How do you size an intercooler?

Uncharted territory for me.

- I've seen some general things like motoiq typically prefers thinner cores to maintain adequate airflow to whatever is next in line in the cooling stack (in my case oil cooler or radiator depending where in the vertical direction we are discussing)

- vertical flow might be better? more short paths for cooling air vs fewer long paths?

- Shorter piping is better (think filling a big truck tire with air vs a miata tire with the same psi)

And that's pretty much where my knowledge stops. I've also noticed an interesting phenomenon where various companies have wildly different ratings:

Vibrant IC Core 24x8x3.5 625hp rating

Treadstone Core 25x6x3.5 1300hp rating

Despite being smaller the treadstone has more than double the rating, is their design actually twice as good? 

Basically what I'm working on is this, my current intercooler core is 18x8.5x3" it has some massive end tanks which fill up approximately 1/3 of my grille. My idea is that I can upgrade to a bigger intercooler and get better cooling while also increasing airflow to my radiator/oil cooler.

The thing I am hoping to fix is creeping oil temps and to a lesser extent coolant temps.

My grille opening is approximately 28" across the top, 7" in height and 22" across the bottom.

Contact each manufacturer and ask to see a plot of pressure drop vs. airflow volume for the intercoolers in question.  That will tell the story.  Anybody can arbitrarily say their product supports X horsepower, but it really doesn't tell you much about the characteristics of the intercooler.  Pretty sure Corky Bell's book "Maximum Boost" covers this topic, but I can't find my copy.  Now that I think of it, I think I loaned it out!

First, what can you make fit.

Second, what can you make fit with as many tubes as possible?  Short wide intercooler (long tanks) is way better than long narrow (small tanks) since most of the cooling is done in the first six or eight inches of tube, longer tubes than that mostly just contribute to pressure loss.  In your situation this means you'd need top and bottom tanks with a vertical flow intercooler. 

Third, what can you make fit.

Wish I could post the datalogs from a certain twin turbo 4th-gen F-body that I dealt with.  It had tiny little intercoolers mounted horizontally in front of the front wheels, and no airdams behind them to force airflow.  Charge air temp would get up to 140F within about three or four seconds of hitting max boost (about 9psi).  They'd cool right back off again, but still it was disconcerting.  I've been spoiled by Grand Nationals with gigantic RJC intercoolers that would keep charge air temps within 10-15 degrees of ambient, even with running boost in the 25-30psi range.  Granted, at 25-30psi you are never in boost for more than a few seconds before the pass is over with.

Oddly enough, the stock GN intercooler is almost as good as long as you keep the ducting and the crank pulley mounted fan to force airflow across it.  The intercooler core is a lot of tubes, not a small number of long tubes.

What issues are you having with your current I/C set up? Any overheating issues? I'm guessing the car is used for track days? There are so many variables and compromises, you pretty much have to stack up the variables that you can control in a manner that best suits your application. An I/C that makes the most power on a dyno or drag strip may not be the best on a road course. End tanks on the long sides are better in theory, but are highly dependent on the internal design of the tanks to divide the airflow more evenly through the core. And the two extra 90 degree turns they introduce can reduce some of the benefit. 

If I were you, I'd take some measurements of you current system, and see where you stand. Then see if you can make any improvements with ducting first. 

Knurled, I've posted the image in my original post down at the bottom of this reply. But with the dimensions I'm working with. See also the attached behind the bumper shot, I've got quite a lot of room to work with and am fairly handy with cutting tools so even more space could be made if desired. I'm currently seeing 30-45 degrees above ambient pending how hot it is outside, cooler track days I'm closer to 30. Hotter days like ATL in August I'm 45 or a bit more.

Jason K, 

I'm not currently having issues with the intercooler setup, I'm having issues with the radiator/oil cooler behind the intercooler getting too hot. I measure IAT's and typically hover 30-45 degrees above ambient when I come off the track. Car is used for time trials and lapping days (and hopefully also One Lap of America next year.

I'm not sure which measurements you are after or thinking would help so I re-hosted the original image with dimensions on it. My thought was the current end tanks are blocking a fair bit of the grille opening and preventing some flow to the radiator/oil coolers. A new intercooler setup would hopefully lower IAT's closer to ambient and allow more airflow to the other coolers, win-win?

The car currently has pretty good ducting, I have modified the factory M3 ducting with weather stripping material on both sides and around the intercooler. All surfaces are fully sealed, any air coming through the kidney grilles or the lower grille must flow through the coolers before exiting the engine bay. The only spot I could see where I can practically gain better ducting is by making a chimney type vent out the hood (which would require a different intercooler/radiator setup which is the path I'm heading down)

Here's a picture without the intercooler so you can get an idea of how things are ducted, the kidney grilles are also ducted straight to the radiator.

