SVX Engine cooling "Again & Again"

[Tapani]

http://www.surrealmirage.com/subaru/files/SAE_boxer.pdf

The cooling system is designed to balance heater and radiator return ratio so that the coolant temperature going into the engine is always constant.

This is the basics to have thermostate inline with the return hose down at the water pump.

If the thermostate is moved to the upper hose the engine will receive coolant of uncontrolled temperature (cold) and this might have drastic effects in an engine not designed for it. Think about crank shaft alignment.

Any comments?

Tapani

[Dessertrunner]

Originally Posted by Tapani

http://www.surrealmirage.com/subaru/files/SAE_boxer.pdf

The cooling system is designed to balance heater and radiator return ratio so that the coolant temperature going into the engine is always constant.

This is the basics to have thermostate inline with the return hose down at the water pump.

If the thermostate is moved to the upper hose the engine will receive coolant of uncontrolled temperature (cold) and this might have drastic effects in an engine not designed for it. Think about crank shaft alignment.

Any comments?

Tapani

Sorry to disagree but we have done computer modeling using excell to show that what you say is incorrect. Your assumptions above do not mention revs, it is impossaible for the return water and the radiator water to always be in the same ratio as each path has a different friction coefficient at different flows.

As far as I am aware Subaru are the only guys in the world that have the thermostate on the water pump athe inlet to the block. THe problem they face is that the ideal ratio to keep a engine cool is variable based on the radiator return temp. Before all my mods I was travelling in the mountains and the outside tempreture was around Zero C. I watched the engine heat up then the thermostate open dump cold coolant into it suddenly the engine temp dropped 30C then the thermosate closed and the engine heated up till the whole cycle was repaeted.

Pretty sure a engine moving up and down 30C is not good for it.

Next point All my SVX's have the thermostate on the top pipe now and they are getting a lot better fuel encomy, my wifes car just did 630k for 51litres of fuel, this on a car and injectors that have 260,000k.

When the cars are driving along the road if the air temp gets to 44C outside we never see the car engine get hoter then around 88C. On my car I have the fans hooked to the dash so I can see when they turn on and you cand drive all day without them turning on. The only time you need them is if the car parked or moving very slow were to gets poor airflow.

By the way all, Bazza and I are going to run some extra test on the cooling system at hi revs to see if we can get rid of the air build up issue on race cars.

Tony

[Dessertrunner]

Tapani,
Just had a quick look at your link,
I agree with most of what they say I think the engine cooling system design on a Subaru is very good its just that they screwed it up put the thermostate on the pump.

Thanks for the link.
Tony

[Tapani]

Hi,

Is the Fig 1 in the manual page 2 in the cooling section correct?

There's an outlet fitting on the pump discharge - but it's not presented in the Figure 1, though. Both the TB and the heater core supplies and a separate bypass line seem to be tapped from the crossover pipe and all three are discharged into the pump inlet / thermostat housing (to open up the thermo).

Thanks,

Tapani

PS The Porsche 996 concept is basically the same (see attachment) and it seems it is shared by others too, like the Boxter. What do you think?

[Dessertrunner]

In answer to the Porsche first,
I don't think its the same as we can't see what the thermstate housing looks like and therefor can't tell which side of the state the flow is. Also the heater gets turned on and off by tap 11, so if it was a control point to open the thermostate it wouldn't happen if the heaters off.

Okay smart parts about this design which I have been trying to figure out how to fit to my engine is 5 and 6 especially 6. I worked out last year that once we controled the engine coolant temp we could use it to control the oil temp and keep it constant, its the only way to go. My problem I haven't found a water to oil cooler that would suit the job. As I see it this is by far the best way to grantee consistant engine oil temp.

Next smart part is the tank 7 open taps at 8 to get the air out of the system then close. It also has a spring so there must be a valve to let air bypass, ( think I just figured that out). Line 9 supplies coolant to the pump direct if it trys to caviate that would creat a vacum in tank 7, spring valve opens and sucks any air building up in the raidator and oil cooler.

Interesting how the radiators are in parallel not series makes far more sense, in effect they get more flow supplied to the pump with less suction.

Its smart which is what you would expect from Porsche, we need to take this knowledge on board and use it for our engines. Just a thought I know the stat is showing near the pump on the diagram it could be anywere on the engine, its just a daigram.

Anyone come up with a image of the state housing.

Good work Tapani.
Tony

[Dessertrunner]

Originally Posted by Tapani

Hi,

Is the Fig 1 in the manual page 2 in the cooling section correct?

There's an outlet fitting on the pump discharge - but it's not presented in the Figure 1, though. Both the TB and the heater core supplies and a separate bypass line seem to be tapped from the crossover pipe and all three are discharged into the pump inlet / thermostat housing (to open up the thermo).

Thanks,

Tapani

The diagram does not correctly reflect the amount of flow by the size of the pipe drawn. Also the two return pipe go tot he pump each not joined together as the diagram shows.

Tony

[Dessertrunner]

Can anyone tell me the first date that Porsche made a water cooled engine?
Tony

[Dessertrunner]

It seems that most of the thermostat the Porsche use have flap to shut off the coolant bypass. Still digging that makes sense just need more info.
Tony

http://www.ebay.com.au/itm/Porsche-9...-/350520848554

This one for the 996.

http://www.ebay.com.au/itm/Porsche-9...-/350520848554

[neverLift]

Originally Posted by Dessertrunner

Can anyone tell me the first date that Porsche made a water cooled engine?
Tony

The first 911 was 1998, the 996, though it looks like you found that! What I've heard from a friend who works at an independent Porsche/BMW/Audi repair shop, is that there were lots of issues with the 996 cooling system (along with the RMS and intermediate shaft), though I can't recall specifics. What I mean is just because its Porsche doesn't mean its right/perfect/applicable for us. Definitely worth looking at though, just we can accept things blindly.

