DIY air to water intercooler

[NSFW]

I agree the boost/afr plot looks pretty reasonable.

 

And the AF correction scatter looks famliiar to me as well. I did tweak my MAF scaling to straighten it out somewhat, but I think there is something weirdly nonlinear happening around 5 g/s and I'm not sure it matters. I always thought that was weird that the stock AF learning has the ranges that it does: 0-5, 5-10, 20-10, when the 5-10 range is used so little. I think it might be because that range acts quite a bit different from the ranges above and below it.

 

Your car knocks a lot more than mine in cruise. I'd probably get 20-40 knock sum increments in two hours or driing. My only guess is there's a rattle, but I've heard from other people who have the same issue and can't find anything rattling so I dunno.

[LGTspecB1657615261]

I have heard from others that I get more knock than they are used to.... so I am starting to think that number is high. I am pretty sure it was high before though too.... prior to the new AWIC. So its related to something else.

 

I have a theory that my car started acting "different" after I installed the injectors. The tuner tuned for the injections just as he always had, and claimed they tuned fine. But it seems like that was the point in time when I started getting odd stutters, etc. The car started driving differently in CL.... at least most of the symptoms occur at low load, low throttle.

 

So far I have no idea how the intercooler is performing other than the car just works and drives normally.

[LGTspecB1657615261]

I pushed it a little further. I hit more g/s than when I dynoed in 4th, but of course a road log is a little different than a log while on a dyno. Either way, I am making the same power, that is for sure.

 

I stopped at 5700 RPM, but it looked solid. No knock in the high end. What does it take for the FLKC chart to zero back out in the high end? My LV looks about the same.

 

Still not sure why the A/F learning numbers all shifted..... my OL AFR's look good though. High 10's... so I guess that is what is important, right?

romraiderlog_20091201_182911.csv

[E-Tank]

So f***ing cool. Can I be your apprentice? I bet there's a lot more knowledge where that came from.

[NSFW]

AF learning will wander around +/- a couple percent, that's totally normal.

 

The 1.4 knock correction will be decreased in increments of .35 when you spend time in that RPM/load range without knocking. I think it also requires being in that range with a load that's not changing rapidly - just crossing through it won't count. I've seen logs where it un-learned the correction in seconds, but my car (an 05, like yours) takes a while. Maybe a few seconds cumulative time spent in that cell? I haven't watched it closely enough to really be sure.

 

1.4 is how much timing the car pulls for a single knock event, and even the stock tune knocks once in a while, so it doesn't necessarily indicate a problem IMO. Numbers bigger than 1.4 are the ones that really matter - that means it pulled timing and still knocked.

 

EDIT: or two corrections side-by-side...

http://www.legacygt.com/forums/showpost.php?p=2727824&postcount=18

[LGTspecB1657615261]

^^ thanks NSFW! Good to know. I don't see knock in any of my logs in that range, so I think it will slowly go away.

 

UPDATE:

The car is running solid. Today I broke the rear tires in 3rd gear. The ground was a little wet, so I cheated a bit... but I was still shocked I was able to do it. I think that with the AWIC setup, I am getting colder air temps in city driving. I don't think the stock TMIC does great in city driving since there is no airflow across it. It does ok, don't get me wrong... but with the AWIC that water is staying VERY cold. I can be driving stop-and-go and when I get to my destination the water hoses feel COLD to the touch... and those are rubber hoses. Even if I drove the car hard, they still feel cold. I attribute this to the large heat exchanger and 8 GPM flowrate. The heat exchanger is large so the coolant spends a larger percentage of its life there... i.e. it spends more time in the cooling phase than it does in the heating up phase.

 

We are working out some kinks in the micro-controller code... but soon I will be able to measure inlet air temps. I still need to run some wires for that too. I am excited for that part...

[ClimberDHexMods]

Interesting heat exchanger. Can't wait for your hard data. I have the one that's only 7" tall, but 2" thick. How thick is yours? Wondering what's the better option. Obviously for fitment yours sounds very good...

[LGTspecB1657615261]

Interesting heat exchanger. Can't wait for your hard data. I have the one that's only 7" tall, but 2" thick. How thick is yours? Wondering what's the better option. Obviously for fitment yours sounds very good...

 

Mine is 2" thick and 7" tall. It fits perfectly and I think made specifically for this application. It has a brass drain valve on it, and comes with a temp sensor port if you want to install a water temp sensor. Its nice. It came painted silver as well.

