Coolant Boost Pumps

[Mechie3]

I was talking with Bob tonight and he shared his thoughts on some of the engine failures. It was sort of discussed in someones build thread, but I can't find it now. The coolant has several ways to go when exiting the pump. All of those paths are essentially the same except for the path to and from the radiator. One of Bob's theories (and it made sense to me) was the with the addition of the corrugated tubing, the extended length, that additional bends, and the additional ups and downs of the 818 radiator system that flow through the radiator is likely severely reduced as flow is instead pushed through the other channels and the coolant has the chance to become steam and cause engine problems. A picture was posted of the FFR car using an inline booster pump.

I use an electric booster on my F500. The engine is from a snowmobile and uses a single radiator in stock config. The motor requires quite a bit of cooling as it is raced in 90F weather when it was designed to run in freezing temperatures. To combat this, we run two radiators in parallel. To help flow I use a small electric bosch water pump.

Is this required? I don't know. What we really need is the ability to datalog a car with sensors pre intercooler, post intercooler, and sensors in the block for coolant and in the line going to and form the radiator. Until someone does that I'm going to assume that an inline pump can't hurt. Does anyone have recommendations. I'm seeing everything from the standard AWIC pumps, to $400 mezier and steward pumps.

[Bob_n_Cincy]

Hey Craig,
Thanks for your hospitality tonight. It was great to share ideas.
Bob

[Harley818]

Hi Mechie and Bob ,

I agree that alot of us are worried about the potential cooling problems.
I am considering AWIC just because I don't know the problem, and AWIC will probably solve it... but its expensive.

Metalmaker and others have had problems.
Is it AWIC related? is it coolant flow and temp related? what is the problem? Doing a datalog of a completed car is important to the basic configuration we are all building. If we know there is a problem, we can design an build to suit. If we don't know what the problem is, we are all taking shots in the dark.

Once we know what the issue is, we can all add AWIC, or coolant pump or whatever...... We need to know what the problem is.
Is there someone who has basic 818 set up like FFR manual that is willing to document temps and other data?

[Mechie3]

Bob,
It's always good to chat and share ideas. The best ideas often come through collaboration.

Harley,
You are 100% correct that we don't fully know what the problem is. Lot's of theories regarding intake charge temps (though a few cars with AWIC have had issues) though without data, no one knows for sure. FFR had issues with cars puking coolant out the overflow which is indicative of a cooling problem.

My line of thought is that the most expensive boost pump I've seen was $400, lots in the $200 range. A factory shortblock is $1800, new oil cooler is $250, gasket are $300. Ooof. FFR has blown a few motors, Brando blew a motor, or two, metalmaker blew a motor, I thought someone else did too. I'd rather spend $200 I don't need to potentially save ~$2500 + lots of labor time. Now, I'm not ready to dump $3k in to a dry sump system just because, but $200 seems worth it.

Until someone has the ability to datalog and we actually find out I'd rather play it semi safe.

[RM1SepEx]

Good idea to data log, how hard can it be to add sensors to data log for water temp in and out of the motor and air inlet per and post IC? How many are already there?

[Mechie3]

There is 1 sensor located on the aluminum crossover pipe under the manifold. Adding sensors is easy enough with the hardline adpaters. You just need something to actually do the datalogging with.

[Santiago]

Good idea to data log, how hard can it be to add sensors to data log for water temp in and out of the motor and air inlet per and post IC? How many are already there?

Well, the most obvious answer is: FACTORY FIVE

C'mon, I know this is a kit car, and I know we're all "builders" but this seems like the sort of thing the manufacturer should be diligently working on. FFR has got far more resources, equipment, and experience than just about any of us. Their mules are already carrying data-logging capable dashes (on the R-cars), and they've already had engine woes, so I think it reasonable that they should be taking the lead on this inquiry.

This is a pretty serious matter that goes to the core of the car's reliability...and ultimately its fledgling reputation. One guy blowing an engine is nothing new. A few of the first guys to complete the kit doing it, that's another thing. FFR, please put somebody on this and let us know what is (or has been) found.

In the meantime, I'm sure some other folks will start to look into it, but we would be better off if a couple folks could do it.

Best,
-j

[Mechie3]

^^ We had a very similar conversation last night. If the issue is oiling and the solution is a $3k dry sump, that's a huge market turn off. If the issue is an AWIC and it's $600 to build it yourself and $1400 to buy a kit, that's still an issue but depending on your fab resources is manageable, If the issue is coolant and it's a $200 pump? Meh, not a big deal. Problem is, until we figure out what it is it could be any, or all three, and then it starts to become an issue for the kit.

I just reread the UTCC thread from 2014. The red car had cooling issues at limerock (supposedly bypassed hoses), then had issues on the first timed session that was attributed to a bad radiator cap. They went out again and had the same issue with water from the front spraying and determined it was a headgasket. They then took the bluecar out and it had issues with the cap on the coolant tank and they attributed that to just a loose cap. I haven't seen that many issues with Subaru radiator caps going bad in my 8 years in the Subie world. At VIR they blew up both cars (one overboosted, another lost coolant through the banjo bolt and it lost compression). Issues with cooling seem to pop up often and it makes me wonder if there isn't something else to blame aside from the easy target of "we just had a loose cap, no problems here". I'm not road racing mine (not at first) so I'm not terribly concerned, but after having rebuilt many motors I don't have the desire or time to rebuild more. Still, I want the car reliable.

[RM1SepEx]

I dramatically changed my build schedule and philosophy due to the multiple booms... I've addressed the oiling issues with an sti pan etc and will run more oil, my side mount IC should fix the intake temp issues. If I was to road race a dry sump would be an excellent idea since the inherent design of a flat four is really an issue with high cornering loads. I would not road race a flat four w/o a dry sump. For short duration high lateral G forces in autocross and on the street we should be OK.

Cooling issues too? I don't see where we should have problems due to the design there. The long tubes are essentially more radiator, unless it just can't keep up with the engine heat. Smooth coolant tubes should decrease flow losses but it seems like FFR and others have extensive experience using these types of tubes in such an application. Coolant flow isn't high volume or pressure so losses shouldn't be significant.

[Mechie3]

I don't know if there are or aren't, but for $200, I don't see a reason not to put a pump in especially if FFR just did on their red car. Just wondering if anyone has experience with certain pumps vs others.

[Scargo]

I am worried about mass hysteria here. I and my friends have been tracking 350~400 WHP STis at the racetrack for years without too many failures. Some, with no failures for years! Heating was not an issue. Head gaskets were/stock pistons were. Overboost was.

As RM1SepEx (http://thefactoryfiveforum.com/member.php?4528-RM1SepEx) said, the flex tubing just adds radiator. In my mind, as well. JeromeS13 (http://thefactoryfiveforum.com/member.php?7233-JeromeS13)has posted about pumps and gave links to testing of various water pumps, if you want to pursue that.