This article discusses some issues, though mainly the liners.

The 924 was the first water-cooled Porsche, in 1976. Didn't have a flat-six though.

[Dessertrunner]

Originally Posted by neverLift

The first 911 was 1998, the 996, though it looks like you found that! What I've heard from a friend who works at an independent Porsche/BMW/Audi repair shop, is that there were lots of issues with the 996 cooling system (along with the RMS and intermediate shaft), though I can't recall specifics. What I mean is just because its Porsche doesn't mean its right/perfect/applicable for us. Definitely worth looking at though, just we can accept things blindly.

Linner issues can be caused by heat as well. The reason I asked the date was I wonder if Porsche fell into the trap that others have assumed Subaru had it sorted. You would guess the pipe feeding the pump from the tank on the diagram indicates they have caviation issues.


Okay new thinking based on todays discussion ,
cooling design system for each bank coolant out of the bank on one bank and into heat exchanger for each bank then into the radiator. The oil side is out of the dry sump pump split flow and into the heat exchange out of it and into the oil gallery on the respective bank.

This will keep oil at a constant temp and balance coolant flow to each bank. Undecided if thermostat is after heat exchanger or on block, current thinking after heat exchanger and for it to have a bypass shut off then it will stop sending coolant around the radiator when more flow is required. 2 Thermostats.
Tony

[Tapani]

The 996 was the first one, though it has had its share of teething problems none of them is ralated to the basic cooling arrangement. It is shared by more recent models too. Here you can see some pics of the Boxter:

http://www.pelicanparts.com/techarti...t_and_Pump.htm

I talked to Ross Farnham who is into using Sube engines in aircraft - also building a twing turbo EG33 for this application. He said that there are a number of people running our engines and none of them experience what we see in the last 10 years. All running thermostats in stock housings and stock water pumps/pulleys and with 19mm or 25mm radiator hoses. Oh, they all drill a 3mm hole in the thermostate as a safety catch in case it gets stuck mid air !

They run cross country races at full song for several hours.

Ross said they all have a tapped the cross flow pipe with a 3mm hose to an expansion tank and his opinion was that most probably what we see is caused by air trapped in the system. Also, he pointed out that the bypass returns are cruicial to control the coolant temperature - they are the only signal for it to open.

What do you think?

Ross also offers standalone engine management systems - for the EG33 too.

www.sdsefi.com

Kindest regards,

Tapani

PS Check this out too: http://www.subaruoutback.org/forums/...ets-print.html

[Tapani]

Originally Posted by Dessertrunner

The diagram does not correctly reflect the amount of flow by the size of the pipe drawn. Also the two return pipe go tot he pump each not joined together as the diagram shows.

Tony

Thanks,

Are the two fitting on the pump truly returns? I do not have a pump to look at, but judging from the photos and figures one goes clearly into the pump inlet and the other seems to be drilled into the discharge cavity. Is this correct?

Cheers,

Tapani

[Dessertrunner]

There are 2 pipes into the side of the pump one larger then the other, the larger of the 2 comes from the cross over return pipe at the right bank just near the outlet from the block it goes via the heater, the smaller one goes comes out of the top cross over pipe near the out let of the left bank and goes straight to the side of the pump.

In spite of the left bank having a simpler path to the pump it flows less then the one that goes all the way around through the heater. Before my thermostat opens I can get 10C difference between the 2 heads, it is pretty well allways between 3 to 5C at the best of times.

Not sure whats going on with the plane engine, it would seem they have two radiators some times. Also the higher they fly the colder it is so I not sure how that effects things.

Porsche link is interesting, Harvey and I were debating why the engine needs to run hotter, the point about poluation makes a lot of sense. Its sure not for fuel consumption as we have proven that.

Tony
Tony

[Tapani]

Thanks,

The two fittings on the pump body are not in the same location axially. See attachement. Can you verify if both do connect to the thermostat housing (pump inlet) or is the smaller one at the discharge cavity?

The small aircraft cross country racing is not that high altitude. Remember that mine spits coolant even when driving on ice !

Do you guys think that there's any possibility that this has nothing to do with thermal balance but only circulation / air trapment?

Those who have the EZ30 reference booklet 104H6 look at the Figures 72, 74 and 75. Even though the casting is of open deck type the head gasket forces the cooling flow front to rear and only then to the heads rear to front, then exit separately. There are only very small purge holes at each bore center to make certain any air / vapor is not trapped.

If you look at the EG33 open deck casting there are pockets around the head studs which might be locations for air/vapor traps - just like Boxersix pointed out in one of his earlier posts.

Another thing I am wondering about is why does the coolant spit out from the LH wheel well and not stay in the expansion reservoir? Mine never overflowed, but the coolant "leaks" from somewhere that's not really visible. When I get a chance I will inspect the expansion system for leaks / cracks or what ever. If there's a leak if will suck air into the rad when cooling and at high revs this might end up in "right" places.

Just trying to figure out what's going on

Kindest regards,

Tapani

[Dessertrunner]

The two fittings on the pump body are not in the same location axially. See attachement. Can you verify if both do connect to the thermostat housing (pump inlet) or is the smaller one at the discharge cavity?

Both are going into the pump thermo housing.

If you see coolant coming out near the left hand front tyre you could of blown the heater core and the coolant is coming from there. Other wise you overflow bottle could have a crack in it.

One of my SVX got a split on the small pipe under the intake manifold and coolant was leaking from there. I think it might have caused the engine to do a head gasket, just not sure yet. My guess is you have a leak some were.
Tony