 

http://www.siliconeintakes.com/images/product/radiator_26x7x2_picture.jpg

 

http://www.siliconeintakes.com/images/product/radiator_26x7x2_specsheet.gif

[LGTspecB1657615261]

I am not sure which is better - large surface area and thin, or smaller surface area and thick. This is another option you could probably go with:

 

http://www.siliconeintakes.com/images/product/radiator_24x12x1_specsheet.gif

 

I would actually think large surface area and thin is the best because of the surface area / volume ratio (higher the better for thermal exchange), but I would also imagine you might lose some velocity there. Maybe that is a good thing, gives it more time to cool. I was afraid I would start running into the grill and/or hood latching mechanism, so I did not want to try a tall and thin one. It would be close. Since the inlet and outlet are on the same side, it seems like the flow path of the water will have a gradient. Guess that is not too big of a concern.

 

The one I went with definitely looks good behind the bumper - it looks just like an FMIC core.

[ClimberDHexMods]

Ah ok we have similar versions. Mine is a little different, but same basic thing and same measurements/inlet/outlet. I thought you had the tall skinny one cause of the image in your first post. There is the 3.5" one available too, not sure how necessary that is over this one, but this style is an excellent choice for the bumper area we have.

[LGTspecB1657615261]

So it is below 20 here, one day it was about 10 degrees. The pump still seems to be working fine and everything looks good. I was not sure how the pump would react to colder temperatures and colder liquids.

 

This is the first winter for my car and the shift boot crinkles when I shift because it is so cold. I guess the leather is very dry. I should condition it or something. Never seen it so stiff like that. (that's what she said)

[Merc6]

LMAO @ shift boot. I used a few products on my seats over the time of ownership. If I remember correctly it's best to warm the interior of the car for any of the products before you apply the cleaner/conditioner. How long it takes to fully asorb 20 below i'm not 100% on.

 

Glad to hear it still works in the cold. I subscribed a page or 2 back and lurked for results and more info.

[urfsin]

Good job on the A/W intercooler!! Nice install and fab up skills, it looks good.

 

You have shown it works well in the winter, but I wonder how it will hold up in the summer.

[kb2spn]

subscribe

[LGTspecB1657615261]

car is still running good even with the cold temps

[musty_hustla]

excellent thread and data prep

[LGTspecB1657615261]

I am starting to feel chatter / vibration in the brake pedal when sitting at a light idle. From what I can tell, the TMIC is not close to hitting anything, and nothing else I installed would cause this. Perhaps the cold weather is just stiffening up my motor mounts?

 

Anyway, nothing has broken yet .

[Fraggle]

<subscribe>

Very interesting

[kinder]

I'm about to order up the stuff for my AWIC... I still need to see how much room I have to play with.

 

1) I was thinking of going with a bigger cooler. The type 4 or possibly larger. I was looking at some of the larger ones just to get the water in/outlets on the same side as that would make the cooler technically a dual pass design which I feel is a more effective design. How much room under yours do you have? I am curious if I can run a 2.5" silicone tube under a 4.5" thick cooler to an inlet on the side.

 

2) I didn't notice if you were running more than 1 spal fan. I was looking at running 2-3 fans wired in serise. Also I was considering mounting the cooler near the top of the grill for more of a sleeper look. Did you notice if there was room for that with the rad you chose? Also you comment it was a dual pass but your pics kinda imply its a single pass. (water lines comming fully from right and left sides.)

 

3) What direction is you water running when looking at your engine bay? Clockwise or counter clockwise?

 

4) Any inlet-outlet temps? I know its cold right now but it would give an idea of effectiveness.

 

Thanks and nice looking install!

[LGTspecB1657615261]

I'm about to order up the stuff for my AWIC... I still need to see how much room I have to play with.

 

1) I was thinking of going with a bigger cooler. The type 4 or possibly larger. I was looking at some of the larger ones just to get the water in/outlets on the same side as that would make the cooler technically a dual pass design which I feel is a more effective design. How much room under yours do you have? I am curious if I can run a 2.5" silicone tube under a 4.5" thick cooler to an inlet on the side.

 

2) I didn't notice if you were running more than 1 spal fan. I was looking at running 2-3 fans wired in serise. Also I was considering mounting the cooler near the top of the grill for more of a sleeper look. Did you notice if there was room for that with the rad you chose? Also you comment it was a dual pass but your pics kinda imply its a single pass. (water lines comming fully from right and left sides.)

 

3) What direction is you water running when looking at your engine bay? Clockwise or counter clockwise?

