Know subaru engine problems when tracked

[Desertrunner]

I am going to give this a go, don't have a 818 but have a lot of time spent with Subaru motors right back to the 1,300 cc in 1972.
The two big issues as I see it is cooling at high revs and oil flow.

Subaru in recent year decided in there wisdom to move the thermostat from the outlet of there motor to the inlet which leaves them and Porsche about the only car companies that do that. I won't go into it in detail at this point.
If you doubt its a problem take your 818 out on the track, watch the temp gauge, do a few hard laps then look in your overflow bottle for the radiator, it will fill up with coolant. The longer you are driving the more coolant over flows and this happens even when the engine is cold, in fact at high revs it will happen in the snow.
The problem is that the pumps are to good and with the crazy idea of the thermostat on the inlet to the water pump the pump suffers from cavitation. This then results in air in the coolant which in tern results in the volume increasing which intern blows it back to the overflow bottle. At some point you get to much air and to little coolant then its all over the engine dumps the rest of the coolant and boils. Generally it won't show up on the temp gauge till it all to late.

I have a EG33 engine and I run a thermostat on the outlet of the block, I also run temp gauge into each head at the back were the plugs are to remove the gugon pins from.

When sorted it should be possible to race all day long and get only a slight bit of coolant back to the overflow bottle if the cooling system is set up correctly

Oil problem are generally due to surging in corner and can be sorted in the usual ways.

Tony

[Mechie3]

Is your thermostat between the block and the water crossover pipe, or is it on the outlet of the water crossover pipe itself?

[Desertrunner]

Its on the end of the outlet pipe just were the hose is attached, I have made a proper housing and welded it to the pipe,
The theory behind having the thermostat down on the inlet is so you new send coolant at a constant temp into the motor.
Lets just work the logic of why it doesn't work,
If we assume to that the thermostat opens (understand it opens over 7 degrees) at 80C so the ability of the engine to cool is calculated by the flow times the difference between 80 C and 100 C. If the engine is generating more heat then the BTU's that the coolant can carry its all over. Equally as you can see there is no limit on the discharge temp, absolutely no control of the head temp, it can be all over the place. You could have a freezing cold radiator return temp and still boil the engine. You can end up with localized boiling in the head and not know.
Thermostat on the top, it will make sure that provided you have a radiator that is providing the cooling you engine is always at a constant temperature.


Top Thermostate 1 (6).JPG

[FFRSpec72]

For an road race R wouldn't the simple solution be to remove the T-Stat, with pre-grid and pace lap there is plenty of chance to get engine up to operating temperature.

[TouchStone]

The problem is that the pumps are to good and with the crazy idea of the thermostat on the inlet to the water pump the pump suffers from cavitation. This then results in air in the coolant which in tern results in the volume increasing which intern blows it back to the overflow bottle. At some point you get to much air and to little coolant then its all over the engine dumps the rest of the coolant and boils. Generally it won't show up on the temp gauge till it all to late.

Cavitation from the pump wont cause a large enough displacement to fill your overflow bottle. The vapor formed by cavitation will implode quickly and return to liquid not causing a significant volume change. I do agree that the thermostat on the inlet is a poor design decision. Since as soon as the engine is hot and the thermostat begins to open it gets flooded with cold water from the radiator and immediately starts to close again before a significant amount of coolant cycles through the engine. This is why Subaru had to run the coolant bypass hose to the thermostat to help fix this. I like your relocation of the thermostat.

[C.Plavan]

For an road race R wouldn't the simple solution be to remove the T-Stat, with pre-grid and pace lap there is plenty of chance to get engine up to operating temperature.

I thought about this, but the T-stat is supposed to slow down the flow a bit to make the water stay in the radiator longer to cool. I am installing a "Race" thermostat tonight. It opens sooner than stock.

[FFRSpec72]

I thought about this, but the T-stat is supposed to slow down the flow a bit to make the water stay in the radiator longer to cool. I am installing a "Race" thermostat tonight. It opens sooner than stock.

I run a thermostat restrictor disc in my challenge car, slows down flow but has no thermostat http://www.jegs.com/p/Moroso/Moroso-...rPage=90&pno=1

[DodgyTim]

I've previously made a simple (temporary) inline thermostat housing for an offroader by using two of these
809-10108470[1].jpg
Bolt the two together with a thermostat inbetween and connect inline the top (return) radiator hose. It cost about $20

[Desertrunner]

Cavitation from the pump wont cause a large enough displacement to fill your overflow bottle. The vapor formed by cavitation will implode quickly and return to liquid not causing a significant volume change. I do agree that the thermostat on the inlet is a poor design decision. Since as soon as the engine is hot and the thermostat begins to open it gets flooded with cold water from the radiator and immediately starts to close again before a significant amount of coolant cycles through the engine. This is why Subaru had to run the coolant bypass hose to the thermostat to help fix this. I like your relocation of the thermostat.

On the EG33 I can get it to cavitate so bad the coolant stops flowing, how I got involved was guys using these engine in buggies (sand rails) were noticing that the coolant in the radiator stopped flowing. When we did mods to correct that when they were racing for say 3 hours all the coolant blew back to the overflow bottle. As I say its just my experience.
If you move the thermostat to the top you need to make sure you drill a hole in it to let some coolant passed it while its closed that way it will see the engines actual coolant temp. Don't do it and you can boil the engine before it opens.

