heat soaking IAT sensor or temp curve in TunperP
09-25-2017, 01:23 PM
#1
Supreme Member
Thread Starter
heat soaking IAT sensor or temp curve in TunperP
All this playing around with the IAT sensor and MAT tables lately...
(This is more out of curiousity than anything else, since the car is still running really well).
My cold air tube is entirely non-metallic and that IAT sensor (an LT1 unit) was located right in front of the throttle body. On hot days (say on a 90F day), it could get up to 130F. I always found that a little puzzling to say the least, as with a non-metallic tube, one wouldn't expect the air to heat up that fast traveling through it (much less so at higher air flows than at idle at that). Rather you'd expect the temperature to be just a few deg over ambient.
So after playing around with the MAT Compensation Counts table the last few days, I relocated the MAT sensor to right at the air filter. Air filter is located under the car. If this was a third gen, it would be directly under the chassis where the battery is located.
In a location like that, you'd DEFINITELY expect the air temp to be just slightly over ambient (especially with the IAT sensor taking readings right inside the air filter itself. But, I'm only getting a marginal improvement (it's mid-80's over here and inlet air temps are hitting ~115). It'll cool down a little further on the freeway, but not enough IMO to make logical sense.
The IAT sensor is sitting in a rubber grommet, in a non-metallic part of the air filter. I literally drilled a hole in the snout of the air cleaner and put the sensor there.
Which all leads me to wonder if I'm really getting an accurate reading of the air temperature. Seems like three possibilities exist...
1.) The temperature curve of the LT1 sensor is not what the ECM is expecting... Or it doesn't match the curve in Tunerpro (and therefore TP is telling me something different than what the sensor is telling the ECM, which would make this a reporting issue and not an operational issue)?
2.) The sensor somehow is getting heatsoaked to the point where it's giving a temperature reading of what's its touching rather than the air. This seems unlikely to me given how thermally insulated the body of the sensor is from everything else. The only thermally conductive part of the sensor that is exposed to anything is the element and that's only in contact with the air stream.
3.) Somehow the IAT sensor wires are heating up and conducting that heat to the sensor element where the air flow is being used to dissipate that heat, but in the meantime the sensor element is reporting on that wire heat rather than heat in the air stream. Copper is thermally very conductive, so I'm starting to wonder.... any heat the copper builds up is simply going to race over to the nearest heatsink... in this case it could be the sensor element dissipating that heat into the airstream. My IAT sensor wires are passing by the engine side of the radiator for a significant distance (plenty of opportunity to absorb heat), and just being in the engine compartment itself, I'm wondering if it's absorbing heat and delivering it to the IAT sensor. I may test this by wrapping MAT sensor the wire harness in some sort of heat shielding sleeve. It makes a pretty long run separated from the rest of the harness, so it'll be easy to wrap it.
(This is more out of curiousity than anything else, since the car is still running really well).
My cold air tube is entirely non-metallic and that IAT sensor (an LT1 unit) was located right in front of the throttle body. On hot days (say on a 90F day), it could get up to 130F. I always found that a little puzzling to say the least, as with a non-metallic tube, one wouldn't expect the air to heat up that fast traveling through it (much less so at higher air flows than at idle at that). Rather you'd expect the temperature to be just a few deg over ambient.
So after playing around with the MAT Compensation Counts table the last few days, I relocated the MAT sensor to right at the air filter. Air filter is located under the car. If this was a third gen, it would be directly under the chassis where the battery is located.
In a location like that, you'd DEFINITELY expect the air temp to be just slightly over ambient (especially with the IAT sensor taking readings right inside the air filter itself. But, I'm only getting a marginal improvement (it's mid-80's over here and inlet air temps are hitting ~115). It'll cool down a little further on the freeway, but not enough IMO to make logical sense.
The IAT sensor is sitting in a rubber grommet, in a non-metallic part of the air filter. I literally drilled a hole in the snout of the air cleaner and put the sensor there.
Which all leads me to wonder if I'm really getting an accurate reading of the air temperature. Seems like three possibilities exist...
