IAT/ACT sensors behavior

[lito]

We have been discussing this for quite some time so I wanted to make a litlle log so people can see what I mean when I say these sensors suffer from big latency and need to be read differently. The thing is that you can't read actual temps and take them as actual instantaneous values and see the actual slope to have a better view of what is going on IAT/ACT wise.

Test was conducted this way, with a cool engine bay we removed the MAF sensor, this was done with a 2011 Ford Ranger that still use a Hitachi type voltage MAF and a powerPC PCM like the 05-10 mustangs and used "lab grade" ambient and hot air generator equipment, an small 110V fan and a Vidal Sassoon hair drier :-D. Both were being run constantly so their temps were quite stable and alternated over the MAF trying to keep an steady flow, cold, hot, cold and see the time it took the temp sensor stabilize on the new temps read.

Here is an screenshot of an extract where we got to keep quite an steady flow and how it took over 7 seconds to stabilize on the new hot reading.

So the idea behind this is that when you read an air temp you need to see its curve and study its slopes, if you see a raising slope, the actual air temp will be higher, depending on how high is the slope, the bigger the delta between actual and read temps, the same thing works on the other way, a failing temp means the actual air is colder than shown and the slope will help us guess how big is the delta.

So when you see things like this:

You can be absolutely sure your IATs are quite high and you are in trouble. In that case, actual intake temps could easily be over 200F at times.

 

[JeremyH]

So if I understand correctly your saying the iat might have been hotter but didn't have time to stabilize before the pull/log was done. Interesting, so does the pcm infer anything based on the slope and mass air flow?

 

[Pentalab]

So obviously the oem thermistor used is not fast enough. I was going to say the Fluke 62 IR, point and shoot thermometer I use is instantaneous.... but it does not read air temps, only surface temps.

Flip side to all this is... the MAF sensor is relocated to the aluminum intake manifold, if a blower is used, like the roush units. Cruising at 50 mph in OD, 1.5 krpm, IAT's could be 120 F... even though outside air is 70 F. the air is moving past the manifold slowly...and picks up heat. Hit the gas... or drop down a gear or 2... and increase rpm. IAT drops...since the same 70F air is now moving faster past the manifold. It doesn't get a chance to pick up as much heat from the 200F manifold. I can modulate the IAT's with my right foot.

Nothing new here. There may be better, faster thermistor's out there, but I doubt any of em are instantaneous. They have used a less than optimum device for the job. The yellow lines (measured) should be tracking the blue lines ( actual, real time) in the above graphs. They don't. The yellow is averaged over XXX secs.

Its bad enough on the 1st graph.... but the 2nd one is worse. You are alternating rapidly between 186F and 74F ..and the thermistor fails miserably.

 

[lito]

So if I understand correctly your saying the iat might have been hotter but didn't have time to stabilize before the pull/log was done. Interesting, so does the pcm infer anything based on the slope and mass air flow?

Basically that people should not take a single IAT/ACT read as a correct value and see its trending because of this latency. And no, nothing that I've seen on that, they just play safe.

So obviously the oem thermistor used is not fast enough. I was going to say the Fluke 62 IR, point and shoot thermometer I use is instantaneous.... but it does not read air temps, only surface temps.

Flip side to all this is... the MAF sensor is relocated to the aluminum intake manifold, if a blower is used, like the roush units. Cruising at 50 mph in OD, 1.5 krpm, IAT's could be 120 F... even though outside air is 70 F. the air is moving past the manifold slowly...and picks up heat. Hit the gas... or drop down a gear or 2... and increase rpm. IAT drops...since the same 70F air is now moving faster past the manifold. It doesn't get a chance to pick up as much heat from the 200F manifold. I can modulate the IAT's with my right foot.

Nothing new here. There may be better, faster thermistor's out there, but I doubt any of em are instantaneous. They have used a less than optimum device for the job. The yellow lines (measured) should be tracking the blue lines ( actual, real time) in the above graphs. They don't. The yellow is averaged over XXX secs.