Since that picture was taken I've also removed the AC Fan and condensor which opened up quite a bit of extra surface area to the radiator.

You both mentioned that the vertical core would be the way to go which is great because that's what I was hoping to do from both a weight savings and shorter path standpoint. Jason do you think in this case the disadvantage of the 90 degree turns would out-weigh the previous pipe layout?

Current pipe layout has the cooled air exit on the passenger side and route back under the engine over to the driver side before coming up into the intake:

The proposed design with a vertical intercooler would look more like:

and dimensions of the current setup:

The grille opening is 7" the current intercooler is 8.5" tall and tipped at an angle to allow most of it to fit in the opening.

So based on all of that, I was leaning towards the Treadstone CV25 intercooler core because it's  25" long, 6" tall, vertical flow and you can buy these pre-made end tanks which seem convenient:

this brings up a question I have always had. Why do people put the intercooler down low, usually under the bumper? I know from playing with a thermoprobe that air a few inches off of the asphalt is a LOT hotter than the air a foot or so higher.

FWIW, the "short but wide" argument was used for a lot of early Miata intercooler designs.  The theory sounded good, but it turned out that in practice it didn't make much difference at all.  The packaging of top/bottom tanks winds up being a PITA in most cars, so everyone uses long and narrow ones now.

FWIW in my case (not an off the shelf bolt on application)

I can make a short but wide design work without much issue. I have dead space behind the crash beam where no airflow is anyways to house the upper piping. While a street m3 would wind up with the lower pipe hanging below the bumper by a little bit I have an extra 2" of lip and splitter below to keep the piping "inside" the body of the car.

There's a ton of factors in intercooler choices but based on what you're actually trying to accomplish (which has nothing to do with already-adequate charge temps) im taking the KISS approach and saying stick with your original idea. Get a wider core that replaces the flow-blocking endtanks in your grille opening with more intercooler core. 

Vigo said:

There's a ton of factors in intercooler choices but based on what you're actually trying to accomplish (which has nothing to do with already-adequate charge temps) im taking the KISS approach and saying stick with your original idea. Get a wider core that replaces the flow-blocking endtanks in your grille opening with more intercooler core. 

Cool, so I'm on the right path then, now the question is, vertical flow or horizontal flow?

I should go turbo that way I wind up with boost entering on passenger side and exiting on driver side for efficiency sake 

codrus said:

FWIW, the "short but wide" argument was used for a lot of early Miata intercooler designs.  The theory sounded good, but it turned out that in practice it didn't make much difference at all.  The packaging of top/bottom tanks winds up being a PITA in most cars, so everyone uses long and narrow ones now.

Some of that depends on the application.  In higher power cars where you need to move a lot of mass flow through the intercooler, it makes a significant difference.

MY opinion, is to use water/air.  The water radiator up front doesn't have its size constrained by needing to attach large charge air tubes with large end tanks.  The intercooler unit itself can be placed anywhere convenient for ducting.

Cool, so I'm on the right path then, now the question is, vertical flow or horizontal flow?

I should go turbo that way I wind up with boost entering on passenger side and exiting on driver side for efficiency sake

I would be surprised if you got a larger intercooler core that ended up measurably worse in function, so i would pick the style based purely on packaging and ease of charge plumbing.  Again, im simplifying here based on the fact that your setup is already functional and reliable and you're really only concerned about improving airflow to the other coolers.  In general longer tubes will shed more heat but be more restrictive (shorter tubes the opposite). But, more restriction also means the turbo ends up spinning faster and making more pressure to have the same manifold pressure, so some of that additional heat loss through the intercooler could be offset by additional heat going in due to the compressor having to work harder. I dont think it's net-zero but it's diminishing returns. I think going too far down the rabbit hole is basically a waste of time unless you're trying to build in headroom for some future massive power increase. One way or another you're going to end up with a core with more surface area which probably means its going to work at least as well if not better than the current one. You're going to wind up with less blockage in your airdam flow path which probably means your other coolers will work better. I just think it's going to work. 

I agree with Vigo. Concentrate on solving the engine cooling issues by getting the I/C tanks out of the way. You are unlikely to hurt the I/C functionality with a larger core, as long as it isn't overly restrictive and the piping is as efficient as possible. That said, I think that end tank leaves something to be desired. One of the main reasons why the top-bottom I/C's don't deliver the theoretical benefits is because the end tanks don't do a good job of evenly distributing the air charge through the wide core. It could benefit from a divider to direct more of the air charge to the far side. But in the grand scheme, if may work well enough, especially if you are well within the limits of what that core will support as far as airflow and charge temp.