I think the Subaru motor is a sound platform. There are growing pains with this combo, no doubt. There are people rushing to get to the track with the 818 and little understanding of the motor and its idiosyncrasies. No... you don't have to have a dry sump.
Yes... you need a good bottom end and tight clearances for the oil pressure.
When you start pushing that old donor motor, all of these factors become more that a re-tread, 75K stock motor can take. Don't be silly.
My engine builder (a mechanical engineer that designs engines) said "Look up xxx and do some reading." I recommend that for those that wring their hands about the boxer motor.

[Wayne Presley]

We ran the prototypes for 4 solid days with no overheating issues, I've run 4 cars on the dyno for extended periods with no overheating. The last one was an 410 RWHP car that I could reach down and touch the coolant return line with my bare hand. All of them got to the thermostat temp and sat there.

[Jim Schenck]

Factory Five has never had any coolant system operation issues with our cars, that is why we haven't pursued any other avenues to try and change the cooling system. Of the three engine failures we had two of them already had an additional pump in the system, however the pump was never put there to help with keeping the car cool once its running, in both cars it was put in to speed up the air bleeding process if we have to do it at the track in a hurry. We ran the blue car and the red street car for many test sessions with no auxillary pump and nothing special at all with the cooling system and never once had an issue. At the 2013 UTCC it was over 105 degrees for the whole day and the blue car never once had a cooling issue and at the time it still had a completely stock system

The three engine failure we had were as follows:
Red race car engine number one: Blown head gasket due to pushing the limit of boost on a 175k mile engine. If you watch the video where that car blows coolant out of the rad cap the temp is just barely over 200 degrees, not enough to boil over but it blows the cap due to compression leaking past the gasket into the cooling system.

Red race car number two: Over boost, thrown rod. This car never got above 180 degrees even on the dyno.

Blue race car: Cooling hose to the turbo came undone at the banjo fitting and drained the entire system of coolant. Engine seized after running dry for about half a lap or maybe more. (I realize this is part of the cooling system but this is a preparation issue not an issue with the design)

Both the blue car and the sliver car (previously red) have an auxillary water pump to help bleed the system. In my opinion these are not worth carrying around the extra weight unless you will always be rushed during the coolant filling process.

[Mechie3]

I'm not doubting the the soundness of the boxer platform (I've driven them for years and rather like them)
Not trying to incite hysteria, just promote discussion as there have been several failures with very quick root cause analysis that were fast to dismiss any deeper problems such as "oh, it was just this" but then the failures continue after that was fixed. It makes me wonder if we aren't fixing symptoms and not true root cause. I'm just asking the 5 why's and wondering if there isn't a different issue potentially) such as the heads lifting and pushing coolant out http://www.iwsti.com/forums/2-5-liter-litre-factory-motor/95480-update-overheating-pushing-coolant-change-head-gaskets.html Edit: Jim just alluded to this on one of their cars

The flex tubing adds radiator, yes, but also adds turbulence, length, flow restriction. Have people with freshly rebuilt motors blown them up? Yes,. think everyone (including FFR) rushed to get to the track (it seems every high profile event they attend they blow up a car, or two, always in a different manner). The concern (rather I should say, area of interest, as the word concern seems to incite riot and not promote discussion) is the difference between the factory setup that has been road raced (like Scargo has) vs the setup in the 818. The very fact that many Subaru's have been raced without failures in general make me wonder why the 818 platform is experiencing trouble.

The system was designed to balance flow to all areas with the radiator being 6" from the pump/motor. In the 818 setup there are several more bends, ID changes, elevation changes, and length increase to the radiator circuit only. The factory pressure cap is rated for ~16psi. Simple online calculators with the assumptions of a 1.5" diameter coolant line, 16 ft of line, 100gpm, and assuming aluminum piping gives a pressure loss of ~5psi (meaning to flow the same as a stock Subaru arrangement, the system needs to provide 5psi more, or 1/3 more pressure). Add in the sudden expansion and contraction losses (FFR doesn't provide any smooth transition couplers) and the flow to the radiator is reduced even more while flow to other sections of the motor (ie, heater core circuit) are increased. Does the flex add radiator? Sure. But if it doesn't flow enough, it doesn't matter. Cooling capacity and the capability of the system to utilize the capacity are different.

Lets say all of that is moot. Fair enough. But I'd rather discuss it and then rule it out rather than ignore it preemptively. Without a running car or datalogging I can't really test the theory and can only rely on anecdotal evidence until someone that has datalogged or FFR chimes in.

Edit: Jim, where are you measuring coolant temperatures from? Also, now that we have input from FFR as to why they added the aux pump there's less reason to believe it is a problem, though I'll wait and let others be the guinea pigs for a while. ;)

[Jim Schenck]

I don't agree at all that issues were quickly dismissed, in every failure we dissassembled the engine looking for any signs of additional failures or any possible causes. We also poured over the data from the Racepak comparing it with exact on track locations, trackside observations, driver feedback, and even in one case we called another driver on the track after the fact looking for info about what the car was doing when it passed him 30 seconds earlier.

In every case there was a clear explanation, and none of the three had any common symptoms or failures, and I can't state this enough times, none of them ever ran hot before the failure occured.

Bottom line is we pushed the limits of preparation time over the summer trying to get as much in as we could, and at the same time trying to push what has been done with the platform. Pushing the platform is something I still think benefits everyone, however doing it in private testing and not high profile events is certainly a better way to go.

[Mechie3]

Jim, I don't mean yours. There have been several failures of customer built cars. I'm not trying to PKI t fingers just starting a conversation where we can talk about potential weak points and how to address them. Its what any engineering firm does when they do a dfmea. They aren't pointing blame but are looking at potential failure modes, causes, effects, and then measuring the probability of those failures to see where to focus efforts.

[metalmaker12]

Hi Mechie and Bob ,

I agree that alot of us are worried about the potential cooling problems.
I am considering AWIC just because I don't know the problem, and AWIC will probably solve it... but its expensive.

Metalmaker and others have had problems.
Is it AWIC related? is it coolant flow and temp related? what is the problem? Doing a datalog of a completed car is important to the basic configuration we are all building. If we know there is a problem, we can design an build to suit. If we don't know what the problem is, we are all taking shots in the dark.

Once we know what the issue is, we can all add AWIC, or coolant pump or whatever...... We need to know what the problem is.
Is there someone who has basic 818 set up like FFR manual that is willing to document temps and other data?

In stock tmic and the 818s with windshield it has been proven you will need to run an awic to get safe intake air temps. The air to air can work, but major modifications to the body and ducting will be needed along with most likely losing the full winshield. The hardtop might fix this.