Its bad enough on the 1st graph.... but the 2nd one is worse. You are alternating rapidly between 186F and 74F ..and the thermistor fails miserably.

We can't do anything as of now, there are faster sensors out there but mostly sure they don't share our temperature/electrical behavior with what's in the calibration and the transfer function for the IAT/ACT have not been mapped on any of the tuning platforms available that I now.

 

[one eyed willy]

I just took the easy route.....turbo ftw! Good info!

 

[Department Of Boost]

I just took the easy route.....turbo ftw! Good info!

All psi increases create heat, it doesn't mater what method of compression is used. In a "lab" or on paper you get a 9.89deg/psi increase. Of course nothing is that efficient at compressing air.

A well sized turbo will make about 12.5deg/psi.

A well sized screw blower is 11.75deg/psi.

A TVS screw is about 13.5deg.

A roots improved about 14deg.

There are no free lunches with boost.

 

[Department Of Boost]

Great info Lito. Now I'm scared to look at my logs!

 

[hamish]

All temperature probes or instruments will have a delay if temperatures are changing.
It's on the control side that you can be proactive or reactive.

 

[Pentalab]

All temperature probes or instruments will have a delay if temperatures are changing.
It's on the control side that you can be proactive or reactive.

Agreed. By factoring in gas pedal position, etc, etc, etc, and knowing the typ delay on the measurement device used, and typ temp rise over the same delay, the software could be tweaked to account for the delay. It's as good as you can get.

Interesting test he did though. You could take 2 x identical hair dryers, one with low heat position on, and the other with heat off, and both with identical cfm.... and alternate between em, which lito did (using a fan for the no heat position) and experiment with various temp sensors.

No doubt the software tune has factored in the delay. Suffice to say, the iat's are higher than what your aeroforce or other gauge used.... plugged into obd port..when wot.

 

[s8v4o]

All psi increases create heat, it doesn't mater what method of compression is used. In a "lab" or on paper you get a 9.89deg/psi increase. Of course nothing is that efficient at compressing air.

A well sized turbo will make about 12.5deg/psi.

A well sized screw blower is 11.75deg/psi.

A TVS screw is about 13.5deg.

A roots improved about 14deg.

There are no free lunches with boost.

No free lunches but at least with a turbo or centri you can eat steak instead of ham sandwiches.

 

[Department Of Boost]

No free lunches but at least with a turbo or centri you can eat steak instead of ham sandwiches.

Huh?

 

[Bobby]

we are getting extreamly low iat on our turbo kit. and i dissagree on the heat. my car proved that at mustang week in the dyno shoot out. all the twin screw guys cars didnt make the power they normally made because of the heat but my car made what is allways does. i think it is more in how it is set up (placement and efficiency of the kit and i think we have it pretty good)

 

[s8v4o]

Huh?

What I'm getting at is packaging. IMO a centri or turbo car allows for more intercooler, especially without breaking the bank.

 

[Department Of Boost]

we are getting extreamly low iat on our turbo kit. and i dissagree on the heat. my car proved that at mustang week in the dyno shoot out. all the twin screw guys cars didnt make the power they normally made because of the heat but my car made what is allways does. i think it is more in how it is set up (placement and efficiency of the kit and i think we have it pretty good)

Boost is boost. It's physics and thermodynamics. It's like gravity, there is no cheating it.

10psi makes XXX amount of heat. 20psi makes double that. 30psi makes triple.

"When air at atmospheric pressure is mechanically compressed by a compressor, the transformation of air at 1 bar (atmospheric pressure) into air at higher pressure (up to 414 bar) is determined by the laws of thermodynamics. They state that an increase in pressure equals a rise in heat and compressing air creates a proportional increase in heat. Boyle's law explains that if a volume of a gas (air) halves during compression, then the pressure is doubled. Charles' law states that the volume of a gas changes in direct proportion to the temperature. These laws explain that pressure, volume and temperature are proportional, change one variable and one or two of the others will also change, according to this equation"

Link:
http://www.compair.com/About_Us/Compressed_Air_Explained--02What_happens_when_we_compress_air-q-.aspx

 

[Department Of Boost]

What I'm getting at is packaging. IMO a centri or turbo car allows for more intercooler, especially without breaking the bank.

