Need some input, I'm building a turbo 370 with air to air intercooler .
I was reading that gm iat will saturate with heat and possibly can have as much as a 15 second delay reading the correct temp, that seems to be a problem if Temps are high.
What sensor is everyone runNing, or am I reading to much in to it,
Be easy on me this is my first turbo build.
Thanks in advance

 

Im running a stock sensor for mine and it reads fine, no delays. Im air to water and leave the line near 60 and reach low 90s at the end of the 1/8th.

 

a popular IAT sensor is the GM typhoon stock IAT. Used in stock turbo applications and is an open element plastic cage type that reads quickly.

Uses a common 2 wire pigtail and 3/8 npt threaded body.

 

All thermister type sensors have delay. I'm not aware of any way to eliminate it with another sensor. The sensor you do not want to use for IAT is the encapsulated water temperature style sensor. That one will lag a ton if used for air temperature.

A thermistor is a resistor that changes it's resistance with temperature change. They're just a little tiny thing that's dipped in an epoxy to seal them. Even in the "open element" sensors used for IAT, the actual sensing element is coated. Although fairly small, the mass of the epoxy (and the actual thermistor) takes time to either absorb or expel thermal energy until the temperature of the thermistor matches the temperature of the air.

Some guys have modified their IAT sensors by removing the thermistor and replacing it with a smaller one with less mass in an attempt to get them to read quicker. From memory, the results are improved, but still far from the near instant response a pressure transducer will provide.

The only way I can think of to get around this is a software based solution that calculates the rate of change against the current reading to predict what the actual air temperature would have to be to create that rate of change.

Shouldn't be too difficult to do. But, to my knowledge, it hasn't been done yet.

When you see a log of air temperature and it's climbing over a run... MOST of that climb is the sensor catching up to what the temperature actually is the moment you hit target boost.

As a numbers guy, that bugs the hell outta me.

The short answer for your question is, just use an open element IAT sensor. Nearly any one that's compatible with your ECU will be just fine and has produced all the great results you've heard and read about. Definitely not something to geek out on unless that's just your thing. Wouldn't give it any further thought if I were you.

 

here's a half *** hack idea. the iat sensor is linear right? couldnt you just skew the scale so it reads a little higher

you could test it to find the delta and approximate off of that

just spitballin. i'm not a numbers guy

 

Spit balling is all I've got too. I'm a numbers guy, but certainly not a scientist with a PHd in Physics. I'm sure, if I brought up the thought with the right guy, he'd have it figured out by the end of the day.

Seems it wouldn't be too difficult if you knew (or could deduce from testing) the thermal mass of the thermistor and the thermal transfer qualities of the epoxy coating material...

If I absolutely had to, or just have the drive to figure it out sometime (which could be a fun little challenging pet project), it wouldn't be too difficult to rough something in and iron out the wrinkles, to an illustrious "almost there" rendition of a working concept. LOL

 

Megasquirt has an option to Ignore mat correction while after start enrichment is active to get rid of any perceived heat soak

On that note, on hot days I've staged late against the all motor cars with hoods. I'm intercooled and have a correction table, those nasty small blocks don't and suffer some as under hood temps rise!

 

Are you melting all your ice? I looked at a 1/4 mile datalog from last year with my a/w setup and I left at 78*, dropped mid run to 73* and ran the 1/4 at 85*. I'd melt 8lbs of ice in the 1/8 and would melt 16lbs in 1/4

 

No, they aren't linear

 

derp, 15 seconds of googling got me this

 

Stock GM plastic IAT (open cage) smacked into my truck intake air purge port) Datalogs show a 18* drop in a 4 sec boost pull with a small nozzle of 50/50 water meth and W2A intercooler. Id say your over reading this lol.

Dont get me wrong they aren't the fastest sensor but the new open cage ones aren't that slow either. I've used the FAST versions with the brass screw in 3/8"NPT style and worked fine as well.

 

Maybe use a fast acting thermocouple instead of thermistor? They are pretty linear and more accurate. also usually read much higher. PLX makes a nice one. Though I'm not seeing a delay time listed in the specs. I had planned on using it on a hot air setup and wanted the higher temp range.

Becomes much more affordable when you start stacking other gauges with it. You can use one of their gauges and run several sensors to it. I use one for boost and WB02 at the moment.

http://www.plxdevices.com/Air-Intake...7346002214.htm

 

Nice find Forcefed86! I was looking at another OB2 port gauge and its like $250-300 for just the gauge. I tried a EGT K thermocouple in m intake charge pipe and it heat soaked and skewed the readings but it also didn't work with a damn for accurate readings... kinda just stayed at one temp and would just rise but never go down. But it wasn't the open style Kthermocouple like the one you posted. $250 could be pretty sweet for an IAT gauge with the ability to provide all my stock perimeters.

 

Also with the MS3 (or any aftermarket ECU really) you can get away with the box and sensor only. No real need for a gauge.

Their gauges are cool though. Touch screen and fancy like. You can get the larger OBD gauge if you use the factory ECU and it will display multiple inputs at the same time, lots of useful information. The IAT sensor will also plug into the OBD gauge.