making them equal on top means they are unequal as a system. it would be best to run both pipes on the RH side of the engine. The coolant has to flow into the block farther on the RH side so the make it as equal as possbile having a shorter exit would essential try to balance the two halves out. Tom

The distance in respect of each circuit is an interesting fact, which has not previously been addressed. Ford obviously had this in mind years ago, with their twin pump and twin pipe set up on the old flat head V8.

Keep the ideas coming, Trevor. ;)

[RallyBob]

If we are ever to get good track horsepower out of this engine, we need to put torque out at 6000/6500 rpm, the engine needs to run between 5000 to 8000 rpm. If we don"t, successfully cure this problem, all the engine work is wasted.


One concern I have is that the standard cooling system has proven to be substandard for even a bone-stock engine that is run continuously hard (6500 rpm max).

Even if the inherent flaws are cured for that standard engine, will it be enough for 300+ whp and 8500 rpms N/A, or 500+ whp and 7000 rpms w/forced induction and sustained loads that might be seen in off road desert racing and road racing. After all, more power = more heat.

[SilverSpear]

Well 12 pages of theories and still nothing applied. I think the only way to do this is through Tony's method. Water flowmeter, a temperature meter and begin the journey of trial and error in modding the cooling system.

Keeping with the speculations without this will lead to more pages, more confusion and more arguing. We need to test.

Tony any luck wtih the flowmeter? What is the cheapest you were able to find so far?

[yt]

Take this to heart when thinking about the radiator's efficiency.

A stock engine can run at 5000 RPM all day with a stock radiator and never get too hot.

My supercharged car making an estimated 350wheel horse runs all day without ever getting too hot with a PWR radiator

So double the wheel horsepower of the engine without revving it out to 6k+ RPM and you will see that double the power generates more heat. This heat is not too much for the radiator to cool to proper levels.

So to put the radiator out of our minds, we are stuck with the rest of the engine. It will take step by step investigation and modification to actually find and address the root cause. Now, Tony has been a huge help in testing the modified stock water manifold and has brought back results telling us it is not enough. Jack and Bob have also done quite a few things to their engine, which has helped but never fixed the issue at hand. I am going to list a couple things below which have been tried but have not proven any great results

- Electric Water Pump
- Modified Water Manifold
- Aluminum radiator
- Modified Thermostat
- Water Wetter

Lets work from here with all this in mind to address it logically. I will try my best to get the custom manifold done soon but I cannot make promises on a time frame yet.

TOm

[Dessertrunner]

Still not there with flow meter but this is heading in a simlar direction I want to ask everone a question. That maens I want everyone to just say yes they agree or no they don't. Don't debate the issue just agree or don't agree.

The Question do we all agree that,
The flow in the Left Head is LESS then the Right Head and I submit as evidence the higher discharge coolent temp on the left head.
Remember the answer is to be Agree or Don't Agree. When everyone answers I will explain why I ask the question.
Tony

[SVXRide]

agree.....



I'll also agree that we have localized boiling /gas pockets being formed ;)



-Bill

[SilverSpear]

Ok Tony, what is the issue?

[Dessertrunner]

What do you other guys thinK?

[oab_au]

Don't like the wording, but yes I agree.

Harvey.

[Dessertrunner]

Great Harvey, words were never my strong point. Just need a few more to agree. I am trying to target one problem then fix it.
Tony

[yt]

Agree with Harvey here. Wording is not ideal but the idea is there. I agree that this is main focus point we need to pay attention to

Tom

[svxistentialist]

My point about bubbles, cavitation, steam, voids, whatever we call it, was in support of Tom's theory above and relating to your earlier hypothesis on pressure drop boiling, which Trevor dissed.

The whole system may be equalised by a relatively small bore interconnecting pipe at the top of the heads that would bleed any superheated gassy coolant back to the better circulating right side and vent any gas to the tank. With adjustable ball valves if they are a help.

Joe


It should be kept in mind that a limited section balance pipe, between circuits at any critical point, will balance pressure and flow. There should be terminals where one or more could be fitted, without too much strife. It is a matter of fluid mechanics, not an airy fairy mystery. :D

What you suggest here Trevor sounds a lot like what I mentioned earlier, and what YT is saying. I do think the core of the problem is fluid mechanics. Two separate heads being cooled by a pump on one side.

Joe

[svxistentialist]

Also I agree with Tony's analysis.

Too much heat on one side, or ineffective cooling on one side, however you want to see it.

[I'm sure Tom meant to agree with Tony, not Harvey? Tom?]