As far as the cooling system, I am really unsure if there is any real problem with it. My car experienced a coolant leak because the lower radiator hose made contact with steering rack boot and it created a leak. I was loosing coolant and sucking in air through serveral continuous runs and also my iats got way out of control. My coolant temp never got to the over heated state. But both head gaskets went and #4 piston ringlands has some failure. It is something that can happen on any car. I since moved to solid coolant lines, which move the line away from the rack more, and I also put larger spacers under the radiator for even better clearance. Once up and running I am going to log it often and check multiple spots of the charged air and coolant systems temps.There might be something that has been missed, but it could just be that the cars were put together quickly and simple stuff was missed.

I drove my 818 all around pushing it pretty good for 1,000 miles prior to autocross. So the problem is when you push it hard for sure. This spring I will be testing it out. I will keep you updated.

[D Clary]

I do not have a lot of Subaru experience, that been said I do have quite a bit of experience with racing and engines. The Subie engine is very sound with the possible exception of the "G" forces on the oiling system which is yet to be determined. Looking at the cooling system in my 818 I see absolutely no reason for any concern about heating. The main issue is greed and the ease of horsepower through the turbo. My policy is build a 400 horsepower motor and dial it back to 350. Les boost, less timing and richer mixture. My 4 cyl Mustang turbo runs 18 psi boost on pump gas. I have been running this way for three years of auto cross and track days. I use the megasquirt ecu. It has great timing and mixture control. I expect to have the same result with Waynes ECU and the Subie motor.

[Scargo]

A lot of this goes to what I was saying. We're building a kit car. There are lots of miss-steps, variables and little is proven other that we are wild and crazy guys who like to take the path less traveled.
Almost everything is anecdotal and there's not a common link, as I see it, other than many do not know the needs of a Subaru motor and start off with a tired, stock motor, push it hard and expect it to perform like a Porsche race motor.

Lifting of heads is classic and well documented. You need good gaskets and ARP studs. You need forged pistons. You need an 11mm oil pump. You need an oil cooler, a good intercooler and possibly water spray or AWIC. Some factory Subaru cars came with water mist for the intercooler and heat-soak is a well known problem. IMHO, having the AWIC anywhere over the hot engine is a bad idea.

[RM1SepEx]

increased boosting a used engine with unknown history is a crapshoot, higher g forces will cause oil to slosh around, very common issue for all motors, esp a boxer where the oil goes to the heads, easy to resolve. No one should do any racing (even autox) with a stock 02-05 wrx oil pan. Extra oil won't hurt. I'm crossing my fingers and hoping my luck is good with a 70,000 mile donor but if it blows in a week... I'll just have to accept it, smile and review options! JDM???

Flow rates are not enough to be a real issue in the water cooling, just be sure to burp well. It's easy to miss something building your own car... expensive OOPS will happen.

FFR's cars were raced, seriously stretched. When racing, **** happens! ;)

[Mechie3]

My donor had 70k miles. It the motor had 20k since i last rebuilt it. It ran 21psi and ran well in the WRX. I just don't feel like rebuilding it for a fourth time.

[Santiago]

Thanks all for input on the discussion so far. If nothing else it reinforces the need to be very careful with the initial prep (no rushing) and to go over the car very thoroughly after several of the initial runs. Looks like developing a very thorough pre/post-run checklist is in order.

Best,
-j

[Turboguy]

In every case there was a clear explanation, and none of the three had any common symptoms or failures, and I can't state this enough times, none of them ever ran hot before the failure occured.


Out of curiosity, where in the system is coolant temperature sensor(s) located?

[JeromeS13]

Out of curiosity, where in the system is coolant temperature sensor(s) located?

It's in the upper crossover pipe, just before the coolant exits the engine and enters the upper radiator hose going to the radiator. This is when the coolant is at its hottest.

[metalmaker12]

Like a said up above I am going through the car with a detailed punch list and documented data before I ever really boost it.

My current punch list:
Psi check awic core again to 50psi
Go over all vacum lines again
Check all sensors and plugs are tighten to spec
Go over Aos routing and check for any problems
Check all torque specs on intake/ exhaust manifold, etc
Make sure oil press is up to 15-30psi to start.
Burp coolant system and once running check for hot/cold spots
Install awic and Run/bleed coolant
Install motul gearbox fluid
Check over all inlet, intake and IC tubing to check it's all tight and secure
Tune for larger injectors and mbc
Eliminate factory fuel pump controller for better amp control of aero 340 pump
Run a safe break in tune for like 1,000 miles, with break in oil for 50-80 miles than Rotella T6
Etc etc etc
I am missng something, but you get the idea.
Did I mention I am pretty much done, lol for all of you building, but this all does not take too long

[BrandonDrums]

Not owning one of these cars (but wanting to) I hope we get some stuff nailed down. I'm counting on you guys!

I would certainly say a booster water pump would make sense. That's a lot of piping to push a pull water through when the OEM water pump is designed for a setup where the radiator inlets and outlets are literally inches away from the engine.

Datalogging is one thing but just having some analog gauges would be better than nothing for getting water and intake temps at different locations. It would be like an aircraft where you just have tons of data on the dash. Could look cool and it most likely would be less expensive than getting a full computerized data logging setup that could compile that data alongside the ECU's logs. Air temp is hard to find but water temp ones can be had for $30 each.

[wildhorsesracing]

Having raced 600+ hp Mustangs and assorted BMWs over the years I had to deal with cooling issues often. One thing I did was to remove the center of the thermostat and simply put the "washer" that is left back in. It has always worked well to improve flow and cooling, has anyone done this with their 818?

[Jim Schenck]

Jerome is correct for our coolant temp sensor, some were in the crossover and others were in a tee coming off the line coming out of the crossover, where the coolant should be about as hot as it will get.

We also have checked the coolant temps at different spots with a gun while the car was on the dyno, just looking for hot or cold spots etc.. Not something you can do on the street but if you have an extra guy at a dyno session it is pretty easy.

Some of our cars have run with the thermostat gutted but the washer left in as a restriction, that is why the red car temps ran below 180 at the last event and on the dyno. Doing this was something we thought would help burp the air out of the system but it really didn't make much difference.

[Mechie3]

Thanks for the input Jim.

[Zach34]

One of Bob's theories (and it made sense to me) was the with the addition of the corrugated tubing, the extended length, that additional bends, and the additional ups and downs of the 818 radiator system that flow through the radiator is likely severely reduced as flow is instead pushed through the other channels and the coolant has the chance to become steam and cause engine problems.




I would certainly say a booster water pump would make sense. That's a lot of piping to push a pull water through when the OEM water pump is designed for a setup where the radiator inlets and outlets are literally inches away from the engine.

I'm not sure this is a correct line of thought. The whole flow restriction principle sounds like a compressible gas phenomenon - exhaust for example. We're all used to the concepts of flow restriction with exhaust systems - you want smooth bends, few of them, and a short-length system overall. That's because gas is compressible and we're trying to avoid pressure differentials throughout the exhaust tube. Water is not compressible. The coolant system is hydraulic. As long as there are no leaks, you get the same pressure at the end of a tube as you have at the beginning. In a closed-loop system, elevation changes don't matter because any hill that is climbed is also descended on the way back to the origin. There's probably some friction involved with the long corrugated tubes, but we're talking about water here - couldn't be THAT much friction.