 

4) Any inlet-outlet temps? I know its cold right now but it would give an idea of effectiveness.

 

Thanks and nice looking install!

 

1) I would not get a bigger cooler. It won't fit from a piping stand-point... it would be a pain. I am not sure same-side is a better option. Plus, those big guys have 3" inlets and outlets, which is BIG for up in the TMIC area. Not enough room even for reducers. Once summer hits, I will be able to test how effective my small cooler is. The heat exchanger, water, and water flow-rate are most important. I have a pretty large heat exchanger, a pretty good flow-rate, and a good coolant/water-wetter mix.

 

2) There is no room behind the grill really, the hood latch is in the way. Paint it black and put it where I put it. Trust me, the options I outlined here are the ones that will fit. I would not deviate too far from that. There is not much room for fans. Actually, the factory radiator fans pull air past it believe it or not. I am going to have 1 fan in the center, that is all that will fit. Not sure what I am going to use to control it yet.

Mine IS a dual pass. I routed the hoses down both sides of the engine bay, but they meet back up on 1 side of the heat exchanger. Look at the pics more closely.

 

3) The hot side of the coolant lines is the hot side of the engine bay (turbo side). Counter-clockwise if looking down from top.

 

4) I have not installed the IAT sensors yet.

 

It will be a long while before I get good data. It obviously works great right now, its freezing cold out. The rubber hoses always feel ice cold, the liquid going into the cooler is pretty much ambient.

[LGTspecB1657615261]

UPDATE

 

Today I installed my Dual IAT sensor setup. I have not installed my TMIC inlet temp yet, only outlet temp. I still need to weld a bung on the inlet side of the TMIC.

 

Under normal city driving with the engine warmed up, my post-IC temps were 68 degrees. I stopped to get a quick haircut. Once I got back in and started the car, it measured 75 degrees. As soon as I started the car (and the water pump kicked on), it jumped to 71. After about 2 minutes of driving it was back down to 68.

 

So, as the days get hotter, this effect will become more predominate. But, I was glad to see immediate results when the water pump came on. With the flowrate of the water, it very quickly cycles the water in the TMIC.

 

The air temp in my area is about 45 degrees. I did not pull out the laptop to see what the stock airbox IAT sensor read.

[LGTspecB1657615261]

Another update.

 

Drove around for a bit, it settled around 66-68 degrees for city driving. Let it sit for a while, but did not let the engine cool. After startup IATs after the IC read 100 degrees. Within the first 30 seconds of starting up it dropped to 90 (just sitting in the garage). This must be due to simply cycling the fluid out of the IC. Then I pulled out into the road and drove a steady 40 mph. Within the first few minutes it was down to 66 degrees.

 

So this is all without fans. Once I install a fan, I can manually turn it on when I first start the car... that will accelerate the temp drop. Once the temp stabilizes (or I start driving at a decent speed), I can shut the fan off. I found that once you hit about 40+ mph steadily, the outlet temp starts dropping very fast. So, the heat exchanger is definitely working well.

[ClimberDHexMods]

The heat soak issue while stopping to run errands has irritated a lot of AWIC folks. One thing I figured before was to create a timer on the pump when the car is shut off (and could be tied in with the fan too or separately) which would cycle on the pump for XX seconds every 5 minutes or so. One thread suggested time delay relays, which I'm sure is very easy for you to setup. Then have them auto stop cycling after 45 minutes or whatever. It would dissipate some of the heat soak by bringing water to the heat exchanger where it is not nearly as hot. The other (and better) idea was to just wrap the intercooler in heat shielding. This would be the uglier, and definitely the most effective method. Also a combination of the two would be a definite win.

[LGTspecB1657615261]

It drops quick IMO - all you need to do is run a fan to simulate 40+ MPH driving. You could easily create a circuit that cycles the pump and fan every 5 minutes and stops after 30 minutes or an hour.

I might be able to do this with the microcontroller I already have.

 

Yesterday I was getting outlet temps that were just 15 degrees over ambient.

[LGTspecB1657615261]

So I have learned this... one way to look at a AWIC is like a capacitor. The mass of water is the capacitor. It takes a while to cool down and heat up, but when it gets there it stays. What is nice about this, your IAT's do not rise that much sitting at a stop-light. Very little actually. And if you are not moving that fast, you can still have low IATs. The issue is, when you stop for a while (10+ minutes) and the TMIC full of water just sits there and heats up in the engine bay.