Joe :)

[yt]

Don't like the wording, but yes I agree.

Harvey.

Agree with Harvey here. Wording is not ideal but the idea is there. I agree that this is main focus point we need to pay attention to

Tom

Yes, do agree with both if thats what you meant

Tom

[svxistentialist]

Yes, do agree with both if thats what you meant

Tom

:D

I meant that Tony was the one who asked for a specific Agree/Don't Agree answer.

[shotgunslade]

Agree. Lower flow through left bank. Probably causes local boiling.

[Dessertrunner]

Okay, today (it is now Sunday 6.30am) I will change the pipe on the top of the engine. Based on what we all agree if the tempreture of the left bank runs cooler then the right bank it means that the flow has become higher then the right, if on the other hand we get the same tempreture in both heads we would assume that the flow is constant. So lets see what what happens will tell you at the end of the day.
The reason for asking if everyone agrees is that we can"t flow test left and right bank that easily so we need to all agree that there is a different method to find the answer.

The Biggy for the day,
Second point I need some Hp power numbers of the standard engine at different rev, 2,000rpm, 4,000rpm, 6,000 & 8,000. I would also like them for the target engine. I have been thinking about what Rally Bob said and think we need to build a quick & dirty, Tony "type" rough computer model to show us what would happen under different scenarios. It won't matter is the Hp number are right or not just that they are indictive of the engine.
The model will deal with the follow variables and I was going to us excel.
Change in Engine Revs,
Changes in air tempreture past radiator,
Change in coolent flow rate,

If anyone feels like giving it a go please speak up, other wise I will start on it tonight. Have a great day team.
Tony

[oab_au]

Okay, today (it is now Sunday 6.30am) I will change the pipe on the top of the engine. If anyone feels like giving it a go please speak up, other wise I will start on it tonight. Have a great day team.
Tony

What are you changing the top pipe to?
What modification have you done to the pipe?

Need To know.:)

Harvey.

[Dessertrunner]

Harvey,
Think of the moded one that Tom posted a photo of first. Haven't got the time and parts to build a "T" one today but want to see how the simple mod goes as step 1.
Tony

[Trevor]

One concern I have is that the standard cooling system has proven to be substandard for even a bone-stock engine that is run continuously hard (6500 rpm max).


The crux of the problem is that cooling efficiency deteriorates as RPM rise, i.e. as pressure rises. Increased pressure does not equate with an efficient increased steady cooling flow. This is not difficult to comprehend, but is being ignored. All other issues transcend this fact and should be accorded similar status, but the cart is before the horse. :)

The pressure does not rise at a linear rate. Cooling deteriorates accordingly on a slippery slope, at hot spots and points of resistance, as a result of increasing pressure. Balance within the system is upset to an increased degree, as a result of increased pressure. Cavitation at the pump is more likely at increased RPM, against what is a fixed resistance.

Increasing pressure in line with RPM, is surely the essence of the subject.;)

[Dessertrunner]

Harvey,
Forgot to ask, any ball park guess on the HP numbers, it just needs to be a best guess.
Tony

[yt]

A stock engine generally produces somewhere in the ballpark of 175 wheel horsepower if it is in good shape w/ a std. trans. Around 160 w/ the auto

Tom

[Dessertrunner]

Well the cross over pipe didn't fit, its to high as was suggested. I need to cut and weld to get around the problem.
Any way I will photograph the fix when done and keep you posted.
Tony

[SilverSpear]

Crap... is he gonna pay you for it?

[dynomatt]

Post edited

[Dessertrunner]

Sorry for going off before I was unfair.

Good news, I got the pipe changed this morning and drove the car to work and guess what the left and right bank were running at the same temp. It was possiable to sometimes see 1c difference but was either way, above or below right bank temp.
Other interesting thing is that the dash temp guage ran slightly lower which would indicate to me that the thermostate is functioning better.

Hooray team we have progress, we are heading in the right direction on this project.
So what next? Have a great day.
Tony

[FriendlyTurkey]

Great Job! Thank you for taking action and doing experiments for the group!

I think the next thing to do would be the following:

Observe temperature of each bank at different rpm points: 3000, 3500, 4000, 4500, 5000, 5500, 6000, 6500, 7000, 7500. Record temperature, and difference in temperature between banks.

Observe under which conditions that the present combination will overheat. (rpm, temperature, time)

[yt]

I thought you had said the modified water manifold did not help?? Now it does?? What is going on here?