1.) The temperature curve of the LT1 sensor is not what the ECM is expecting... Or it doesn't match the curve in Tunerpro (and therefore TP is telling me something different than what the sensor is telling the ECM, which would make this a reporting issue and not an operational issue)?
2.) The sensor somehow is getting heatsoaked to the point where it's giving a temperature reading of what's its touching rather than the air. This seems unlikely to me given how thermally insulated the body of the sensor is from everything else. The only thermally conductive part of the sensor that is exposed to anything is the element and that's only in contact with the air stream.
3.) Somehow the IAT sensor wires are heating up and conducting that heat to the sensor element where the air flow is being used to dissipate that heat, but in the meantime the sensor element is reporting on that wire heat rather than heat in the air stream. Copper is thermally very conductive, so I'm starting to wonder.... any heat the copper builds up is simply going to race over to the nearest heatsink... in this case it could be the sensor element dissipating that heat into the airstream. My IAT sensor wires are passing by the engine side of the radiator for a significant distance (plenty of opportunity to absorb heat), and just being in the engine compartment itself, I'm wondering if it's absorbing heat and delivering it to the IAT sensor. I may test this by wrapping MAT sensor the wire harness in some sort of heat shielding sleeve. It makes a pretty long run separated from the rest of the harness, so it'll be easy to wrap it.
Last edited by ULTM8Z; 09-25-2017 at 03:23 PM.
09-25-2017, 07:00 PM
#2
Supreme Member
Thread Starter
Re: heat soaking IAT sensor or temp curve in TunperP
I found this...
https://ls1tech.com/forums/pcm-diagn...mperature.html
In which someone posted this...
http://redshift.homestead.com/Schema...mp_vs_Ohms.jpg
It seems to match the resistance curve of the GM coolant temp sensor (which the TPI's are set up for).
http://www.slantsix.org/articles/dibiase_efi/chart.jpg
https://ls1tech.com/forums/pcm-diagn...mperature.html
In which someone posted this...
http://redshift.homestead.com/Schema...mp_vs_Ohms.jpg
It seems to match the resistance curve of the GM coolant temp sensor (which the TPI's are set up for).
http://www.slantsix.org/articles/dibiase_efi/chart.jpg
09-26-2017, 08:06 AM
#4
Moderator
iTrader: (1)
Join Date: Mar 2002
Location: Chasing Electrons
Posts: 18,399
Likes: 0
Received 215 Likes
on
201 Posts
Car: check
Engine: check
Transmission: check
Re: heat soaking IAT sensor or temp curve in TunperP
There is self heating due to the pull-up resister.
The road surface also gives off plenty of heat.
RBob.
The road surface also gives off plenty of heat.
RBob.
Last edited by RBob; 09-26-2017 at 08:49 AM. Reason: spelling
09-26-2017, 09:21 AM
#5
Supreme Member
Thread Starter
Re: heat soaking IAT sensor or temp curve in TunperP
I guess what I can try is to take a thermometer and hold it near the air filter after a long drive and see if it reads the same temperature as the sensor (or how far off it is).
09-29-2017, 07:48 AM
#6
Supreme Member
Thread Starter
Re: heat soaking IAT sensor or temp curve in TunperP
Well, I can rule out scenario 3 above. I got hold of another LT1 sensor and this time I zip tied it to the chassis right next to the air filter.
Drove the Camaro into work this morning (about a 25 min drive) and when I got in, I unplugged the sensor in the filter and plugged into this spare one and they were within 4 deg of eachother. If the one that was in use was getting significantly heated by the wires, I would have expected TP to read a large temperature drop upon plugging in the other one. Just to get as much heat on the wires as possible, I turned on the A/C as well so that the condenser heat would be getting blown into the engine compartment as well.
Reaching for the MAT sensor, the air around the filter did feel a warmer than the ambient air outside the car. So then my guess is air from the engine compartment is still bleeding through the gaps in the chassis panels and mixing with the ambient air around the filter prior to getting into the filter. When driving on the freeway, the heated air in the engine compartment must must be bleeding over to the filter as well.
I'm kind of inclined to discount scenario 2 since it seems to just flat out violate the laws of physics!). Scenario 1 appears to be incorrect as well based on the temp vs resistance curves found online.
