I attached a pic, pretty much like you described, I just made a template out of cardboard, transferred it to aluminum, and it just sits around the filter to isolate it from the engine bay heat.

For every degree you drop going into the turbo, you drop more than one degree out of the turbo. I think it works out to about 1.12 degrees. What this means is the cooler you can make your pre-turbo temps, the less work the IC has to do. Another thing to bear in mind is that the ECU is designed to work with the stock airbox, which draws in cool air from the grille. I'm not sure exactly how the EVO ECU is, but on a DSM the ECU will start to pull timing if charge temps get over 85F. I'd imagine the EVOs have similar safeguards built in.

Here's a great calculator that you can plug numbers into and see what effect inlet temps, compressor & IC efficiency, water/meth injection and VE have on charge temps & power,...

http://www.not2fast.com/turbo/glossary/turbo_calc.shtml

 

I monitor my AIT with a GM sensor in my intake manifold. I have a stock IX turbo peaking at ~25psi and tapering down to 21 psi at redline, Nisei aluminum pipes, a Perrin FMIC, and a MAP v3 ported intake manifold with heat dispersant coating and phenolic spacer gasket. During WOT pulls my temps are always within 10F of ambient.

Here is a graph comparing my manifold intake temps to the MAF intake temps during a 3rd gear pull. Ambient temp was in the low 70's. The sensors or calibrated correctly and read identical when the car is off and fully cool.

 

Nice setup. Maybe you can put the stock snorkel back on as well and cut an opening for that? It may help feed cold air into your box. I still have to make a box similar to that myself.

Anyway, here are the tables from the stock ECU that controls when and how much timing gets pulled due to intake air temp (as measured at the MAF) as well as the corresponding min load table:

 

That's a neat calculator. I have bookmarked it. It does bear out what i am saying though. I used the base numbers the tables come with and increased the inlet temp by 10 degrees and it resulted in a post intercooler outlet temp of 3 degrees.

You are right about the computer pulling timing based on air temps, the Evos do this too. It is definitely worthwhile keeping the air as cool as possible.

 

I was thinking of making a scoop out of alum to feed it from the stock location, haven't gotten around to it but I'm pulling pretty much ambient air in as it is.

Good info, I hadn't really gone out of my way to dig for it but I figured it had to be there. Looks pretty similar to how DSMs are scaled.

 

Bear in mind those numbers are Celcius.

 

very good info guys!!!
Steve, that setup looks nice. I didnt even think about the ECU pulling timing by charge temps. Makes sense though as those higher temps could possibly aid to detonation.

 

thats why pre turbo WI was so succesful at lowering charge temps. altho controversial on turbo damage

 

So if we are considering fabricating a heat shield of sorts around the filter, wouldn't it be best to make the shield out of plastic (e.g. stock airbox) vs. aluminum as it would transfer heat less ?

 

Based on the high initial air temp values, it looks like you have a conefilter setup. Is that correct? If yes, is there any sort of heat shield or cold air ducting?

Where did you get the phenolic spacer?

 

This is one place http://tunertools.com/proddetail.asp?prod=GZM-G600X

 

I think most just use a sheet of Al since it's easy to work with and source. But if you look online, there are plenty of places that sell sheets of plastics, like HDPE (which I think is what the stock box is made out of). That's what I am going to make mine out of.