I'm not an engineer, and I could be wrong, but I doubt the long corrugated tubes are affecting flow much.

[DodgyTim]

As long as there are no leaks, you get the same pressure at the end of a tube as you have at the beginning. In a closed-loop system, elevation changes don't matter because any hill that is climbed is also descended on the way back to the origin. There's probably some friction involved with the long corrugated tubes, but we're talking about water here - couldn't be THAT much friction.


Thats only true when the water is not flowing. Once it starts to flow there is a pressure difference from one end to the other. Diameter is the main factor. Simplistically friction loss in a length of pipe is

= (friction factor x length x velocity squared) / (2 x g x diameter)

when you make the pipe small,for a set flow rate the velocity goes up, and the friction loss goes way up.

The corrugated tube raises the friction factor. When you get into the maths of flow in corrugated pipes, it gets really complicated and theoretical really quickly:cool:

My gut feel is the thermostat opening would make a bigger difference than the corrugated tubing:confused:

[Junty]

Hey guys, My 818s has just returned from a week at the Dyno. It required a little more to get things going, as I started with a base MAP in the M800 Motec - which really has no targets. So the 818 off and on spent about 8hours running during the past 10 days, lots of stop starts and overnight cold starts to get everything just as it should be. Engine temperature is certainly the least of our problems down south with a 2.5L JDM engine. My donor engine is however near new only 14K miles before being moved to 818. The only modification to this engine was adding new fuel rails to get ride of all the aluminium spaghetti that these 2012 stock engines come with, and completely removing all the TGV valve assemble (allowing more inlet flow). Running at a very conservative 13PSI I've ended up with a tune that quickly pulled to 345 WHP and 460ft pb torque.
Feedback from the tuner:- Engine temperature and cooling system is perfect, engine gets up t temp, thermostat opens and remains at constant perfect temperature. From my limited experience fans on a dyno can contribute to slightly less airflow than open air driving. So I'm very confident with cooling system.
The tuners did try running a little higher boost - however the inlet air-temp (top mount IC) did climb and with increased boost lead to early (slight) engine detination.
As we should all appreciate - air temperature as it increases, and you increase boost = a whole lot less O2 getting into engine and detination (big contributor to engine failure) starts.
I have some excellent numbers, with very conservative boost on standard pump gas. I'm going to run it this way for a long while before eventually adding AWIC before any more boost.
This entire discussion leads me towards wanting to know all the differences between USA and JDM Subarus? Why does a JDM engine go so well?

[Scargo]

Water pump, Subaru PN 21111AA026 is closed vs the typical open pinwheel looking one. It is supposed to be better. There are expensive, supposedly tweaked "high performance versions of it, too.
35977better VS OK? 35978

While I'm at it, I have done passage matching on my EJ 257, though I cant report what, if any, effect it has had. I haven't run it yet. Let's just say that if you line up the head gasket on the head and also on the case you will see that there is room for improvement, enlarging and smoothing (at least on mine).

[metalmaker12]

And the heads are way better stock, bigger ports, better parts. Did I mention the engines are pretty much blueprinted and the tolerances are just spot on compaired to the US stuff. Guess they want the best for themselves. Kinda why I used one.

[metalmaker12]

Hey guys, My 818s has just returned from a week at the Dyno. It required a little more to get things going, as I started with a base MAP in the M800 Motec - which really has no targets. So the 818 off and on spent about 8hours running during the past 10 days, lots of stop starts and overnight cold starts to get everything just as it should be. Engine temperature is certainly the least of our problems down south with a 2.5L JDM engine. My donor engine is however near new only 14K miles before being moved to 818. The only modification to this engine was adding new fuel rails to get ride of all the aluminium spaghetti that these 2012 stock engines come with, and completely removing all the TGV valve assemble (allowing more inlet flow). Running at a very conservative 13PSI I've ended up with a tune that quickly pulled to 345 WHP and 460ft pb torque.
Feedback from the tuner:- Engine temperature and cooling system is perfect, engine gets up t temp, thermostat opens and remains at constant perfect temperature. From my limited experience fans on a dyno can contribute to slightly less airflow than open air driving. So I'm very confident with cooling system.
The tuners did try running a little higher boost - however the inlet air-temp (top mount IC) did climb and with increased boost lead to early (slight) engine detination.
As we should all appreciate - air temperature as it increases, and you increase boost = a whole lot less O2 getting into engine and detination (big contributor to engine failure) starts.
I have some excellent numbers, with very conservative boost on standard pump gas. I'm going to run it this way for a long while before eventually adding AWIC before any more boost.
This entire discussion leads me towards wanting to know all the differences between USA and JDM Subarus? Why does a JDM engine go so well?

460 wheel torque @ 13 psi??? I am a bit shaky on that. Show sheets!!
What's your setup, turbo injectors etc

[Hindsight]

Agree.... Even on ethanol that seems quite high. Would be very interested to hear more about the setup.

[Jaime]

345/460 means that the torque peak is well below 5250rpm. That sounds like either an undersized turbo or a very long stroke. Neither sounds EJ20-ish.

[Wayne Presley]

Sounds like he's running more boost down low and then tapering it down at higher RPM.

[C.Plavan]

I'm not too concerned with the cooling system. I was worried about the other things (TMIC, Dry Sump). I'll know in less than 2 weeks when I test the 818R on the track. It's sad when my buddies and I are already joking about blowing a motor. So hopefully, none of that will happen and I will have good things to report after some hard driving.

[BrandonDrums]

I'm not sure this is a correct line of thought. The whole flow restriction principle sounds like a compressible gas phenomenon - exhaust for example. We're all used to the concepts of flow restriction with exhaust systems - you want smooth bends, few of them, and a short-length system overall. That's because gas is compressible and we're trying to avoid pressure differentials throughout the exhaust tube. Water is not compressible. The coolant system is hydraulic. As long as there are no leaks, you get the same pressure at the end of a tube as you have at the beginning. In a closed-loop system, elevation changes don't matter because any hill that is climbed is also descended on the way back to the origin. There's probably some friction involved with the long corrugated tubes, but we're talking about water here - couldn't be THAT much friction.

I'm not an engineer, and I could be wrong, but I doubt the long corrugated tubes are affecting flow much.



Thats only true when the water is not flowing. Once it starts to flow there is a pressure difference from one end to the other. Diameter is the main factor. Simplistically friction loss in a length of pipe is

= (friction factor x length x velocity squared) / (2 x g x diameter)

when you make the pipe small,for a set flow rate the velocity goes up, and the friction loss goes way up.