So, it looks like the readings are probably accurate. In which case, I'm going to have to isolate the filter even more and somehow build up some sort of air duct that pulls air from in front of the car... some sort of scoop perhaps.
More to come!
Drove the Camaro into work this morning (about a 25 min drive) and when I got in, I unplugged the sensor in the filter and plugged into this spare one and they were within 4 deg of eachother. If the one that was in use was getting significantly heated by the wires, I would have expected TP to read a large temperature drop upon plugging in the other one. Just to get as much heat on the wires as possible, I turned on the A/C as well so that the condenser heat would be getting blown into the engine compartment as well.
Reaching for the MAT sensor, the air around the filter did feel a warmer than the ambient air outside the car. So then my guess is air from the engine compartment is still bleeding through the gaps in the chassis panels and mixing with the ambient air around the filter prior to getting into the filter. When driving on the freeway, the heated air in the engine compartment must must be bleeding over to the filter as well.
I'm kind of inclined to discount scenario 2 since it seems to just flat out violate the laws of physics!). Scenario 1 appears to be incorrect as well based on the temp vs resistance curves found online.
So, it looks like the readings are probably accurate. In which case, I'm going to have to isolate the filter even more and somehow build up some sort of air duct that pulls air from in front of the car... some sort of scoop perhaps.
More to come!
Last edited by ULTM8Z; 10-02-2017 at 08:44 AM.
10-07-2017, 04:04 PM
#7
Supreme Member
Thread Starter
Re: heat soaking IAT sensor or temp curve in TunperP
Interesting developments...
I built some heat shielding around the air filter to keep hot engine compartment air from getting to the filter. Actually works very well.
Went out to San Bernadino today for the Route 66 Rendezvous... about 100 degF outside, yet my IAT temperature never got above 115 degF. There's a cruise route within the show, so this was even cruising around at 10 mph in 100 degF heat.
The nice thing is, once I get up to speed (not even freeway speed), it cools down to within 5 degF of ambient temperature almost immediately. On the freeway, it's right at ambient temperature!
I had to reset my MAT Compensation Counts table to factory and am starting again. But it seems like it's going to remain closer to factory values than where I was headed previously.
I built some heat shielding around the air filter to keep hot engine compartment air from getting to the filter. Actually works very well.
Went out to San Bernadino today for the Route 66 Rendezvous... about 100 degF outside, yet my IAT temperature never got above 115 degF. There's a cruise route within the show, so this was even cruising around at 10 mph in 100 degF heat.
The nice thing is, once I get up to speed (not even freeway speed), it cools down to within 5 degF of ambient temperature almost immediately. On the freeway, it's right at ambient temperature!
I had to reset my MAT Compensation Counts table to factory and am starting again. But it seems like it's going to remain closer to factory values than where I was headed previously.
Trending Topics
10-13-2017, 08:13 AM
#8
Supreme Member
Re: heat soaking IAT sensor or temp curve in TunperP
I moved my MAT to the air filter housing just after the air filter itself. In that area is also the ambient air temp sensor for use with the Automatic HVAC system. So I get to see temp readouts as I drive. Its not completely analogous to intake air temps but it gives me a range. Idling is where you see the greatest variability in the temps because of the reduced air flow and heat coming off the radiator. C4 Vettes are "bottom breathers" so road surface heat also plays a role.
In the stock calibration, the "neutral values" for the IAT were basically at the normal operating coolant temps, ie 80C+. With the IAT further forward, I ended up using similar values but at the much cooler operating temps of the air flow
just behind the radiator which is 23-39C depending on weather and current air flow.
While datalogging with the car idling, I note the intake temp and watch the BLM.
BASICALLY, THE VALUES i use are to keep BLMs in the same range as the intake air falls or rises.
In the stock calibration, the "neutral values" for the IAT were basically at the normal operating coolant temps, ie 80C+. With the IAT further forward, I ended up using similar values but at the much cooler operating temps of the air flow
just behind the radiator which is 23-39C depending on weather and current air flow.
While datalogging with the car idling, I note the intake temp and watch the BLM.
