where are you guys putting your IAT's?

[740ATL]

Guys, 2 questions... now that y'all have been running MS for a while now and have a good deal of experience, where are you putting your IAT sensors? I've seen them screwed into the backs of the b230 intake manifolds and installed mid-way up the cold side pipe.

Got any favorite locations?

Also, where exactly do you find them on GM cars? Next time I make a JY run, I want to pick up a couple...

many thanks.

Mike

[Matt Dupuis]

I never liked it in the intake mani myself. Picks up too much manifold heat, and in that location there isn't very much air movement to keep the sensor reliable.

I tried it in the IC-TB pipe, both right at the exit of the IC and right before the TB. Heat soak was an issue in both locations, but it was worse closer to the engine.

I tried it in the filter-compressor pipe, right behind the filter, but it was also too close to the radiator. It really should have been in front of the rad support.

My best results have come from zip-tying the sensor to the front bumper shock. Most consistent temps and the AFR doesn't really change that much. Heat soak is totally eliminated, and hot starts are MUCH better. Personally I feel that you can tune your VE table to adjust for what's really a pretty predictable temp gain in boost, and you don't really want that hotter air pulling out fuel like the IAT sensor will do when it sends a hot signal.

IAT is more important for timing anyway, and there's almost zero IAT compensation for timing anyway.

[kyle242gt]

Hi Mike-

I put my IAT in the driver's side of the cold side pipe, at the big bend near the IC. Since I went NPR, it's more like the top of the pipe. You don't want it oriented in such a way that it can get soaked with oil from the turbo. (ie bottom of pipe)

I found mine in a smallish GM nothing car (88 or so); V6, mounted in the airbox. I also found some in Corsica/Beretta (sp?) but they had a non-threaded mounting surface.

[Matt Dupuis]

Oh yeah - I meant to add that if you have an aluminum-tanked IC, I'd think the best mounting location would be threaded into the cold side tank, just by the outlet nipple, provided that this is in front of the rad support/engine compartment.

I also get my IATs out of either turbo Pontiac Grand Ams or second-gen Quad-4s with the cast aluminum "log" intake manifold. Both times they're open cage elements and both times they're threaded into the intake mani, but unlike our intake manifolds the sensor is mounted such that air always flows past them.

Edit - I've thought of using a simple K-type thermocouple that's squeezed between the IC and the silicone hose, and it'll respond MUCH quicker to IAT without responding to underhood temps. Boris has something like this, though I don't know if he's mounted it in a heat sink or not. Do an easytherm on the thermocouple and all should be well.

[linuxman51]

Anywhere heat soak can be minimized is a good spot. On SwedishFish's car we stuffed it in behind the air filter (hope it doesnt come loose for his turbo's sake), and it did fine as near as i can tell.

Density change after measurement isn't likely to be as important, as there shouldn't be any change in the amount of air once its inside the system.

[740ATL]

Thanks guys!

I really appreciate it!

Mike

[ovlov760]

Mine is mounted mid-pipe in the cold side ic pipe

[The Aspirator]

Mine is mounted mid-pipe in the cold side ic pipeDitto for the 242 and 244. Search for Boris' thread about his thermocouplers, they're wicked.

So Matt, you're saying that MS doesn't really need to know the temp rise during boost? Weirdly interesting...

[PetesDustyVolvo]

I have mine in the cold side intercooler pipe right above the alternator.

I have to throw in my relatively unqualified disagreement about the motor not needing to know about the temp change in boost. I would think if you were roadtripping in the west coast, and for some reason didnt have a wide band plugged in, or not enough controller authority on the gego then you would run into problems on big hill climbs even if you werent boosting that much. While your intake temps would be rising steadily, your temp sensor would be picking up a steady decrease in temperature as you gained altitude. It can happen fast at night.

While your VE table might have compensations for temp rises from WOT boosting, you would be caught unprepared for high load over a long period of time. If you disagree, go climb a 5000 mountain and watch your intake temps.

I have found the heat soak problem to be annoying as well. During hot starts it can make navigating a parking lot before you can get some air on your IC very annoying.

[hockey930]

I had a plan to put mine in the ic piping however now that I read this I don't know what to do. I am confused.
Eamonn

[benflynn]

coldside pipe

[hockey930]

What part of the cold side? The middle? and we are talking about the pipe from the ic to tb right?
Eamonn

[Matt Dupuis]

I have to throw in my relatively unqualified disagreement about the motor not needing to know about the temp change in boost. I would think if you were roadtripping in the west coast, and for some reason didnt have a wide band plugged in, or not enough controller authority on the gego then you would run into problems on big hill climbs even if you werent boosting that much. While your intake temps would be rising steadily, your temp sensor would be picking up a steady decrease in temperature as you gained altitude. It can happen fast at night.

While your VE table might have compensations for temp rises from WOT boosting, you would be caught unprepared for high load over a long period of time. If you disagree, go climb a 5000 mountain and watch your intake temps.

Okay, so examine what happens when the IATs climb: Air becomes less dense, and less fuel is needed to maintain a set AFR. Also, the air and fuel are that much closer to ignition temps, plus the EGT goes up, so spark lead needs to be reduced to compensate, or the engine needs to get slightly richer to slow down the burn (might increase EGT, but is a side effect to the more important detonation control).

Because Megasquirt's IAT calculation is flawed (and I stand by that, based on the -45 to 100 Deg. F situations I've personally driven the car in), the car WILL get much leaner than it deserves to be once you do your uphill boost. The IAT sensor applies WAY TOO MUCH fuel correction. I've cut it in half with good results, and cut it into 1/4 with better results, but it gets to be a PITA to recompile it with my personal airdenfactor.inc file every time a new version comes out. Hacking airdenfactor doesn't get rid of heat soak issues, though, and that's my biggest beef.