Tom

[Dessertrunner]

Tom the last time we tried the manifold was on the dune buggy (sand rail), that was what we were talking about. This is the first time we have tried it on a SVX car. You and I have a bit of a challange to get them to fit under the intake as the idle valve plug hits and the vacum chamber hits at the front. But other then that its all good and works well.
Tony

[oab_au]

Sorry for going off before I was unfair.

Good news, I got the pipe changed this morning and drove the car to work and guess what the left and right bank were running at the same temp. It was possiable to sometimes see 1c difference but was either way, above or below right bank temp.
Other interesting thing is that the dash temp guage ran slightly lower which would indicate to me that the thermostate is functioning better.

Hooray team we have progress, we are heading in the right direction on this project.
So what next? Have a great day.
Tony

OK that is what it is like on high speed commuting. Now we need to test it under racing conditions. Take it out on your favourite test track, use say 3rd gear (don't want to run too fast:)) do a full throttle acceleration up to top revs, off the throttle, jump on the brakes, down to 5000, then back to full throttle, to full revs again, jump the brakes, etc, etc. Try to keep going, watching the temps.
Don't spend too much time off the throttle, you need to have a high pump speed and good heat input from the engine. Just like you are racing. Don't run off the road, don't hit a roo.:D

Harvey.

[Dessertrunner]

It was hot yesterday and the car cooling did some funny things, I have made some changes overnight and will see today. I will come back to you guys with an explanation tonight after we see how it performs today. In short the engine ran cooler yesterday but the radiator went to high, so I am checking things out.
Tony

[SilverSpear]

Nice progress Tony, please keep taking pictures of the progress you are making...

[Dessertrunner]

I have been talking to PWR today about radiators and I am going to get them to build a radiator with some differences.
Top and bottom pipe to be larger as I think the current size is to small for the flow at 6,000 rpm (200lpm). On the bottom feed pipe to the pump it could cause the pump to cavate so to get round any possiability of this I will change the pipe on the pump as well. Harvey's diagram of putting the top pipe return in the centre is the way to go so while I am at it I will do that change and increase the pipe size.
Next change is the number of fins per inch. Currently when you slow down the return tempreture of the coolent going into the pump shoots up. Also if I am traveling at 110kph and turn on the fans the return coolent temp drops which means the PWR air flow is poor. I read that a formula 1 car has 26fpi but if that was put in a normal car it would boil at normal travel speed on a average car. PWR told me they have 16fpi in the SVX ones they build and suggest that the OEM unit would have the same. My concern is that the PWR is twice as thick as the orignal but same fin number. The air can't flow through it properly this has the added effect of causing of the whole engine bay running hotter. I noticed that the SVX I have with the PWR boils the water out of the battery more often then the car with the standard. If you feel both bonnet under the same driving condiations you will find the PWR engine bay hotter.
Guys with the sand rails tell me they are going away from PWR as they don't cool the engine as well as other brands. If I get stuck behind a truck on a hot day the return coolent temp goes up to a dangerous level as well. so what would happen on a race track.
Sorry to have such a long post but needed to share my thinking. I am going with 12fpi reason 2 surfaces twice as think as the OEM means 48 equlivant to the existing 32 surfaces but maintaining airflow(got the quick crazy logic). Have a great day team will keep you posted (Are you still out there Trevor).
Tony

[oab_au]

Sounds good to me mate.

Your car is a bit different to the normal SVX, in being raised on the suspension. This reduces the negative pressure that is generated under the car by the venturi effect of the airfoil shape of the under body.

This creates a negative pressure in the engine compartment to enhance the airflow through the heat exchangers.
So it may well be different to the std one.:)

Harvey.

[Dessertrunner]

DO you happen to know how many FPI on the standard radiator?

[Trevor]

Have a great day team will keep you posted (Are you still out there Trevor).
Tony

I have given up posting, as a result the lack of logic being recorded. From afar one can only base thoughts on what is stated. Your very first post, all of which I took in, advised:-

What I have found recently is the following:-
- Multiple pass radiator "DID NOT" slove the problem.
- A bigger Radiator "DID NOT" slove the problem (buggy radiator is 3 times bigger then my PBR).


You are spending many dollars, I hope in the right direction. You are making several changes at once, which is not good practice. Therefore you will be unable to detect which has made any improvement which takes place, or the exact reason for same.

I wish you well, Trevor.

