Intercooler vs. No Intercooler
 

This conversation began under a different thread, but I thought I may get better responses by starting a new thread.

Background: I'm running .7 BAR with 8.0:1 compression and twin plugs. I'm comfortable driving this on the street only, but I'd like to maximize the reliability. I have been in contact with Pat Keefe regarding water/methanol injection and will continue looking into it. I have a stock Carrera whale tail with the long A/C condenser, and I would like to stay away from a turbo tail if I can. I have a design for an intercooler that would fit under the lid: 20.8x5x2.25 with a flow rating around 900 CFM according to Bell Intercoolers. Gerhardt at Bell Intercoolers was concerned, however, that with the minimal ambient air flow & engine heat soak I should ideally have a larger core size to compensate.

I know the better solution would be to relocate the condenser or get an entirely different tail. But barring those options, do you guys think that installing an IC as described above would be better than no IC? Even if I could get a 20% efficiency, I'd like to throw something in there. Is it possible that an intercooler, in this situation, could actually have a negative effect on the intake charge temp?

Thanks in advance for your advice, as usual.

IMHO, anything is better than nothing. When the car is moving, heat soak is less of an issue, so even a small intercooler will help to cool the air by a few degrees & add to reliability.

Re:size, I would have thought you can get away with a slightly bigger/thicker core even with the carrera tail? How is this core oriented?

BTW - I worked with Gerhard to design my full bay intercooler a few years back (in fact, mine appears in the book Turbocharging Performance Handbook :D) and he's great to work with & very knowledgeable.

A 3" thick core would fit, but the overall core size would have to be pretty small. There are things I could do to free up some additional space such as relocate the A/C condenser and chuck the AFM and rear wiper assembly. Those changes would allow for a 27x6x3 flowing around 1500 CFM. That's a possibility down the road, but for now I'd like to keep those things in place.

The intercooler I have in mind will be placed directly behind the fan housing and proceed back over the intake manifold. Kind of like this setup except the intercooler will be about 3-4 inches closer to the fan.

http://forums.pelicanparts.com/uploa...1218811598.jpg

I remember Gerhard mentioned that while shorter tubes are better, there's also the "too short is bad" factor. He mentioned something like 10" minimum. So with the core shown in the picture, I'd have thought going side-to-side would be more effective even though you'll have more pressure drop (which he says, crank up the boost/bigger turbo :) )

I think this is starting to make sense. So the length of the end tanks largely determine the flow (CFM) of the IC while the length of the tubes/fins along with the core thickness determine the "cooling capability"? If I'm understand this correctly, the above intercooler sacrifices cooling (length of tubes/fins) for minimal pressure drop (end tank size)? By contrast, the below intercooler (excuse my art work) would have worse flow (350 CFM according to Bell IC's) but would be more effective at cooling the charge:

http://forums.pelicanparts.com/uploa...1218826617.jpg

Gerhardt calculated that I would need around 700 CFM to support 400 rwhp. He estimated a 2-3psi loss in pressure with an IC that flowed that poorly relative to the HP output. Would there be any negative effects (aside from turbo wear & tear) to turning up the boost to compensate for this pressure drop? I suppose alternatively I could go with the better flowing IC and just add water/methanol injection to drop the charge temp further.

What are your thoughts? Am I thinking along the correct lines?

I think the most noticeable negative is it takes longer to get to the same boost, i.e. throttle response will be worst too.

If your turbo is big enough to flow, say, 16psi of air, losing 2-3psi simply means it takes longer to get there. I wouldn't even say more wear & tear.

I'm not familiar with water/methanol injection - what does it do & how does it work? Just curious :)

Understood. Does anyone know have an educated guess on what the post-IC temps would be on these two intercoolers? If there is not a large difference, I would much rather not sacrifice the throttle response. With twin plugs, 93 octane gas, and potentially water injection sometime down the road, I think I should be fine with the higher flowing, less cooling IC - especially since my boost level is not that high.

Any thoughts?

Water/Methanol injection is the controlled squirting of water/methanol into the intake charge. The system involves a reservoir, pump, hoses, and pressure switch. The water/methanol absorbs the heat and effectively lowers the intake temperature. Pat Keefe has had a water injection system installed on his for Turbo SC for a while now without any issues. It's common to leverage the windshield washer fluid reservoir to feed the injection system. A boost activated pressure switch is commonly used to trigger the injector (e.g. you can buy a switch which kicks in at 3 PSI). It's a relatively simply system and quite effective/inexpensive.

If not having an intercooler vs putting an intercooler in an unusual spot is your goal then you should design your engine to that limitation.
Start by finding the most efficient ball bearing turbo for the pressure island you will have. Optimising the turbo will significantly reduce discharge temperature. Add twin plugs, good gas, exact cam and ignition timing, and you may not need water/MeOH injection.
Give Marty (copbait73) a holler. He is an expert at this application.

Interested in seeing what you come up with here Andy.

The first intercooler that I built for my car fit under a Carrera style tail. It was a side to side design that was about 6" wide not 8" like above. The pressure loss was 4.5 psi. That was too much for me. The motor has to work that much harder if you want to keep the same boost level as no intercooler. So I bought a big base Carrera style tail and built a bigger front to back intercooler. Now I have less than .5 psi pressure drop.