The corrugated tube raises the friction factor. When you get into the maths of flow in corrugated pipes, it gets really complicated and theoretical really quickly:cool:

My gut feel is the thermostat opening would make a bigger difference than the corrugated tubing:confused:

Also, heat differentials can cause small pressure changes in a fluid. Fluids aren't easily compressed but that's not to say that they don't expand under heat like everything else. Not to mention that corrugated piping laid parallel to the ground leaves a LOT of places for bubbles to get trapped. It seems you'd have to burp the system with the corrugated pipes standing vertically to ensure everything is worked through when standing still.

Perhaps it makes sense to add a small but tall coolant reservoir in front of the engine with a vented cap. Not unlike the turbo coolant reservoir. Just having a high point at the end of the tract before entering the engine could solve a lot of vapor issues as well as act as a coolant 'surge tank' of sorts to ensure coolant flow if g-forces or changes in pressure interrupt flow for any reason. It would have to be either baffled or have staggered inlet/outlet points of course to allow the bubbles to rise under high flow but that doesn't seem to complicated.

Not that flow would be interrupted but seems like a simple stop gap measure to deal with one or two potential issues.

[Bob_n_Cincy]

Red race car engine number one: Blown head gasket due to pushing the limit of boost on a 175k mile engine. If you watch the video where that car blows coolant out of the rad cap the temp is just barely over 200 degrees, not enough to boil over but it blows the cap due to compression leaking past the gasket into the cooling system.

Jim,
When this red car blew its head gasket, I see 2 numbers on the video that are out of wack.

The boost pressure was above 40 psi. I read somewhere that the vacuum hose going to the wastegate got melted and caused this. Is this correct?

The other number was the "air intake temperature" above 200 degrees.
The stock sensor is right after the air filter so I know it wasn't located there.
Was the AIT sensor located before or after the intercooler?
Thanks,
Bob

PS. Thanks for explaining that the thermostat was gutted. That explains why the motor was slow setting up to temperature.

[metalmaker12]

I'm not too concerned with the cooling system. I was worried about the other things (TMIC, Dry Sump). I'll know in less than 2 weeks when I test the 818R on the track. It's sad when my buddies and I are already joking about blowing a motor. So hopefully, none of that will happen and I will have good things to report after some hard driving.
I would take it easy and really watch the iats. That is the motor killer. The cooling system seem good to me, maybe even to efficient. You should be fine, and actually I want your motor to last and last, cause we need success at this point.

[Jim Schenck]

Jim,
When this red car blew its head gasket, I see 2 numbers on the video that are out of wack.

The boost pressure was above 40 psi. I read somewhere that the vacuum hose going to the wastegate got melted and caused this. Is this correct?

The other number was the "air intake temperature" above 200 degrees.


The stock sensor is right after the air filter so I know it wasn't located there.
Was the AIT sensor located before or after the intercooler?
Thanks,
Bob

PS. Thanks for explaining that the thermostat was gutted. That explains why the motor was slow setting up to temperature.

Bob,

The blown head gasket occured in the first engine, a 175K mile donor 2.5 out of a Legacy GT we were running at 23 lbs of boost. This is not the same engine we had the overboost issue with. I think a lot of concern has been generated because of this failure but the results really were not unexpected, we hoped to make it one hot lap to get a time in but we ended up making it 2/3 of a lap 3 times. As to why the heck we would go the track with such a ticking time bomb, thats a long story, but one that I can't spill the beans on yet and ruin what is coming but an explanation is on its way. I do promise though it was not our original plan or even our backup plan and not also just sheer ignorance!;)

The second engine was the one that had the small fire and overboost. After dissasembly I don't honestly know what order those things occured in, however I am sure that one of them led to the other. The crazy high IATs on that engine on that hot lap were for sure a problem, something that had not shown up on the dyno. We added a reservoir to the system between the dyno and the track which maybe was the cause of the problem (I always look at the most recent change first and this was really the only change between the dyno and the track.) We tested the system with the tank for flow at the shop and it was working but during the run the speed at which the IATs cllimbed and the volume of fluid in the system makes me think the fluid just simply wasn't circulating enough. The tank was hot after the run so there was at least some circulation.

The IAT sensor is in the tube between the cooler and the throttle body, so those are post cooler temps. We were seeing 130s on the dyno, which is why we added the tank, plus we were considering running ice just because of the time trial nature of the event. We also thought that the stationary dyno pulls were not allowing the heat exchanger to work to its full potential and we would have better cooling once the car was at speed.

Without the thermostat the engine never got above 180 degrees during a whole day of running on the dyno, it wasn't just slow to warm up, it stayed at those temps under full load and not great airflow.

[Mechie3]

Jim,
What size heat exchanger were you using? I recall seeing photo of one about half the width of the radiator, but don't know what car it was on. I also remember seeing a very large AWIC core on one of the cars. Was this with a stock turbo?

Thanks.

[ssssly]

As far as the previous logging discussion. Get a tractrix cable and an old laptop.

Can log every single sensor in the engine. More data than you can shake a stick at.

From everything I have seen however, the coolant system isn't much of a problem. With the corrugated pipes, just bleed the system thoroughly to avoid air bubbles in the ridges.

I would suggest parking the car on a slight hill, bum up, to get the filler tank good and above the radiator. Open the filler tank and jam a tall skinny funnel in it and fill until the entire system and the funnel is half way filled. Start the car and rev it slowly.

If you get a good rite you will see the fluid in the funnel sucking into the engine and then spitting all the bubbles out as it fills again.

If you have big enough air pockets it will suck the funnel dry. Just fill it up again.

(Note: Wait until the fluid cools before pulling the funnel out. And its going to make a bit of a mess. So you might want to throw some rags or nasty towels under the tank.)

Air bubbles can cause cavitation in both the impeller and the tubing causing premature pump failure.

The far bigger concern of mine is the IATs. High IATs will blow a boxer very quickly.

I strongly suggest everyone tune their engine to pull timing if IATs are over 120 and to go into full limp over 150 or 160 (depending how brave you are).

Hopefully I will get my 3d printer and scanner soon so I can start working on the air diverter vents. Car definitely needs more air to the TMIC if running more than a td04.


metalmaker12: I wouldn't suggest going straight MBC unless it is going to be a dedicated track car. Twinscrolls are particularly prone to PTFB issues with MBCs. The last thing you want is full boost at low RPM going up a slight hill in a high gear. There is no way to tune for it and things will pop. Quickly.

I would suggest a hybrid EBC/MBC setup. Works quite well when set up properly with one way check valves.

[Bob_n_Cincy]

As far as the previous logging discussion. Get a tractrix cable and an old laptop.
Can log every single sensor in the engine. More data than you can shake a stick at.
I strongly suggest everyone tune their engine to pull timing if IATs are over 120 and to go into full limp over 150 or 160

Hi ssssly,
I am running a stock 04 FXT engine. My IAT sensor is before the turbo so it doesn't get hot.
I can add in another IAT sensor but I have no way to log it.
What does it take tuning wise to change my IAT to post turbo?
Bob

PS: I had to look up all the acronyms you used in your comments to Metal.