[Dessertrunner]

Hi Trevor good to see you are alive.
Not spending that much, my current PWR radiator is reaching the end of its life so it time to get a new one. Last one only lasted for 250,000k.
My wife says I don't explain myself well, I am doing one step at a time, we have fixed the flow to the heads but now shifted the problem to the radiator. Coolent is coming back to high in temp so instead of buying the same radiator again I am looking at something different.
Need to ignore the work on the buggy as I have no control over it and there are to many different factors to compare it to here.
I figured out a way to test air flow in different radiators so I want to give it a go. Will keep you posted.
Tony

[yt]

If your radiator is faulty, thats one thing. I still feel we are barking up the wrong tree with it as a solution though

Tom

[Dessertrunner]

Tom,
I think we need to have an open mind and try different things. If the pump flow is low we don't have much room to move in the radiator department.
May be I have confused the issue the big change has to be the pipes to try to get more flow through the system at high revs. Ignore the rest.
Tony

[Trevor]

Tom,
I think we need to have an open mind and try different things. If the pump flow is low we don't have much room to move in the radiator department.
May be I have confused the issue the big change has to be the pipes to try to get more flow through the system at high revs. Ignore the rest.
Tony

Tony,

Thanks for the run down on the radiator. This makes good sense. Also good that you are making any changes separately.

What factual evidence do you consider indicates that, "more flow through the system at high revs." is required?

You will get there, Trevor.

[yt]

Couple reasons I thinkwe should steer away from the radiator being the issue...

Jack is running a different rad in the EG33 RS

Almost every dune buggy is not running our radiator either. While there may be inherent flaws in the radiator you are using, this may also be tainting the test procedures...

In the engine you are using, have the head gaskets ever been changed?? At this age, every EG33 w/ original gaskets is suspect to needing a refresh. I think a proper test bed will need to have them done before we can take anything as an absolute

Tom

[Trevor]

Couple reasons I thinkwe should steer away from the radiator being the issue...

Jack is running a different rad in the EG33 RS

Almost every dune buggy is not running our radiator either. While there may be inherent flaws in the radiator you are using, this may also be tainting the test procedures...

In the engine you are using, have the head gaskets ever been changed?? At this age, every EG33 w/ original gaskets is suspect to needing a refresh. I think a proper test bed will need to have them done before we can take anything as an absolute

Tom

Tom, logical stuff. :)

By the way by advising, "Thanks for the run down on the radiator. This makes good sense." I was referring to the cost of the radiator change, rather than the reason for the change, and agree with your thinking on this aspect.

Too many conclusions to date, have no absolute foundation.

Good one. Trevor.

[Dessertrunner]

Tom what do you think we need to look at next, I accept your point about Jacks car. We need to remember that his top pipe is orignal.
Trevor all current H6's pump more coolent as was mentioned by Matt, he made a point and I think we need to keep it in mind why else would Subaru increase the flow.

Guys I think that we need to remind ourself that not long ago we were just guess about the Left bank now we know it to be fact (10c difference). We were guessing about the top pipe being to small and now we have confirmed it to be a fact (after the "Tom Pipe" was installed temps were the same on each bank). So I think its alway important to remind ourself what we do know to be fact. Have a great day.
Tony

[Trevor]

Trevor all current H6's pump more coolent as was mentioned by Matt, he made a point and I think we need to keep it in mind why else would Subaru increase the flow.
Have a great day. Tony

This is a hardly a definitive reason and relates to what will be a quite different flow path.;)

What you need to do is have a look at the coolant flow and check for turbulence etc. You should also be able to get an idea of the flow rate and observe any change as a result of different mods. Seeing is believing. :D

For a start, you could replace the top tube with transparent neoprene, silicone or Goodyear Tygon tubing, or insert a section of same using short metal connectors. This tubing is available with a suitable temperature rating. Examples ( Read to the end for specs.) :-

http://www.goodyearinternational.com/tubing.html

I am not sure about the common stuff available your way, from plumbing suppliers, but not hard to check. The temperature rating may surprise you.

Cheers, Trevor.

[yt]

Tom what do you think we need to look at next, I accept your point about Jacks car. We need to remember that his top pipe is orignal.
Trevor all current H6's pump more coolent as was mentioned by Matt, he made a point and I think we need to keep it in mind why else would Subaru increase the flow.

Guys I think that we need to remind ourself that not long ago we were just guess about the Left bank now we know it to be fact (10c difference). We were guessing about the top pipe being to small and now we have confirmed it to be a fact (after the "Tom Pipe" was installed temps were the same on each bank). So I think its alway important to remind ourself what we do know to be fact. Have a great day.
Tony


I would like some more confirmation that the two sides of the engine are running at equal temps now.

