I had bought a OEM intercooler off of EBay and had Autobahn Garage redo the end cap and we talked about that on another post. We are working on getting my car and a few other's on the Dyno and will do some runs with both IC so we can saw yes it works or no it does not. Today I cut off the end cap on the ban saw of my IC and really took a good look at how much air could be getting to the front side of the IC. I can tell you that ther can not be a whole lot. Because of the pie shaped wedge of the feed end cap ther is not very much room for the air to get to the front 1-1 1/2 of the IC. Once air enter's the IC it stays in it own channel till it exits the out going end. I would guess that from the side of the core to the wedge shaped end cap there is going aft 1 inch the opening is less then 1/4 of an inch wide. When you add the thickness of the end plate in, ther just can not be a whole lot of air getting to the first 1 1/2 inches of the leading edge of the core meaning that the whole 1 1/2 inches of the leading edge of the core is not being used or not much air can be getting through. We are talking boost or air under pressure and the air will flow through but it will also go the least resitant path meaning the back 2/3 or more of the IC. BY redoing the end cap and getting air to the front of the IC we will be getting more cold air to the engine and should'nt this give us more power??

Before some of you start talking about flow number's and dyno time you need to have cut this end cap off to see what I am seeing. Unless you have actually cut the end cap off, your just not seeing the whole picture. I left the IC at the shop and will get some pictures of it tomorrow.

Wolf Pack 88 951

 

So you cut the end off...where are the cool pictures??
Show us

 

I brought home the end cap but left the IC at the shop.

Cliff

 

My money is still on the math and theory behind the design of the original intercooler. That "restriction" you refer to will cause an increase in flow velocity and a decrease in ambient pressure, which will subsequently drop the ambient temp, assuming its not so small that the air flow is restricted to the point of starvation, which would be pretty unlikely in this scenario, given the overall volume that is in question. The other, larger side allows it to draw air. Mathematically, this should yield a higher flow of air with a higher decrease in temp than a design that has equal volumetric flow on both sides. Think about it this way: the engine obviously isn't starving itself of air or your a/f ratio would never meter out, so it must be flowing enough, therefore we can assume that the capacity of the intercooler isn't the issue. By reducing the size of the pipe entering the intercooler, you immediately reduce pressure and temp (as per Bernoulli) before the air has been drawn through the intercooler. The engine will pull the air it needs through the intercooler, which will get a further benefit of cooling as it passes through. If you take away the "restriction", you take away the pressure drop and will subsequently force any temp drop to come solely from the cooling effect of the radiating fins. Because this occurs at a higher temp, pressure goes up and you wind up reducing flow. And this doesn't address another concern I have with the whole "swap the end cap" idea, which is creating inconsistencies in the surface temp of the unit. I don't have sufficient math data to be able to model this, but I do know that consistent surface area temps will aid flow characteristics. To what extent in this application would be a question worth asking.

The unit that Tim from SFR used to produce his dyno results is a good way to go. He's got a larger surface area to work with and moves the intercooler into a high pressure area, so its got a better cooling effect. A very nice alternative, and I like his suggestion of putting a larger oil cooler where the old intercooler went as well. Theoretically, because he has a larger intercooler to work with, the flow characteristics through the intercooler aren't as important so the impact of the size of the end tanks in minimized.

These are my musings based on my knowledge of flow mechanics. I have NOT done any real world testing on this, nor is it intended to dissuade you from pursuing your project, but maybe you will find a useful tidbit of info in all this. I'll be interested in your dyno results.

Regards,

 

Scott,

I respect your knoledge on this, but I am some what confused on the end cap issue. After looking at the IC from Sperco and looking at other car set ups from Racing Car's to Ricer car's all with IC. I talked to Sperco on the phone today and they said I would want the end I have yet to see a IC that has a end cap like on the 951. Cutting the end cap off today it was easy to see that ther can not be a whole lot of air reaching the front 1/3 of the IC. The opening in which the air must go through is just not very wide. I do understand that the air will be pressurized but would'nt the air go the easier route of the back side of the IC? Having been a sailor for along time and having Yacht Design as a somewhat hobby I see that if this were water that some of the water would go to the front section of the IC but that most of it would pass through the back side of the IC and it would all mix at the exit end. I would think that the flow of air/water would go faster through the least risistant section and with air would build pressure more on the back side of the IC then the front side.

I have looked at Tim's IC at SFR and it looks like a very nice piece and I also believe that a bigger core section is also the best way to go as long as it has the right amount of flow. Doe'nt his IC go in the same spot as the stock unit? If Porsche did this for a specific reason then why has nobody else done this? I quess I need you to explain this in a simpler form for me and other's to understand this. Thanks for your input on this.

Cliff

 

blah, blah, blah. Fluid dynamics doesn't have a lot to do with what "looks" like a restriction.

 

Cliff, it really would be cool if you could post some pics. I think most of us would like to see what the cap and I/C core looks like once they been separated.

Thanks

 

Cliff-
You have a few threads about your intercooler going at the same time. Along with the thousands of other threads and information you've already been told, how much more do you want people to say?

In all reality, I don't think anyone has done any specific testing on the IC itself. Dyno results really won't tell you much. There are too many variables, typically the airflow across the IC is not proper for real world use, and many people don't have a way to test in and out temperatures or pressures.

Ideally a flow bench would be used to tell us the exact flows and temps of a stock and modified unit. Short of that, none of us can tell you the exact answer.

I have some logging capability for my car (temp and pressure in/out of IC) that I'm using to determine water injection trigger and flow. I do not have enough data to comment on the performance of the stock IC at this time. Perhaps after a month or 2 of logging in various temperatures and swapping to a unit with an "upgraded" end tank can help to provide some data about this topic. By no means will this information be the holy grail on the subject.