BASICALLY, THE VALUES i use are to keep BLMs in the same range as the intake air falls or rises.
10-13-2017, 11:49 AM
#9
Supreme Member
Thread Starter
Re: heat soaking IAT sensor or temp curve in TunperP
Yeah I have an aftermarket rear view mirror with an ambient temp readout. With car "cold" prior to start up, it's within 3-4 deg of IAT readout in TP, so I keep that delta in mind when I do comparisons . Temp sensor is mounted in under the bumper in the front of the car so it's not too "contaminated" by underhood temps during operation. So I'm able to always compare in real time as well.
Last edited by ULTM8Z; 10-13-2017 at 12:03 PM.
10-19-2017, 08:19 PM
#10
Re: heat soaking IAT sensor or temp curve in TunperP
How much of an improvement can be had by moving the IAT sensor from the plenum to the air filter housing like on the V6 cars? I only ask because I just got back from datalogging and the guy who went with me was telling me that the sensor is getting heat soaked. Not sure when the ECM pulls timing when it sees high IAT temps.
10-20-2017, 07:48 AM
#11
Supreme Member
Thread Starter
Re: heat soaking IAT sensor or temp curve in TunperP
It's not just a matter of moving the sensor. You have to also adjust the calibration to take advantage of it. I forget what mask you're running, but in AUJP, you have to do three things...
1.) Enable the switch that enables use of the Inverse MAT table (tells the ECM that you're now running a true IAT sensor, and not the coolant sensor GM stuck in the plenum to read air temps.
2.) Change the Inverse MAT table. I started with the LT1 table and modified it (mostly by seat of the pants feel).
3.) Some people say that you don't need to modify the MAT Compensation Counts table, but (as you can read in this thread), I found that it does make a significant difference. Though you need a wide band (and TP to read it) in order to what I'm doing. Though, you could use just the BLM's to gauge consistency if you don't have a WB. GM programmed in an adjustment to fuel trim based on increasing intake air temps, but I think they overdid it IMO... For the temperature ranges I spend most of my time in, I brought the compensation values much closer together.
The improvement I felt has more to do with consistency rather than making more power, such that the car always feels like right after you start it up and get into closed loop (everything is relatively cool under the hood... that "cold car" feel). BLMs stay consistent as do the PE AFR's. So I guess one could say that it does make more power, in a sense, that when the car is hot, it's not losing any power.
1.) Enable the switch that enables use of the Inverse MAT table (tells the ECM that you're now running a true IAT sensor, and not the coolant sensor GM stuck in the plenum to read air temps.
2.) Change the Inverse MAT table. I started with the LT1 table and modified it (mostly by seat of the pants feel).
3.) Some people say that you don't need to modify the MAT Compensation Counts table, but (as you can read in this thread), I found that it does make a significant difference. Though you need a wide band (and TP to read it) in order to what I'm doing. Though, you could use just the BLM's to gauge consistency if you don't have a WB. GM programmed in an adjustment to fuel trim based on increasing intake air temps, but I think they overdid it IMO... For the temperature ranges I spend most of my time in, I brought the compensation values much closer together.
The improvement I felt has more to do with consistency rather than making more power, such that the car always feels like right after you start it up and get into closed loop (everything is relatively cool under the hood... that "cold car" feel). BLMs stay consistent as do the PE AFR's. So I guess one could say that it does make more power, in a sense, that when the car is hot, it's not losing any power.
10-20-2017, 09:22 AM
#12
Supreme Member
Re: heat soaking IAT sensor or temp curve in TunperP
Before adjusting the MAT table, the idle on my car would get pretty ratty as the motor and intake air would get hotter. This was really noticeable after u took the car for a long hard run in the summer. Now w the MAT table adjusted, the BLMs, particularly at idle, stay much more consistent as does the idle itself.
10-20-2017, 09:44 AM
#13
Supreme Member
Thread Starter
Re: heat soaking IAT sensor or temp curve in TunperP
Yes... consistently smooth idle is also a benefit. Mine never got ratty per se, but it definitely wasn't as smooth with hot IAT temps as it was with cold temps.
Last edited by ULTM8Z; 11-09-2017 at 11:45 AM.