The fact that the modified manifold did not help the buggy is what worries me most.

I still think harvey's idea to have individual manifolds with a merge that agrees with the fluid's velocity is going to be the real winner here.

Tom

[Dessertrunner]

We have to go the way Harvey suggests if what you are saying is the pipe in the centre. There is still a slight difference in temps from left to right but you can hardly notice it but it still worth further improvements such as the centre pipe.
I do think we need to increase the inlet and outlet pipe sizes as it gets rid of any other possiable flow problem espically the suction to the pump.
Tony

[yt]

We have to go the way Harvey suggests if what you are saying is the pipe in the centre. There is still a slight difference in temps from left to right but you can hardly notice it but it still worth further improvements such as the centre pipe.
I do think we need to increase the inlet and outlet pipe sizes as it gets rid of any other possiable flow problem espically the suction to the pump.
Tony

size does not always mean more flow. A smaller pipe will increase velocity as long as it does not have any obstructions like the stock configuration does. Thats why the custom dual manifold system will likely work best. Lets hear more about how the modified stock manifold is working??

Tom

[oab_au]

I don't think increasing the hose size is going to achieve the results, it may help, but you would still have the same diameter pipes on the pump inlet and output, the same flow capacity through the water jacketing, etc, there are many points that you could jump at.

I can't see anything being a greater restriction, than the junction of the two flows are having at the right outlet. Just my opinion, but I see this as the No 1, fault to remove. Then it has to be tested under racing conditions, to be sure it is fixed.

Harvey.

[Dessertrunner]

Okay I may have not explained myself correctly I am sorry for that. First off yes a smaller pipe will flow faster but if this is on the suction side it creats a bigger vacum which airrates the coolent and in extreme case causes cavitation. The pumps we have is basicly a centrifcal type as such in the real world the suction is always larger then the outlet. Given that we need to run at high revs and pump every last drop of coolent there is a lot to gain by changing the pipe. The other point is that veloicty does increase but as we have found out from the top pipe friction goes up to a point that flow is reduced.

I need a new radiator so I was aiming at getting a bigger inlet and outlet.
The top pipe needs to come off the centre cross pipe as you suggest Harvey, if we are running AC there is not enough room to do 2 pipes. If I am building a new take off pipe I will make it larger as it costs no extra.
The cover on the Pump is just that and it will take about 1 hours work to enlarge the pipe on it as well. Then its just a matter of finding new hoses.
Sorry for not explaining that I was going to change all the pipes.

Tom some strange things are happening now that I changed the top pipe, all good but very confusing. On the first day when they happened I figured I did something wrong but as I have watched the temp guages and driven the car I have started to understand that it does make sense. This post is a bit long but here is what I saw.
- For the first time ever the Radiator had times when the inlet temp to the water pump was higher then the engine temp.
- Left and Right bank are with in 1C of each other, if the radiator outlet temp (40C) is really cold there is still a slight difference between both banks.
- I have seen the right bank hotter then the left.
- Engine dash tempreture reads lower all the time.
- A bit early to say but it appears that Maxium head tempreture in each bank is about 5 plus lower then ever seen before.
- Radiator seems to take bigger heat loads then before, meaning, I think that more heat is being transfered from the engine.

Sorry if I am not explaining myself properly.
Tony

[Dessertrunner]

Got a email from Steve with the buggy yesterday and he has parked it in frustration. I am going to contact him today and ask if I can fit temp guages to the back of the head etc to try and find were his buggy is failing. To go forward I need to be working on a engine that is boiling to learn what happens during cooling failure. Will keep you posted. Not sure he will agree but can only try. SVX engines are a dirty word right now.
Tony

[yt]

I guess there is a lot at play with current vehicle in question. You might have some coils in the radiator that are not flowing or cooling so well and causing the heat to either be held up or passed through too quickly

Tom

[Dessertrunner]

Good news guys Steve has agreed to let me plug in Temp guages to his engine and monitor it while he gets it to overheat. He has the new top pipe on so if I do it properly I should be able to find were his heat is coming from. I need to give this some thought as it may be smarter to us a PLC so I can log the data and graph it later. Its a bit more work but would give more accurate info to base our knowledge on also the tempreture probes are of higher quality therefor the info will be more reliable. I may see if I can figure a way to monitor radiator airflow as well. Have a great day and a great weekend.
Tony

[FriendlyTurkey]

Does his current EMS have logging capability? Would make a lot of sense to log with the ecu if it does so that you can look at RPM, Load, vehicle speed (correlates to radiator flow) to compare to your temperature readings.