 

Lets try this... have you ever put your finger over the end of a hose when you are washing your car? Notice how the water comes out with more force/speed? Same principle.

Why did Porsche do this when others haven't? I'd speculate that there are two reasons. First, they have the resources to do this sort of thing, whereas your average tuner doesn't even have a flow bench, much less the math modeling capabilities. Its a pretty time consuming job to design or make a change and then get an accurate and meaningful result. Secondly, Porsche had a specific target they were trying to hit, based on flow efficiency and size that the intercooler had to fit in, plus costing and a host of other factors. Tuners primary consideration will be "what sells" (thats not a slam on tuners or suggesting that they are trying to sell junk... some do, but they tend to get weeded out pretty fast), and will tend to develop their products over time. Take a look at how the MAF and chip markets have developed over the years as an example.

Whats the bottomline in all this? You want to find an easier/cheaper way to produce more hp, right?, and your method in this case is to reduce the overall air temp. At $150 per modded intercooler, its not like you are trying to get rich off this. The question winds up being, is it easier to make hp by trying to get the inside of the intercooler to work better, or by getting more cooling by getting flow over the outside of it (either by routing more air over it or relocating the intercooler, applying a cooling agent (such as water mist), or increasing surface area). We both agree that there are hp gains to be had by looking at intercooler design, we just disagree on the approach. I have given, on several occasions, my reasons why I believe that the physics behind the original design is sound. I will also readily admit that I do not have all the answers, so feel free to prove me wrong.

Regards,

 

Porsche has a large amount of engineers at their disposal. All those guys are total geeks and get paid to geek out on making stuff work to its optimum based on the criteria given.

I just think that to beat that brainpower is not such an easy task and slicing off an intercooler endtank and recontouring it "because it looks like a restriction" isn't going to beat the Porsche brain trust. I think their positions at Porsche are safe...

 

Yeah , care to explain what happened to their superior brain power with 996/986/997 Rms leaks, gear/synchro issues on Gt2/996tt's, wiring harness recalls on 993's, leaking like a toilet 964 cases , lousy valves guides on all of them and the list goes on. Porsche is a capable cutting edge manufacturer but they f*ck up like everybody else and it stands to reason that a intercooler designed 20+ years ago is not the zenith of , like you call them, "porsche brain trust".
BTW, the stock intercooler on the 930 is marginal and that is being polite.

 

I'm somewhat with pole on this ( i.e masochistic clutch service, weak hubs, . . . ). The design of every passenger vehicle component is a compromise of effectiveness and cost. Moreover, it has to be the ultimate falicy to say something can not be improved because one doesnt have the experience/degree/facilities that the original designer(s) had.

That said, I also think Tim's (SFR) solution is the best overall direction - the 951 intercooler works better as an oil cooler in both it's size and location.

 

I'm sitting here looking into my stock IC with a flashlight and a dental mirror, and don't see how there's going to be much internal flow through the front side of the IC--this because of the concave taper of the end tank. Internal flow is going to take the path of least resistance, which in a stock IC is NOT gonna be through the front. So, I'm with Wolf on this.

What these ICs need is better internal flow and better ambiant cooling flow via proper exit ducting.

And, regarding the thumb-over-the-garden-hose analogy, think about this: Sure, the water squirts farther, but there is a lot less of it coming out, due to the restriction created by said thumb. If you're using such water for cooling, you're probably not gonna cool as well as without the restriction, because volume of coolant is probably more important than velocity.

And, regarding the Porsche-did-it-so-it-cannot-be-improved-on notion, consider this: If that's true, then NONE of the aftermarket mods ever done on any of our cars has made a bit of difference. Seems to me, quite a few modded cars are running much better than stock. And, while on the Porsche-must-be-perfect thread, please get back to us after your next clutch job, etc...

 

Here's a simple, cheap, and effective way to test IC internal flow: Water and air have virtually identical flow characteristics at relatively low (i.e., less than transonic) speeds.

So, do a before- and after- test with the stock IC configuration and then after modifying the concave end cap.

Attach a big funnel* and pass, say, 20 gallons of water through the funnel and IC. Measure the time it takes the 20 gallons to gravity drain through the IC is each configuration, the elevation and orientation of the IC and funnel being fixed.

I'll bet the modified end cap (assuming no jagged internal edges or sharp turns) drains the water significantly faster.

* A big plastic water bottle like used in office water coolers would work great, just cut the bottom off to make the funnel, then attach the bottle neck to the IC with the stock IC hose, clamp, and maybe duct tape.

 

I like your water flow test suggestion, except that its just testing one dimension (flow) instead of flow AND pressure. Air changes density, and water remains constant. Given the assumption that air density stays constant, I wouldn't argue much, but it doesn't. Also, in my opinion, the point of the intercooler isn't to flow as much air as possible through it, its to make the air running through it cooler (and subsequently more dense). If the amount of air flowing through the intercooler were the major concern, then how is it all the modded 951's running bigger turbos/chips consistently show a proper air/fuel mix on the countless dyno charts that have been posted here? They'd all be going way rich, if they were getting air starvation from the intercooler, so we can assume that the stock intercooler flows more air than is necessary as is, right? Flowing more air won't get you anything by itself, since the engine is already getting what it needs. We need to measure pressure drop and change in temp, not flow as our measure of success.

One other comment, I do not believe that Porsche holds all the cards or magic beans, and I do believe that there are many things that can be improved on our cars. Look at the chip/software development thats taken place as just one example. It wouldn't have happened unless folks got interested in improving these cars and had similar discussion to whats taking place here.

Good discussion points, Dash!

Regards,