SVX Engine cooling "Again & Again" - Page 95

bazza · #**1411** 10-20-2012, 01:02 AM

Quote:

Originally Posted by Dessertrunner

The volume of the coolant path around the linear is a lot smaller then the port size as I said. Also to get your finger in the hole you can only do it at one side of the port. The majority of that port is blocked by the linear. Looking at both port it appears to me the 22 coolant has a clearer path then the 33.
Tony

Interesting.

The EJ22 is far more restrictive and I couldn't actually get my fat thumb in. I'm thinking this may be a clue as to why the EJ's don't overheat.

EG33 block

EG33 block

EJ22 block

EG33 block

Anyway I got my 2nd spare engine going as I wanted to know what was going on in my engine and test the cams before Tony's awesome dry sump goes in. Basically used the same gauges as my testing with the pool pump and found at 7000 rpm the pre-radiator pressure (top radiator hose) was the same as the bottom radiator hose.

Check this, revving to about 6500 rpm. The gauge that says boost is the TOP of the radiator. Tony tested inside the pump hence the difference in the results. Interesting none the less.

http://www.youtube.com/watch?v=BGdNP-YWM74

1986nate · #**1412** 10-20-2012, 12:48 PM

Just thought I'd pipe in with regard to the EG33 with oil cooler. It was only in Euro markets, for sure the UK. Not sure of other markets. These are the water pumps with 3 coolant ports.

bazza · #**1413** 10-21-2012, 06:42 AM

Tested Tony's thermostat holder up top of engine in my sti, madness - usually I'd sit at 86-92c. Now it sits at 65-70'c!!!

SVXRide · #**1414** 10-21-2012, 07:35 AM

Quote:

Originally Posted by bazza

Tested Tony's thermostat holder up top of engine in my sti, madness - usually I'd sit at 86-92c. Now it sits at 65-70'c!!!

+1 for real world testing!!

Bill

SilverSpear · #**1415** 10-21-2012, 07:43 AM

Quote:

Originally Posted by bazza

Tested Tony's thermostat holder up top of engine in my sti, madness - usually I'd sit at 86-92c. Now it sits at 65-70'c!!!

Great achievement indeed...

but will your engine operate in its peak efficiency at that temperature?

Dessertrunner · #**1416** 10-21-2012, 01:14 PM

Danny we have 5 different temp options for thermostat so we will experiment to find the right temp to run the engine.
Tony

bazza · #**1417** 10-21-2012, 06:18 PM

Quote:

Originally Posted by SVXRide

+1 for real world testing!!

Bill

Thanks Bill. More to follow soon.

Quote:

Originally Posted by SilverSpear

Great achievement indeed...

but will your engine operate in its peak efficiency at that temperature?

That's a very good question although I'm not fussed about efficiency of course, rather performance... it can use twice the fuel for all I care but if I get an extra 20-30 hp at a certain temperature then perfect.

My current concern is regarding the bearing wear. Although given it's a track car, 1000 miles for example is about 2-3 years racing for me... so also not a major issue but needs to be investigated.

bazza · #**1418** 10-23-2012, 05:04 PM

While testing everything in the system I've also been trying to understand why Tony's fix, fixes the issue and why the EJ is not effected and I think I've figured it out. I'm in the process of doing a heap of reading and posting up info here: http://diyracecar.com/cooling/

However this is what I've come up with. From an engineering book on pumps:

“NPSH is Net Positive Suction Head, in simple terms we could say the NPSH is the reason that the suction nozzle is generally larger than the discharge nozzle. If there is more liquid leaving the pump faster than the liquid can enter into the pump, then the pump is being starved of liquid“

This would explain why the bigger thermostat cover works as it simply allows more water to enter the pump with less effort. A pump works by simply creating a low pressure zone in the eye of the impeller. It does not suck per say. In an open system the atmospheric pressure outside of the pump suction piping is higher and pushes the liquid towards the impeller - the book uses the analogy of explain this in simple terms - when you suck liquid through the straw, you create a low pressure in the suction line (compared to the atmosphere pushing down on the outside liquid) and the atmosphere effectively pushes that liquid into the low pressure. In a closed system the same principle applies, however the system pressure in this case pushes the liquid towards to impeller. The small thermostat cover may have been simply restricting the system pressure that could act upon the impeller.