FT

[Dessertrunner]

I will ask its a great suggestion, thanks

[NeedForSpeed]

I will ask its a great suggestion, thanks


So obvious, it's brilliant, thanks again FriendlyTurkey.

[Dessertrunner]

Okay we have new news. I have been working out in my head that I need to find out how much air is getting through the radiator. So over thelast day I have played with computer fans etc but found the best was was us a standard fan and cut the blue and black wires then connect to a multi meter to measure the volts. What happens is when the fan free spins in the wind it generates power. That power varies based on the speed of the fan. So I calabrated it in free air then fitted it in behind the radiator removing the existing one then drove down the road at different speeds to see what would happen.
Enough talking here is the graph which really shocked the hell out of me.
Tony


http://www.subaru-svx.net/photopost/data/500/medium/RadiatorTest.jpg

[NeedForSpeed]

Tony,
I read through your post a few times. You are saying that very little air passes through the stock radiator, based on a slow fan spin and low measured voltage. The PWR flows a little bit more, based on same principle. Neither radiator is anything like free wind.

Is the obvious question, 'how to get more air through the radiator'? It would seem that the off-road rig your friend has the problems with would get plenty of air through his radiator?

[Dessertrunner]

Sorry guys I tried everything to post the graph with in the posting you will need to go to the link.
NeedForSpeed I was dumb founded with the result and the flow was so bad that with out the second fan I boiled the standard SVX.
Slightly different direction I was looking at a radiator for a high performance STI and it had 2 by 40 amp fans as opposed to ours having two 15 amp (could have worked ours it out wrong). Also the STI radiator was bigger then ours again.
Based on everything I saw today you need the biggest sucker fans (By amps) you can fit. Also I seem to recall Subaru turning one of our fans on at slow speed. Hope that helps, to me its a major break through on understanding the cooling issue.
Tony

[Dessertrunner]

Guys I am surprised that no one has commented, did the info shock everyone or did I do someting wrong.
Joe made a point to me about my car height. which would change the result but the data is still worth having. So I may need to clear some thing up.

- I used two different cars, mine and my wife. Both had the sump guards (even the standard plastic one) removed.
- I rechecked the specs on the STI fans and I read it wrong thay were only 5 amp each (far to small).
- I may need to clarify that the units used are volts but they are directly related to CFM of air flow. Its just a different way to measure the number of revs of the fan. If Subaru told us how many CFM the fan generate then I could convert the figures into CFM.
- NeedForSpeed forgot to answer your question, buggys have terriable airflow around the radiator as they hide it behind the driver so it doesn't get damaged.

All comment greatly appricated.
Tony

[legacyau]

Tony i think that the sump/engine guards may actually have a part to play in the aerodynamics of the engine bay by creating a low pressure area behind the radiator? On another point I remember back in the earlier 90's Possum Bournes legacy rally car sounded like a jet engine as it had a huge fan in it, same with the WRC cars in rally oz just recently, the hot air outlets in the bonnet of the fords were like those hand driers in restrooms only 3 times as much flow that you could feel from several feet away, maybe our fans arent cutting it any more ?

regards cam

ps good work

[NeedForSpeed]

Tony,
Yep, that answered the question, thanks. It might be interesting to retest with the sump guards, to see if there are measurable differences? It's the weekend, we should have some input from others soon. Ron

[Dessertrunner]

Cam Thats really interesting about the fan.

[SVXRide]

Tony,

I'm not sure if the results are all that surprising. Take a look at the actual "open space" area on any radiator compared to the "open space" area in the grill in front of the radiator. I'd be surprised if the radiator has even half the effective "open space", with things getting only worse if you take into account the drag around the fins and core lines.

This said, yeah, heftier fans are usually a good thing;):cool:

-Bill

[Dessertrunner]

Okay More data,
Mounted a computer fan in behind the existing SVX fan to see how much improvement on flow they made so here is the data. Seem we need to improve the set up of the fans so they acheive more when on.
Tony
http://www.subaru-svx.net/photopost/showphoto.php?photo=1838&size=big&cat=

http://www.subaru-svx.net/photopost/data/500/medium/RadiatorTest_using_computer_fan.jpg

[svxistentialist]

Tony, don't forget one thing when checking how much throughput of airflow is aiding your cooling.
It's this:

In the SVX the two "sucker" fans are forcing airflow through the radiator and into the engine compartment. No doubt when the aircon is on and both fans working hard you will have noticed the flood of hot air around your feet when you step from the car with everything running.