[Sgt.Gator]

"If you doubt its a problem take your 818 out on the track, watch the temp gauge, do a few hard laps then look in your overflow bottle for the radiator, it will fill up with coolant."

I race my Subaru Legacy GT and never have the issues the OP posted. I have never filled my overflow tank after a few hard laps, or after racing all day! I can do a 4 day weekend of Test &Tune plus 3 race days and never add a drop of water.
I've been around lots of Subaru WRXs and STIs at HPDE events and never seen this issue there either.

[walt555]

I thought about this, but the T-stat is supposed to slow down the flow a bit to make the water stay in the radiator longer to cool. I am installing a "Race" thermostat tonight. It opens sooner than stock.

Does your race t stat already have a small hole in it?

[Desertrunner]

"If you doubt its a problem take your 818 out on the track, watch the temp gauge, do a few hard laps then look in your overflow bottle for the radiator, it will fill up with coolant."

I race my Subaru Legacy GT and never have the issues the OP posted. I have never filled my overflow tank after a few hard laps, or after racing all day! I can do a 4 day weekend of Test &Tune plus 3 race days and never add a drop of water.
I've been around lots of Subaru WRXs and STIs at HPDE events and never seen this issue there either.

Some engine don't have the issue some do, the easy way to find out is to take off the overflow hose from the radiator cap and the rev the motor for a while, if its a issue you will see coolant bubble out. Simple test.

[Buzz Skyline]

Dessertrunner, thanks for sharing your experiences. Are you planning to build a an 818?

[TouchStone]

On the EG33 I can get it to cavitate so bad the coolant stops flowing, how I got involved was guys using these engine in buggies (sand rails) were noticing that the coolant in the radiator stopped flowing. When we did mods to correct that when they were racing for say 3 hours all the coolant blew back to the overflow bottle. As I say its just my experience.
If you move the thermostat to the top you need to make sure you drill a hole in it to let some coolant passed it while its closed that way it will see the engines actual coolant temp. Don't do it and you can boil the engine before it opens.

What size hole did you drill?

[walt555]

For an road race R wouldn't the simple solution be to remove the T-Stat, with pre-grid and pace lap there is plenty of chance to get engine up to operating temperature.

Wouldn't you still want a flow restrictor on the top side to keep water in the block?

[FFRSpec72]

Wouldn't you still want a flow restrictor on the top side to keep water in the block?

Yes, but not a t-stat, just a restrictor to slow down the flow

[508wrx]

Some engine don't have the issue some do, the easy way to find out is to take off the overflow hose from the radiator cap and the rev the motor for a while, if its a issue you will see coolant bubble out. Simple test.

Is this an issue with a specific engine (eg33 vs. ej205 vs. ej257) or are you saying that some engine of the same type are not consistent with their cooling?

[Desertrunner]

- Hole size in top thermostat is 2 by 5 mm
- Problem of cavitation is engine model specific, think the 6 cylinders have the issue more then any four, it is also rev dependent. If you are going for high revs you are more likely to have a problem. EG33 above 4,000 rpm continuously and its a issue. Just run a test when the motor is hot.
Yes the answer is to reduce the size going out of the block back to the radiator, so pumps flow rate is restricted to a maximum flow. Restriction/pressure kills flow which stops cavitation. This is very important, I did a trial with taking the cap off the radiator at 85 c and is the restriction was the correct size there was no back pressure. If you take the cap off at 85 C and there is back pressure your coolant is being air rated which stuffs the efficiency of the cooling system. Do a test.
- I have 4 Subaru SVX's with EG33 engines when we move the thermostat to the top the fuel economy of the cars improved in summer due to the ECU not having to add fuel to cool the charge. Gained 10% improvement.
- Not not building a 818 as the cost in Australia is high, I am building a rear engine clubman with a EG 33 and 6 speed.

Am a serious Subaru nut so just brought two of the Subaru 1235 F1 engines plus spares also brought a FJ1600 which is from Japan and its a open wheeler EA71 engine, goes like a stung cat for the size motor. Car weighs 450kg and has better the 120hp.

The engine are like this,
192erc01zvzn.jpg

FJ1600
563.jpg

[Desertrunner]

I was thinking for the guys that plan to raise the oil pressure and or dry sump the engine I would suggest you check the oil gallery going down to the crank. On the EG33 they have flow restrictors. If using the stand pump and standard flow this is not a issue but if higher flow it would be.
Oil passage 1.JPG

[Scargo]

I know this thread is a bit stale yet I want to put in my two cents. I don't know of anyone with an '04+ STi or late model WRX, who has tracked their car, who has had cooling issues as Desertrunner has mentioned. I don't see it happening unless you have a problem, such as a blown headgasket or insufficient radiator for power increases to the engine.
It may not be a fair comparison to discuss what the EG33 does relative to an EJ25*.
Again, about the EG33 oil gallery restriction, I am not sure if it is relevant to the EJ22~25* engines. Yes, there is at least one similar restriction and there are ID transitions all over the block. Some are unfathomable to me. The return from the filter is large yet it is restricted as soon as it is intersected from above and then again as the gallery goes down to the center main. I would think that Subaru engineers have done this on purpose. Yes, many race motor builders modify the passages and galleries. I do as well. I don't see a need for much fiddling when you are south of 400 CHP.