OK, I have a question I'm hoping some of you guys with lots more engine tuning experience than I can help me with:

I recently read a somewhat technical/how-to article on relocating the intake air temp sensor off the intake manifold to a location upstream in order to isolate it from potential heat-soak from the manifold. The hope being that it would get a more true reading of actual intake air temp. Furthermore, the article provided a very basic chart of resistance values that the IAT sensor would report based on temperature in degrees F. It gave basic instructions on how to either: 1. replace the IAT sensor with a resistor altogether (engine would always think it was a specific temp outside, would require different resistor for every potential situation (not good). 2. Add a resistor in-line with the IAT sensor, fooling the computer into thinking that the the outside air was cooler than it actually is, thus causing the engine to increase the amount of fuel to the engine (I guess through increasing the duration of the injector pulse) and advancing timing, theoretically producing more horsepower. The specific resistor mentioned was a 10k ohm 1/4 watt resistor which is supposed to report an approximately 20 degree colder resistance value than the IAT alone. And finally, 3. adding a variable resistor in line in order to acheive tunability with the same expected results as #2 above.

Is this a bunch of crap or is this a known trick? I have heard of relocating the IAT sensor and actually had performed that mod on a previous vehicle, but never tried to alter the resistance value reported by the IAT sensor. I know, obviously one would prefer to alter timing and mixture though modification of the ECU's program. In my particular car (86 Integra with D16A1 and stock computer) I don't think there's any way for me to effect that sort of change to the ECU and I haven't seen any sort of modified PROM that will work with my computer (vacuum advance car).

To be specific about my car and mods, here's my set-up and it's like this because it is a IT road race car and modifications are very limited by the rules:

86 Integra LS w/ 1600cc D16A1, blueprinted & balanced
custom cold air intake w/ K&N
Lightspeed 4-2-1 header feeding 2.5" pipe with 12" Dynomax muffler, exiting in front of rear wheel
Underdrive pulleys
stock injectors
stock pressure regulator and stock fuel pump

I'm curious if there's any merit to any of this as I'm relatively underpowered compared to most of the competition and even 2 or 3 extra HP would be appreciated. I might have access to a dyno in the near future and was just wondering if this is something that is even worth experimenting with.

Anybody got any knowledge or advice?

Thanks in advance....

 

Adding a resistor in series with the IAT sensor would trick the ECU into enriching the fuel mixture. 10K ohms is too high though, I think. I don't recall the normal resistance range of the IAT sensor. If you can find that data, a 10% increase should be a safe starting point though.

 

The resistance chart from the article:

Temp (F) Resistance value (ohms)
32 - 100K
36 - 81K
57 - 47K
86 - 16K

These are supposedly the values reported to the ECU by the IAT sensor.

Also stating that a 10K ohm resistor inline with the IAT sensor equated to approx a perceived 20 degree drop from the actual temp.

It would seem to me that the computer richening the mixture thinking that the intake air was 20 degrees cooler, you'd wind up so rich that much of the fuel would go unburned and wind up expelled down the exhaust. Maybe I'm wrong? The only way I can see a benefit is if the original fuel curves are set up with economy in mind, not performance, and are so leaned out for fuel efficiency's sake that a small difference in fuel and advance would make a big difference. Are we onto something?


[Modified by 1gTeg90, 12:58 AM 8/30/2002]

 

um, why move the sensor up stream?

it is there so the ECU can actually compensate for the heatsoaked manifold???!!!

the IAT sensor is such a minimal A/F trimer though anyways - 0.1ms to 0.2ms of the injector base pulse width,

whereas the ECT sensor is more relevant - upto 4.0ms of the injector base pulse width...

if you wanted to play with hte mixtures you could consider the ECT sensor a better place to start.

but hey, why not look at adjusting the MAP sensor input if thats what you are trying to achieve?!?!

this signal has most to do with injector pulse width and A/F ratios than all hte other sensors combined (in most circumstances...)

HTH,

t..

 

Not sure if I follow you... You're saying the ECU can actually figure out from one basic resistance value reported by the IAT sensor how much of that reading is the real temp of the incoming air and how much is heat soak that has artificially elevated the temp of the sensor by soaking through the body of the sensor itself? Not being a wise ***, I just don't know the fine points of how the fuel curves are derived that the ECU runs off.. Maybe the guys who thought all this up said "hey, if the ECT says the engine is operating at say, 210 F, and we get a IAT reading equating to 97 F, we know that the intake air is actually 89 F but that the IAT reading has been affected upwards by a heat soak factor of 1.xx." Maybe? Or is the fact that the IAT is located where it is a function of engineers intentionally allowing the heat to affect the sensor so that the mixture will lean out slightly at operating temp, increasing fuel economy? That's what I'm curious about...

That's the kinda info I'm looking for...

How would one go about such a thing? Does a higher resistance value indicate a lower temp, as with the IAT sensor? And how would one richen the mixture with this method? Increasing the resistance, telling the ECU that the engine is not as hot as it actually is? Curious...

Same question as above, I guess...

Humor me, as this is all really vague to me (as you can tell ) and I'm trying to gain some insight on the fine points....

Thanks

 

Humor you?

nah, you are trying to build your own Apexi VAFC...

too hard basket for me.

sorry,

t..

PS do a search on "closed loop" and you will find out why the ECU does not run off a set table of values, and hence is really alot more able to decide its own mixtures.

PPS not being a wise *** but you have really started this backwrds, working from hte least important sensor.

PPPS there is no way for you to adjust the sensor you talk of without using a wideband O2 sensor and a Dyno logging eahc change and monitoring its effect on hte A/F ratio.

PPPPS send me a PM with your email address and i will mail you some good info for reading...

 

PPPPPS the temprature of hte air determines its density and hence it should be measured closest to where it will enter the combustion chambers...

 

gotcha. At any rate, that's the kinda info I was looking for. It probably won't effect enough change to make it worth the trouble.

Thanks

 

This sounds like the jrsc system. Not a good idea if u ask me.

 

EXACTLY!

Any hack that modifies that IAT or ECT is not a good idea. Any fuel modifier out there (with the exception of the ghey JRSC setup), modifes the MAP voltage. Do not play with temperature sensors.

The max resistance of the IAT sensor is 15k. I'm not sure where that guy got his data from.

Sonny

 

I was actually musing with a tech buddy this past week about putting
another IAT sensor in the fliter neck, extending the wires and plugging
it in. He's got several in the shop, it would be interesting to dyno the
difference with the 5 gas and find out whats happening

 

So puting a resistor on to trick the ecu into lowering the air temperature is not good? I just have mine in my cai tube. I'm not going to do this, but...if you took the tip of the IAT sensor and put it in a ziploc bag with dry ice in it and drove around, besides probably getting really crappy gas mileage and maybe too much fuel increase, could it possibly give you some more ponies?Just curious. Thanks.

 

Basically, Tinkerbell hit it right on the head. Air temp sensor has very little authority on fuel mixture. Plus it will only affect "open loop" mixture. Most Hondas are probably set up a little on the rich side anyway (in open loop) and putting more fuel in the motor is apt to slow you down. Need a dyno/wideband or your wasting your time....3gdude