[Bob_n_Cincy]

Thanks Jim for your response in post 43 above, that clears some things up for me.
I am going to work on my AWIC System.
Bob

[Scargo]

Per Jim: "As to why the heck we would go the track with such a ticking time bomb, thats a long story, but one that I can't spill the beans on yet and ruin what is coming but an explanation is on its way."
The saga continues. What can they be up to?

[Jaime]

I can add in another IAT sensor but I have no way to log it.
Sponaugle's recommendation was to wire the second IAT sensor to the TGV sensor, that way you can log it with ROMRaider.

[Bob_n_Cincy]

Sponaugle's recommendation was to wire the second IAT sensor to the TGV sensor, that way you can log it with ROMRaider.

Thanks Jamie,
Right now I am still using my TGV. I did look a my schematics and found the only analog sensor I'm not using is the fuel tank pressure sensor. So maybe I can use that one.
Bob

[Mechie3]

re: datalogging, I wanted more data that just what the ECU spits out. Post turbo temps, post IC temps, water temp at radiator (vs water temp at engine as well). It's good to have the ECU data, but it doesn't always tell you the full story (such as, how efficient is a certain system).

[metalmaker12]

As far as the previous logging discussion. Get a tractrix cable and an old laptop.

Can log every single sensor in the engine. More data than you can shake a stick at.

From everything I have seen however, the coolant system isn't much of a problem. With the corrugated pipes, just bleed the system thoroughly to avoid air bubbles in the ridges.

I would suggest parking the car on a slight hill, bum up, to get the filler tank good and above the radiator. Open the filler tank and jam a tall skinny funnel in it and fill until the entire system and the funnel is half way filled. Start the car and rev it slowly.

If you get a good rite you will see the fluid in the funnel sucking into the engine and then spitting all the bubbles out as it fills again.

If you have big enough air pockets it will suck the funnel dry. Just fill it up again.

(Note: Wait until the fluid cools before pulling the funnel out. And its going to make a bit of a mess. So you might want to throw some rags or nasty towels under the tank.)

Air bubbles can cause cavitation in both the impeller and the tubing causing premature pump failure.

The far bigger concern of mine is the IATs. High IATs will blow a boxer very quickly.

I strongly suggest everyone tune their engine to pull timing if IATs are over 120 and to go into full limp over 150 or 160 (depending how brave you are).

Hopefully I will get my 3d printer and scanner soon so I can start working on the air diverter vents. Car definitely needs more air to the TMIC if running more than a td04.


metalmaker12: I wouldn't suggest going straight MBC unless it is going to be a dedicated track car. Twinscrolls are particularly prone to PTFB issues with MBCs. The last thing you want is full boost at low RPM going up a slight hill in a high gear. There is no way to tune for it and things will pop. Quickly.

I would suggest a hybrid EBC/MBC setup. Works quite well when set up properly with one way check valves.

Thanks for your input but I have used, mbc,ebc, hybrids with success and reliability on twin scrolls and non. You just gotta know how to tune it correctly for each one. I have never had an issue with mbc and all my subarus have had them. I am also not going to run as high a boost as I would normally so peak boost of 16 all the time is not a concern. It's all in how you skin it and like your boost. I like the boost there all the time, it does not have to be really high, just there. I like 25% throttle and full boost on the ej207 running avcs with a 1.5 dom. I can run lower peak boost and feel power sooner and for longer. You just have to complete the map correctly. I think I can get away with 14 psi- 16psi on this setup with the avcs and timing set just right to be plenty of power. The dyno will tell the story. If it feels like too much response than I will hybrid, but I always start off with what I like most first. I feel from my years of owning and playing with these cars that a mbc can be as reliable/safe as any setup, and the most responsive.
Keep in the loop, my car is about a week away from first starting (for second time). And I did not over boost first time and it was the Sti oem soleniod lol
I just have to set up awic and run coolant with My vacum filler.

[ssssly]

For an EJ207 with that size turbo I generally put an 11lb spring in the wastegate and run on spring pressure bellow 50% throttle. Below 11psi, even on pump gas, you can still run pretty aggressive timings below 3k. Allows the turbo to stay nice and spooled up down low with the a good failsafe against ptfb related engine failures on the road.

I then run 75% wgdc at 60% throttle, 100% wgdc above 70% throttle with the MBC setting max boost. I set my boost tables 25% higher than whatever the MBC is set for. And a one way check valve before and after the MBC. So if either the MBC of EBC fails it failsafes to the other. Then on the load table I always run a row above whatever the max load is, at full target boost, and run it rich as hell.

So if it ever overboosts there is plenty of cooling fuel to keep the detonation down.

I'm also running 24 PSI though. So if I have full boost at 4k and low load, is a way bigger deal than you. The ecu can probably get close to pulling enough timing for 5 psi of overboost at low load. Can't come close at 13 psi of overboost. Tuning for 16psi at low load will create a weird revers hump in your timing map though. Haven't tried it, but I wouldn't necessarily suggest it. Could do weird things in shifts and when lifting off throttle going up or down hill. But again, never tried it so it could work great.

I also tune everything to 85% though. If I can't get the power I want out of 85% of current setup, I change the setup. I don't mind blowing stuff up at a track, but I prefer not to. And I definitely don't like blowing things up driving around town.

I have found running like this makes the car feel way more like a turbo charged 6 than 4. The power build linearly with how much throttle you apply. As opposed to mashing the throttle and waiting for the turbo to spool. Unless you mash it off the line. But with a twin scroll turbo and aggressive AVCS and cams, you are only waiting 3k RPM for spool.

Looking forward to seeing you up and running again. Particularly interested in how you get the power to the ground. As a guideline to how far I am going to need to go with suspension and body mods for humongous rear tires.

[Scargo]

ssssly, you said "For an EJ207 with that size turbo". I can't find what you're referring to, though perhaps it is irrelevant to my comment/question. I'm trying to pick a turbo and am worried about excessive wastegating when trying to stay at a 8:1 P/W ration for ST-2. You bring up some interesting ideas.
Could you and others comment on this issue in my thread, Discuss Subaru EJ Motor Builds For Road Racing (http://thefactoryfiveforum.com/showthread.php?14327-Discuss-Subaru-EJ-Motor-Builds-For-Road-Racing/page2), where I think it is more appropriate?

[metalmaker12]

For an EJ207 with that size turbo I generally put an 11lb spring in the wastegate and run on spring pressure bellow 50% throttle. Below 11psi, even on pump gas, you can still run pretty aggressive timings below 3k. Allows the turbo to stay nice and spooled up down low with the a good failsafe against ptfb related engine failures on the road.