When the engineers designed this they paid attention to the airflow under the car and at higher speeds cool air running under the car creates a low pressure extraction area to the rear of the composite sump guard under the engine.

So at say 100k you will have a combination of forces.

* You have incident high pressure air hitting the rads at the nose of the car
* You have low pressure at the back of the rads when the "sucker" fans are running
* You have low pressure areas to the rear of the sump composite cover created by fast air under the car passing the voids left and right.


These two suction areas left and right of the transmission under the car are highly relevant for you in particular. Trevor touched on this point when he mentioned that your car is higher off the ground than standard. This could have the effect of reducing the venturi effect at these areas [call them hot air extraction areas, for that's effectively what they are] and making it harder work for the first two forces to pump through the rad and get shut of the heated air downstream under the car.

In addition to this possible negative effect you have told me you have a sump guard under your engine and extending back under the transmission. You need to take into account that the shape of this guard may well be working against you in your quest to flow more heated air from the rad out of the engine bay. If you think about it, the only way all this heat can get extracted from the rad and the engine bay is through these areas underneath. You may need to look carefully at the shape of these new sump and engine guards that you have put on to ensure that your rad-extracted hot air has a low pressure area to flow into and get dumped in the airstream.

I don't know if I've explained that clearly enough, but you will need to maximise this extraction of hot air underneath to aid the work of your radiator fans.

Cheers man

Joe

[Dessertrunner]

Here is a photo of how I mounted the computer fan.
Tony

http://www.subaru-svx.net/forum/attachment.php?attachmentid=12913&d=1255826087

[svxistentialist]

That's an interesting set-up Tony.

I think it is possible that at higher speeds this fan could spin more and show more voltage if you had the under tray fitted. I can see the floor beneath.

At higher speeds with the tray you may get and be able to measure the benefit of the low pressure suction effect.

Probably at lower speeds or when stopped having the tray off may aid the two suction fans.

Just conjecture. It would be interesting to see readings on it. Your computer fan is a great idea and correct in principle, but I'm wondering if for sensitivity it might need to be a specialised low inertia fan to give good representational figures at low speeds of throughput?

Joe :)

[Dessertrunner]

Joe I agree with you from my experiance the sump guard on my car due to its shape has ment that the car was running cooler (must be creating suction). That said any change will only be relative and not that large. What the trials confirms and I haven't made a big point of is that since I fitted my return coolent temp guage I have been concerned about the high temps coming back from the radiator.
In my opinion this is also a contributing factor to the cooling issue. Fixing the problem is a package deal need to improve coolent flow to the heads and fix left head problem, decrease restriction and get the coolent return temps down.

(Have to do this in Imp) Every 1 degree F lower on return temp of 1lb of coolent is equilivant to 1btu of heat transfer. So if you are running at 6,000 rpm pump 440lb pm of coolent with a runduction of temp of 20F you are getting rid of 8,800 Btu of heat per minute. Added to this the gains from heat transfer effectivency from have a lower return temp to operational temp and you have a winner. Sorry for being to complicated.
Given the numbers we are seeing for airflow I can't see how you can cool the engine by attacking just one part. I am designing changes to the PWR which I will post when I finalize them. Then order the radiator and run a bunch of new tests.
Tony

[FriendlyTurkey]

It is no doubt that the radiator posses a restriction to free airflow. This is why radiator ducting is so important. You need to force air to go through the radiator and not around it. This is why you see a smaller opening compared to the surface area of the radiator face. By going from small to large, you are slowing down the air. It has less of a tendency to turn around, and go around the radiator in this design. You also want a nice smooth path out of the radiator to a lower pressure flow along the vehicles body. Look at airplane and race car design and you will see lots of ideas you can apply to increase the efficiency of the radiator system. You want to have a pressure differential across the radiator so the air wants to travel through it. This is what fans do when they are on and are primarily useful at low speeds where you won't have enough air speed to get proper radiator heat transfer. At higher speeds your ducting can push a lot of air through the radiator without the fans assisting. The fans actually become a restriction at high speeds with proper ducting.

[Dessertrunner]

I have placed the order for the radiator, it has a number of slight changes.
Instead of 16 fpi I have it with 12 fpi my aim here is to increase the amount of air getting through at low speeds and also enable the fans to acheive something when they are on. Currently the fans seem to get no gain from the increase travel speed. I think Friendly has a point, still trying to get my head around it interms of fan speeds, does the fan speed up if the load is less or is it like a AC motor that refuses to go faster which would mean at high travel speed turning on the fans slows the airflow as you have suggest.
Anyone know the answer?
The second change is the top and bottom pipes have been changed to 45mm OD which means they are capable of nearly twice the flow of our current pipes this is because I want to do as Harvey has suggested and have a top pipe Tee.
Will keep you posted.
Tony

[NeedForSpeed]

Tony,

Should be interesting to see how the new radiator works. Too bad you couldn't test the elements separately, meaning, first the tee, then the fpi, then the tee/fpi with larger openings. Something like that.