There is a function that pulls spark lead based on IAT, and I'm sure it's useful, but with the fuel table getting so horribly lean at hot intake temps, I'd gladly run 2-3 degrees less timing all the time and have enough fuel when things get really hot. Alternatively one could hack the airdenfactor.inc file until they got it right FOR THEIR VEHICLE, and then they'd be able to use the ignition/IAT function too, but I'll pass on both, thanks.

For clarification: As the IAT sensor readings increase, fuel is removed from the calculation. MS bases this factor on the Ideal Gas Law, which neither Air nor Air/Fuel follow. Also the Ideal Gas Law would require a reliable measurement of the temperature of the air JUST as it enters the cylinder, not 2-10 feet upstream where it will surely be colder than the acutal air entering the cylinders will be.

The amount of heat picked up between the sensor and the intake valve is a variable - based on how hot the area around the intake piping is, how far the sensor is from the valve, and how fast the air is moving, not to mention air density, thermal conductivity of the intake materials, and the moisture content in the air.

Megasquirt designers have NO idea of what these variables will be when you install their system into your cars, so they gave you something and the rest is DIY. Don't fall into the trap of believing that Bowling and Grippo know best - they designed it to be simple and easy, and to work on any car around. Doing what the OEMs spend millions of dollars on, testing and modifying and re-testing for every model and engine option... that's up to you.

[wildmanben]

Mine's midway up the cold side pipe, and it works great there.
-Ben

[PetesDustyVolvo]

Megasquirt designers have NO idea of what these variables will be when you install their system into your cars, so they gave you something and the rest is DIY. Don't fall into the trap of believing that Bowling and Grippo know best - they designed it to be simple and easy, and to work on any car around. Doing what the OEMs spend millions of dollars on, testing and modifying and re-testing for every model and engine option... that's up to you.

Wow Matt, Ok where to start? You've obviously thought this through a ton more than I have but I have thought about it a fair amount. I agree that atmospheric temperature changes and driving style can screw up your tune based on real-life observation.

To help me summarize your thoughts:

The amount of temperature change between the IAT sensor and the intake valves is variable between cars to the point that Bowling and Grippo have no way of providing an accurate compensation. Then, the VE table cannot provide for this discrepancy because the proper compensation would be too much reflected in driving style and load conditons.

Is that a good blurb?

Also,

I've only recently put my IAT retard to use and it works great for me. I took a roadtrip today to Joshua Tree which was about 200 miles total. It took about 8-9 gallons. So I'm pretty much right where the OEM tune was. 22-24 mpg at like 70-75 mph, absolutely no pinging even though the temperature swung from 45-85 degrees F.... So anyway, at least within the little world of Socal my tune can flex its muscle without trouble. That and I don't have to worry about LH pulling all the timing because it feels like it.

Back to the temperature stuff... if the IAT compensation is colder than it should be, everything should be rich. So if you do your VE table on a hot day under heavy load you should be "OK," if not running at max efficiency. OK I'm pretty much rambling now.

[linuxman51]

I have to throw in my relatively unqualified disagreement about the motor not needing to know about the temp change in boost. I would think if you were roadtripping in the west coast, and for some reason didnt have a wide band plugged in, or not enough controller authority on the gego then you would run into problems on big hill climbs even if you werent boosting that much. While your intake temps would be rising steadily, your temp sensor would be picking up a steady decrease in temperature as you gained altitude. It can happen fast at night.

While your VE table might have compensations for temp rises from WOT boosting, you would be caught unprepared for high load over a long period of time. If you disagree, go climb a 5000 mountain and watch your intake temps.

I have found the heat soak problem to be annoying as well. During hot starts it can make navigating a parking lot before you can get some air on your IC very annoying.


Air in still has to equal air out regardless of how hot your intake manifold is, how ****ty your intercooler is, etc. Think about it that way. If it really mattered pre-intercooler vs post intercooler dont you think volvo would have put the AFM post intercooler?

[Matt Dupuis]

That's a bit different, Ken. In that case, air mass is all calculated in the same place, at the same time, and no matter what the engine is doing load or speed wise, it's all present and accounted for. With a speed density system, air temperature is a vital part of the density equation and cannot be ignored, however in our world of turbocharged engines, the net temp rise from boost minus intercooler is a predictable enough variable and it's perfectly acceptable to put the IAT in front of the turbo. At least, until the intercooler heat soaks and temps rise, for which we must dial in a small margin of safety.

[SwedishFish]

Anywhere heat soak can be minimized is a good spot. On SwedishFish's car we stuffed it in behind the air filter (hope it doesnt come loose for his turbo's sake), and it did fine as near as i can tell.

Density change after measurement isn't likely to be as important, as there shouldn't be any change in the amount of air once its inside the system.

It works well in all the times I have got the bitch hot as **** and still driven the hell out of it...but with no intake tube, maf or filter in front of my turbo now I have wondered where it might go. I think I'll stuff it past the filter in this new fangled pipe I'm making.

[Matt Dupuis]

If you're going to put it before the turbo anyway, consider getting it right out of the engine compartment like I did. Makes everything way more stable, hot starts and in traffic!

[blkaplan]

http://www.pbase.com/blkaplan/image/40214000/large.jpg

[hockey930]

a picture is worth a thousand words.
Eamonn

[coldfusion21]

i put mine where the idle motor tube connects to the bottom of the b230 cold side pipe.

sensor was in some fwd chevy, in the airbox. you have to get the connector as well, and make sure to unthread the sensor, and then rip out the little piece it threads into. its a brass threaded peice that is press fitted into the plastic airbox. drill good sized hole in pipe, jb weld brass peice in, and screw sensor in...