I then run 75% wgdc at 60% throttle, 100% wgdc above 70% throttle with the MBC setting max boost. I set my boost tables 25% higher than whatever the MBC is set for. And a one way check valve before and after the MBC. So if either the MBC of EBC fails it failsafes to the other. Then on the load table I always run a row above whatever the max load is, at full target boost, and run it rich as hell.

So if it ever overboosts there is plenty of cooling fuel to keep the detonation down.

I'm also running 24 PSI though. So if I have full boost at 4k and low load, is a way bigger deal than you. The ecu can probably get close to pulling enough timing for 5 psi of overboost at low load. Can't come close at 13 psi of overboost. Tuning for 16psi at low load will create a weird revers hump in your timing map though. Haven't tried it, but I wouldn't necessarily suggest it. Could do weird things in shifts and when lifting off throttle going up or down hill. But again, never tried it so it could work great.

I also tune everything to 85% though. If I can't get the power I want out of 85% of current setup, I change the setup. I don't mind blowing stuff up at a track, but I prefer not to. And I definitely don't like blowing things up driving around town.

I have found running like this makes the car feel way more like a turbo charged 6 than 4. The power build linearly with how much throttle you apply. As opposed to mashing the throttle and waiting for the turbo to spool. Unless you mash it off the line. But with a twin scroll turbo and aggressive AVCS and cams, you are only waiting 3k RPM for spool.

Looking forward to seeing you up and running again. Particularly interested in how you get the power to the ground. As a guideline to how far I am going to need to go with suspension and body mods for humongous rear tires.

This is pretty much how were going to do it except much less boost, I only want 275-300whp, but am going to try to get the setup to have minimal lag. I also run it rich as hell lol. I am interested in how it will shift off and on boost etc, but I feel we will get it where I want it with some road and dyno tuning. Typically I run 21-26 psi with this setup and make 400-420whp, but the 818 does not need it and I want to try to keep it together.

[ssssly]

Scargo. Referring to a twinscroll Dom 1.5x.

Tuned to around 16 psI it should have no lag if you have at least a 10lb spring in the wastegate. Would think you will have issues keeping no lag and below 350ft lb of torque under 5k. It's gonna want to make more like 360. You're probably gonna have to pull a ton of timing. Or you will have to use more like a 7 lb spring to bleed of boost earlier, which will make it feel laggier down low. Still plenty of juice for this car though.

[iblackwe]

Hi all,

Bringing a thread back up again. I just had a very expensive dyno experience as my car overheated on the dyno and the techs couldn't remove the air lock. I was told that a coolant boost pump would likely be necessary however, after reading all of this I believe the stock mechanical pump is fine and that the system just needs to be burped very correctly. I have done waynes mod prior to the dyno and filled the system myself prior to them trying to purge it as well.

I am looking to see if anyone has come up with a definitive answer as to whether the 818 needs a boost pump or not. As well, does anyone else have a trick to bleed the system of air entirely. I think what I found is that I had a air lock in the cool return line of the radiator going over the steering rack and down alongside of the car.

Please let me know what you guys think, I honestly can't imagine another botched dyno run haha.

[Bob_n_Cincy]

Hi all,

Bringing a thread back up again. I just had a very expensive dyno experience as my car overheated on the dyno and the techs couldn't remove the air lock. I was told that a coolant boost pump would likely be necessary however, after reading all of this I believe the stock mechanical pump is fine and that the system just needs to be burped very correctly. I have done waynes mod prior to the dyno and filled the system myself prior to them trying to purge it as well.

I am looking to see if anyone has come up with a definitive answer as to whether the 818 needs a boost pump or not. As well, does anyone else have a trick to bleed the system of air entirely. I think what I found is that I had a air lock in the cool return line of the radiator going over the steering rack and down alongside of the car.

Please let me know what you guys think, I honestly can't imagine another botched dyno run haha.

Hi Iblackwe,
I have put many hours into the 818 cooling system.
The setup I have now has been drain and refilled at least a 1/2 dozen times with no airlock or other issues.
No special bleeding procedures. Fill system, heat cycle it once, refill and your done.

The number one rule: every high spot in the system must automatically bleed back to the expansion tank.
I have 2 paths.
First I did modification very equivalent to Wayne's Mod. I put an 1/8" pill in that hose to let air flow freely and it would restrict water flow.

Secondly, I add a bleed hose from the top of the radiator back to the expansion tank. (my radiator has a sealed cap.) I raised my expansion tank up into the hump to make it more effective.

The number 2 rule.
Run every hose (bleed or main) with no rises that will trap air.

The number 3 rule.
Run smooth line (tubing or rubber). The pressure out of the water pump has multiple paths it can take. You need to make it run easily to and from the radiator.


Picture of bleed pressure testing:

58245

Picture of engine bleed hose.
58246

With all that said, A boost pump would help.
I don't know what would happen to a boost pump when the thermostat was closed and nothing is flowing.

Bob

[JeffS]

I accomplished the same thing as Wayne's engine mod using one of these drain fittings....

58273

58274


For the steering rack bump I welded a bleeder bung at the highest point on the bend...

58275

I can't tell you yet how well it's going to work though... a few more months to go-carting.

Jeff

[UnhipPopano]

Two things. First, the proper term to use "The number one rule: every high spot in the system must automatically bleed back to the expansion tank." is Vent. All high-point vents can be connected together in the front of the car and then routed to the expansion tank or some other high point vent tank. Note that vent lines can be routed higher than the point that they are venting. Air will find its way to these high points and then be drawn to the vent tank. The second thing is to ask if anyone has added any testing for flow rate of the coolant through the engine. The purpose of the closed impeller cooling pumps are to ensure enough flow by generating a higher head pressure.

If the flow rate is not high enough, then the surface temperatures of the cylinders will be higher. This would possibly cause failure down the road.

[flynntuna]

I was under the impression that flow rate is critical to engine cooling. To fast and the heat doesn't have time to transfer to the radiator fins, to slow and there's to much heat at the heads.

[Wayne Presley]

There is no such thing as too fast. The fast flow prevents localized boiling. The faster flow also cause turbulence in the radiator tubes breaking up the boundry layer which helps in transferring heat to the fins/air.

[STiPWRD]

I was under the impression that flow rate is critical to engine cooling. To fast and the heat doesn't have time to transfer to the radiator fins, to slow and there's to much heat at the heads.
There's no such thing as the flow rate being too fast that heat transfer won't occur. However, if it's too slow, yes the heads will be hotter.

Imagine a table fan blowing on your face, it's not actually removing any heat from the air (like an AC system would), it's simply creating some air flow, which improves convection - so you feel cooler. If the fan velocity increases, you're not going to suddenly feel hot because heat transfer stops. The convection will asymptote to some limit for forced air. In fact, as the flow increases, it will transfer from laminar to turbulent and heat transfer improves drastically with turbulent flow.