The book also mentions one very important fix for centrifugal pumps which is exactly with what Tony discovered to work by trial and error:

* Use a larger suction flange and suction pipe.

This increases the NPSHa (available suction energy) and reduces the NPSHr (required suction energy) and for a pump to avoid caviation:
NPSHa > NPSHr + 3 ft (head)

So basically one minor fix, two major outcomes. Well done Tony.

I think it also explains why the EJ2X doesn't surge. In my testing and research I found the EJ blocks are far more restrictive than the EG blocks due to the two exits in the blocks (30% more flow through EG block exits). So the EG block is less restricted and flows more water than the EJ. The issue with this is the thermostat cover bottlenecks the flow. As I said above, the system pressure is what pushes the water into the low pressure created at the impeller. The thermostat saps the return water of energy and it drops the NPSHa below the NPSHr.

Also the reason why the big cross over pipe with thermostat up top helps (like Tom's with cross pipe) is because the pump is restricted in output due to the thermostat being up top - however the pump has a big reserve of radiator water to use - I also tested the pressure of the water at the base of the radiator with the engine off - 1 to 2 psi (cold water) - at 90'C with engine off it was around 7-8 psi simply due to water expansion. When I tested my thermostat up top, the pressure of the water just before the thermostat (next to PS pump) spiked from around 10 psi @ 6500 rpm (without thermostat up top) to about 25-30 psi (with thermostat up top) and it blew a hose off!!

Clear as mud? Good.

icingdeath88 · #**1419** 10-23-2012, 05:17 PM

^ Makes perfect sense to me. It also makes sense that Subaru would end up with this problem, due to designing the engine from the EJ22.

Out of curiousity, I tried holding the car at higher revs for a while to see what would happen. The gauge I have that reads a sensor in the rad top pipe shot up by ~15*C after about a minute at about 5500 rpms (3rd gear, just cruising, so very little load). Just wanted to see for myself I guess.

teamilluminata · #**1420** 10-23-2012, 05:20 PM

Very clear.

Just about to get my modified thermostat elbow and radiator back from TRF to bolt into my second EG33. I'm trying not to re-install the now tiny-looking thermostat but have a couple of questions:

1. How do people seal the thermostat housing without a thermostat?
2. How can I run a top inline thermostat if I only have a 6" long top hose that's bent 45 degrees?

92 SVX · #**1421** 10-23-2012, 05:25 PM

Quote:

Originally Posted by teamilluminata

Very clear.

Just about to get my modified thermostat elbow and radiator back from TRF to bolt into my second EG33. I'm trying not to re-install the now tiny-looking thermostat but have a couple of questions:

1. How do people seal the thermostat housing without a thermostat?
2. How can I run a top inline thermostat if I only have a 6" long top hose that's bent 45 degrees?

RTV silicone bead around the entire pump cover surface worked for me.

Dessertrunner · #**1422** 10-23-2012, 05:27 PM

I use a large O ring in the housing,

icingdeath88 · #**1423** 10-23-2012, 05:36 PM

Doesn't the thermostat have a gasket?

teamilluminata · #**1424** 10-23-2012, 05:40 PM

Quote:

Originally Posted by icingdeath88

Doesn't the thermostat have a gasket?

Yes so if you don't run a thermostat you don't run a gasket. It fits round the thermostat plate so wouldn't be thick enough to seal without the thermostat to support it.

Incidentally, I would have thought an o-ring would fall out without a groove for it to sit in.

bazza · #**1425** 10-23-2012, 05:45 PM

Quote:

Originally Posted by teamilluminata

Very clear.

Just about to get my modified thermostat elbow and radiator back from TRF to bolt into my second EG33. I'm trying not to re-install the now tiny-looking thermostat but have a couple of questions:

1. How do people seal the thermostat housing without a thermostat?
2. How can I run a top inline thermostat if I only have a 6" long top hose that's bent 45 degrees?

I used the original thermostat gasket down the bottom.

For the top setup - mine was a bit nasty and I haven't got around to making it "nice". Will most likely fabricate something up that can handle the required turns and pipe diameter change - plus I've got to put a bypass to the radiator due to the pressure spike I observed. However this is where it is at the moment. Goes from 34mm OD off the cross pipe to 50 mm OD at the radiator so a bit tricky.