You can answer FriendlyTurkey's alluded question, what happens to the water temperature, at high speed, if the fans are turned off? You are probably the only one here going fast enough to find out :D

Ron

[Dessertrunner]

NeedForSpeed, didn't get what you ment till I had a look at the last graph again, I will fit up the test fan and give it a run in the morning.
Tony

[NeedForSpeed]

I think Friendly has a point, still trying to get my head around it interms of fan speeds, does the fan speed up if the load is less or is it like a AC motor that refuses to go faster,

which would mean at high travel speed turning on the fans slows the airflow as you have suggest. Anyone know the answer?
Tony


Tony, I was referring to Friendly's point above, second part. What happens if you travel high speed, with fans on, then turn off the fans? If the fans actually slow the airflow at high speed, will the water temp drop if the fans are somehow switched OFF, if this allows more airflow?

[Dessertrunner]

NeedForSpeed I did the run with fans on and off(I can control my fans from the dash) at 150kph & 175kph. The engine temp is more effected by the speed but the fans data was a real eye opener, see the graph.
Tony

[SilverSpear]

I was reading Turkey's post and I have a comment:

We used to own a Merc with a 6cyl 24v 3.0L engine and oh boy that engine used to heat up like hell. But the cooling setup is impressive in terms of fans where you can find three of them, one with a clutch driven by the engine crankshaft and two external.

Aside from the crank driven fan, the outer fans were superb in terms of cooling. They had around 12 blades each with motors that can even push a car when in neutral (figure of speech).

This setup is also available on Green1995SVX's 92' 500SL. I think if we can replicate this setup in the SVX, this issue would be resolved.

[FriendlyTurkey]

Is there any way you could do radiator ducting like this to give the airflow through the radiator a place to go?

FT

http://i169.photobucket.com/albums/u210/adamalexaner/Subaru_WRC_Rally_Car_01-1.jpg

[Dessertrunner]

Under some of the Australian race rules that Matt and I will race under we are not allow to add any entry or exit holes in the exteior of the cars body. As they say here "bugger".

On a further point about staged trials, it is very difficult in the real world to make one change and then measure its effect on the cars cooling, then do a different thing and note the change. The reason is that even though you may drive at the same speed I can get different airflow reading if I travel in opposite directions due to wind. Even when the air tempreture is the same but a different day I can get different cooling due to the fact that the levels of humidity are different.
For these reasons I do the changes to the daily drive and you can see over time the "Average" improvement when you have made a real difference. My attemp to put numbers on some of the factors that will effect cooling helps due to the fact things like airflow directly relate to the engine cooling as long as increased airflow is not gained by a greater decease in radiator fin surface area.

Before I change the radiator I will attempt to get some heat data by doing a run at different rev same outside temp but on 3 different days to get the average. This is in an attempt to do as Harvey suggested a while back. Have a great day.
Tony

[yt]

If I remember correctly the fans turn off at 40+mph... I may be mistaken though

tom

[Dessertrunner]

Tom I have hooked (with a diode) into the ECU control wires and put a switch so I can turn them on from the dash. The ECU will turn on based on temp at any speed but I think they turn on low below a certain speed which may be the one you are talking about.
Tony

[Dessertrunner]

Did a test on the way to get some parts. The results are pretty interesting
Target of test was drive constant speed just different revs by changing gear down. In third at 5,500 rpm had to cancel the trial as the car would have cooked as the engine failed to cool. In theory the load was constant the only change was the revs. Haven't had time to graph but will at some point.
Air Temp 30C
6th gear 2,600 rpm return coolent temp 63C engine temp 95C
5th 3,750 80C 97C
4th 4,100 85C 100C
3rd 5,500 92C climbing 105C Stopped trial.

Drove 30 k same speed in 6th and temps return to previous result.
Tony

[dynomatt]

No change in speed, or load, but only revs?...I can't help but come back to water pump cavitation as I originally suggested.

Is there any other option?

Tony, the only observation I can make on the test is to bring the base temperature back to a known state at each rev point for consistency. Heat soak etc, could have affected this.