Edit: Wayne beat me to it

[Wayne Presley]

At least we both paid attention in thermodynamics class

[iblackwe]

Hi Iblackwe,
I have put many hours into the 818 cooling system.
The setup I have now has been drain and refilled at least a 1/2 dozen times with no airlock or other issues.
No special bleeding procedures. Fill system, heat cycle it once, refill and your done.

The number one rule: every high spot in the system must automatically bleed back to the expansion tank.
I have 2 paths.
First I did modification very equivalent to Wayne's Mod. I put an 1/8" pill in that hose to let air flow freely and it would restrict water flow.

Secondly, I add a bleed hose from the top of the radiator back to the expansion tank. (my radiator has a sealed cap.) I raised my expansion tank up into the hump to make it more effective.

The number 2 rule.
Run every hose (bleed or main) with no rises that will trap air.

The number 3 rule.
Run smooth line (tubing or rubber). The pressure out of the water pump has multiple paths it can take. You need to make it run easily to and from the radiator.


Picture of bleed pressure testing:

58245

Picture of engine bleed hose.
58246

With all that said, A boost pump would help.
I don't know what would happen to a boost pump when the thermostat was closed and nothing is flowing.

Bob

Hi Bob thanks for the advice, I will definitely try and remove the high point going over the steering rack. I like jeffs idea in putting a bleed point right overtop of that section and I think I can utilize that drain fitting to do so as I am still using the corrugated tubing and can cut and splice that into the line.

Would i potentially be able to plumb that bleeder vent back into the vent line coming from the top of the front radiator that runs all the way to the back of the car into the expansion tank on top of the manifold? As of right now I have waynes coolant mod coming from the cross over pipe tee'd into the vent line from the front of the radiator and then they both plum into the vent line of the tank on the manifold... I am sure plumbing in a third line from the front of the car wouldn't be an issue. I could also put 1/8" restrictors into each line as I think that is a great idea.

I currently have the expansion tank barb on the front radiator sealed and so I am assuming I should be using a sealed (no vent) radiator cap as well. What pressure rating should I be using for the rear coolant expansion tank (on the manifold) as well?

Thanks everyone.

[flynntuna]

Thanks for the explanation. It inspired me to do further research.

http://iosrjournals.org/iosr-jmce/papers/sicete(mech)-volume2/13.pdf

http://www.maplesoft.com/applications/view.aspx?SID=6403&view=html

[Wayne Presley]

I run a 1.3 bar front cap and a 1.1 bar expansion tank on the motor

[Bob_n_Cincy]

I currently have the expansion tank barb on the front radiator sealed and so I am assuming I should be using a sealed (no vent) radiator cap as well. What pressure rating should I be using for the rear coolant expansion tank (on the manifold) as well?

Thanks everyone.

I think a large air pocket in the radiator is most likely your biggest problem.
Use Wayne's suggestions on the radiator caps.
I would run a 5/16 hose from the radiator bleed port to the expansion tank. I did not put a pill in that line from the radiator. That is the most important place that needs bleeding.

[DodgyTim]

I used one of these temperature sensor fittings in the high point of the coolant line at the steering rack.
Its an easy way of getting a bleed port, and only cost about $5
58297

[UnhipPopano]

The bad part of this Test Report "Thanks for the explanation. It inspired me to do further research. - http://iosrjournals.org/iosr-jmce/pa...volume2/13.pdf" is that it was only testing the Radiator and did not take into account the combined effect on both the Radiator and Engine. As the flow goes up, more energy is removed by the coolant at the engine, and then released by the Radiator. At the same time, energy is being used by the pump to pump the coolant and this reduces the available engine power.

As far as "tank barb on the front radiator sealed ....I think a large air pocket in the radiator is most likely your biggest problem." this is self evident. If the Radiator is in effect not full, then it can not remove as much heat. A Radiator that the coolant is only covering half the cooling fin area will only remove 50% of the heat, assuming no increase in inlet temperature.

[iblackwe]

I think I may have mis led everyone on the front radiator. I have a mishimoto front radiator and the passengers side of it has a barb fitting which I believe is the vent line. So I have that line running to the back of the car to the small coolant tank on the top of the manifold. The barb coming from the radiator cap has been removed and plugged with a set screw as I didn't leave the overflow coolant tank up front, I mounted it in the back and plugged it into the small manifold coolant tank. If I need to put that barb back in under the radiator cap I will definitely do so but my initial impression was that it wasn't necessary.

Thanks

[Wayne Presley]

Front capped is fine

[flynntuna]

Found this on line...http://www.mgf.ultimatemg.com/group2/coolant/ewp.htm
Is this a reasonable approach?

[Wayne Presley]

Found this on line...http://www.mgf.ultimatemg.com/group2/coolant/ewp.htm
Is this a reasonable approach?

I think you are addressing a problem that isn't there in the first place. I've had 2 818's on the track and 8 on my dyno without any cooling issues

[Bob_n_Cincy]

I think I may have mis led everyone on the front radiator. I have a mishimoto front radiator and the passengers side of it has a barb fitting which I believe is the vent line. So I have that line running to the back of the car to the small coolant tank on the top of the manifold. The barb coming from the radiator cap has been removed and plugged with a set screw as I didn't leave the overflow coolant tank up front, I mounted it in the back and plugged it into the small manifold coolant tank. If I need to put that barb back in under the radiator cap I will definitely do so but my initial impression was that it wasn't necessary. Thanks

Sounds like your set up on the front radiator is good.
When the engine is running (water pump spinning). You should get a flow of coolant in each of the bleed(vent)lines to the expansion (degas) tank on top of the intake manifold. Heat cycling your engine should push any air and excess coolant out of the expansion (degas) tank into the overflow bottle. Each time the engine cools, the expansion (degas) tank will suck coolant only from the overflow bottle.
Bob

[JeromeS13]

https://www.amazon.com/Lisle-24680-Spill-Free-Funnel/dp/B00A6AS6LY

Everyone that has a turbo Subaru should have one of these...

[Mechie3]

Bob, how did you run the vent line from the radiator to the rear degas tank without it dipping and causing low spots?

[Bob_n_Cincy]

Bob, how did you run the vent line from the radiator to the rear degas tank without it dipping and causing low spots?

Hi Craig,
It runs parallel to my right side radiator hose.
When the thermostat is closed.
With a 1/8" pill in Wayne's hose.
The water pump develops enough pressure to push water toward the radiator in the 1.25 line and return air and coolant back through the 5/16 bleed hose attached to the expansion tank. I "T" wayne's hose and the radiator bleed very close to the expansion tank.

Here is the post where I was doing pressure/flow testing.
http://thefactoryfiveforum.com/showthread.php?12534-MRG-MotorSports-818S-Build&p=228056&viewfull=1#post228056

Bob

[Bob_n_Cincy]

To avoid humps on the coolant lines, I didn't go over the steering rack, I went under the front LCA.
58429
I moved the fittings in the radiator from the back to the sides to make attaching easier.
Bob