Dean

Thank you for your response, Rarly. I have been wondering if all of those factors you listed could make it possible that no IC is needed. I think that I'm on the right track in terms of your list. I'm going to be running a ball bearing Garrett gt35r modified by Precision to flow even better. I'll call Precision to find out what kind of heat I could expect at 10PSI. I've already got the twin plug ignition in place, and I'll be running a Protomotive chip optimized for twin plugs. I'm entirely ignorant on how to adjust cam timing. I know that my wrench buddy set the cam timing to the exact specs of an '87 Carrera, as indicated by the Bentley manual. Finally, I have 93 octane gas readily available.

Unless the overwhelming majority comes back and says "something is better than nothing" in regard to having or not having an IC, I will start out with NO IC and monitor the intake temps. I'll relay those temps here. Regardless, I'm still curious to hear what kind of efficiency you guys think I could get out of a small IC with low ambient air flow.

One other thought is that Protomotive has had a kit out there for a while for running .5 BAR on a Carrera with 9.5:1 compression and NO intercooler. If you check out this graph which Marty posted a few months back, the net compression on that Protomotive setup would be 12.0:1. I'll be running .7 BAR on 8.0:1 compression, resulting in a net compression of 11.5:1. In addition to that, I'll be running twin plugs with a more efficient turbo.

http://forums.pelicanparts.com/uploa...1218890694.jpg

I never knew this was the case, so many cars i see run the 'side to side' style, I had no idea you could loose that much boost, 4.5 is a huge drop!

I was very surprised as well, Spence. Check out this link:

http://www.bellintercoolers.com/pages/aachart225.html

Column "B" is the length of the end tanks and Column "C" is the length of the cooling tubes/fins. End tank length along with core size largely determine the air flow (CFM) In those terms, there is a large difference between an 18x6x2.25 and a 6x18x2.25. The first yields close to 800 CFM, while the latter yields around 250 CFM (about 3 PSI drop for our cars!). The larger the gap between the CFM rating of your IC and turbo the greater PSI drop will be, provided the IC has the smaller CFM. Intercoolers can be too large as well. Either way, if your IC flow rating is not within proximity of your turbo's output, you'll experience more turbo lag.

Thanks a lot Andy, did you ever see the way supercharging of knoxville did their little intercoolers? Maybe supercharging doesn't increase the charge temp like hot turbos do. If you get rid of the AFM alone there should be enough room right?

Supercharging can generate the same heat increases as turbocharging. The different types of superchargers make a big difference in heat added at particular boost levels and rpms (e.g. Eaton (roots style) vs whipples).

I have lived with air-air intercooling, water-air intercooling and water/meth injection. In my experience, air-air is the most effective for the least amount of aggravation.

My second favorite is water/meth injection (up to 1 bar). You can get a very effective system for very little investment, use cheap windshield washer fluid to get pre-mixed meth, and get the added bonus of a squeaky clean combustion chamber. The downside is that you have to plumb the water source and remember to top it off regularly.

My experiences with going without either intercooling or water/meth injection have been less than satisfying. I always felt like I was underusing the potential of the car or was running too close to the edge of detonation on hot days.

You could always start out with water/meth injection and switch over to something more elaborate later. For example: http://www.alcohol-injection.com/water-injection-kits-1/universal-19/universal-stage-2-12.html

Spence, it frees up some but not a lot. To use the space freed up by the AFM you'd still have to deal with the rubber elbow on the throttle body. With the AFM & rubber elbow out of the way, you could get an intercooler custom designed to discharge directly into the throttle body. I would imagine that would allow for an additional 3" of IC depth - a lot of fab work for little return IMO.

Thanks for your post, Mike. Just curious, have any of your cars run with no IC been twin plugged? Did you feel a need for further cooling? The more I read about methanol injection, the more I'm liking it. I'd be curious to hear about how difficult the install is.

Hey Andy, when you say 'rubber elbow' I asume you're referring to the Porsche OEM plastic elbow part? That's easily gotten deleted, I don't run it as mine kept popping off under boost, you just need to get a hose fitting somewhere in the intake path for the ICV and the brake booster, presently mine are tapped into the AFM but you could tap them anywhere.

you don't think it would be enough? You're crushing my next project plans! You don't think the above posted intercooler would be sufficient?

Ok, I'll chime in a bit. My water injection works very well, just as good if not better than air-air. If it's hot out, you can't get any approach on the condenser lower than ambient. My instrumentation suggests that it is hotter at the air inlet and past the throttle body than one would suspect. I can maintain under 120F temps at the track in the airbox, which is better than an air-to -air unit will get. I have shut the water injection off at the Glen as a test and saw 200F air in the airbox. The downside is a bit more maintenance (filling the water resorvoir counts as maintenance), and keeping the nozzle clean.

I did once speak to Gearhard at Bell, very sharp guy. I aslo spoke to the guys in Phoenix (Turbokraft, maybe), also very sharp guys. If I had to go air-to air, I would get the Phoenix guys to build me one. The biggest problem is the severe lack of room if you want to keep a no spoiler/wing/tray tail, although the 3.2 setup seems to bee a bit more roomy.

Pat

One nozzle, and a flow switch, all the rest is tucked away:

http://forums.pelicanparts.com/uploa...1218930402.jpg

It was a small intercooler. Maybe 6" x 3" for the boosted air.

By my tests blocking the IC and pre/post temp sensors I have installed:
Surprising how little heat is created from short, 5-10 second bursts of boost.

Sticking neck waaay out: No intercooler would be ok in many cases for many people, street use only and just fun mini-blasts of acceleration.