A A2A system can't be compared to a A2W system by "size". Too many other factors.

The reason you see lower cruising/normal driving out of turbo/centri setups is that they make boost less often. Even at WOT a PD blower makes more boost on average than a turbo/centri. More boost = more heat.

 

[Bobby]

i was refering to a ts making less heat then a turbo. it is all about how efficient and placement of the turbos on how well it cools. thats all i was saying. i under stand you heat up air when it is compressed no argument there, i just dont know where they are getting the heat to boost

 

[one eyed willy]

A A2A system can't be compared to a A2W system by "size". Too many other factors.

The reason you see lower cruising/normal driving out of turbo/centri setups is that they make boost less often. Even at WOT a PD blower makes more boost on average than a turbo/centri. More boost = more heat.

I think the turbo has the advantage of getting to cool down and stay cool most of the time. I'm usually only a few degrees over ambient when cruising so when i do give it the boost, my temps go up of course but they are never more then 20-25 above ambient before they level out, that i have seen so far and as soon as i let out it cools right back down. My friends with blowers are doing all kinds of crazy stuff with tanks and pumps and ice just to keep their stuff from pulling timing. that big chuck of metal sitting up there doing the spinny shit loves to soak up the heat.

@15.5 psi, 971 maf counts, I made a pull in 60-65* ambient temps and I hit 84 degrees on my log. when i pushed the go pedal it was 66* IAT. Within 10 seconds i already dropped back down to 70*.

 

[hamish]

Boost is boost. It's physics and thermodynamics. It's like gravity, there is no cheating it.

10psi makes XXX amount of heat. 20psi makes double that. 30psi makes triple.

"When air at atmospheric pressure is mechanically compressed by a compressor, the transformation of air at 1 bar (atmospheric pressure) into air at higher pressure (up to 414 bar) is determined by the laws of thermodynamics. They state that an increase in pressure equals a rise in heat and compressing air creates a proportional increase in heat. Boyle's law explains that if a volume of a gas (air) halves during compression, then the pressure is doubled. Charles' law states that the volume of a gas changes in direct proportion to the temperature. These laws explain that pressure, volume and temperature are proportional, change one variable and one or two of the others will also change, according to this equation"

Link:
http://www.compair.com/About_Us/Compressed_Air_Explained--02What_happens_when_we_compress_air-q-.aspx

There will always be the variable of air composition, humidity and other variables.
Most Industrial air compressors use some sort of moisture controlled system to provide a repeatable moisture content/dewpoint.
But I get what you are saying.

 

[Department Of Boost]

i was refering to a ts making less heat then a turbo. it is all about how efficient and placement of the turbos on how well it cools. thats all i was saying. i under stand you heat up air when it is compressed no argument there, i just dont know where they are getting the heat to boost

A perfectly sized and placed turbo is less heat efficient then a perfectly sized TS at WOT. Fact.

 

[Department Of Boost]

I think the turbo has the advantage of getting to cool down and stay cool most of the time. I'm usually only a few degrees over ambient when cruising so when i do give it the boost, my temps go up of course but they are never more then 20-25 above ambient before they level out, that i have seen so far and as soon as i let out it cools right back down. My friends with blowers are doing all kinds of crazy stuff with tanks and pumps and ice just to keep their stuff from pulling timing. that big chuck of metal sitting up there doing the spinny shit loves to soak up the heat.

@15.5 psi, 971 maf counts, I made a pull in 60-65* ambient temps and I hit 84 degrees on my log. when i pushed the go pedal it was 66* IAT. Within 10 seconds i already dropped back down to 70*.

But, have you looked at your logs? Lito outlined above that your real time IAT's are probably not actual.