for awhile now my engine "idiot light" comes on while i'm driving but only stays on for a few minutes or so then turns off.... So i did a self diagnostic and it seems to be (code # 11) my intake air thermosensor. can anyone give me some info on this piece? Is my car taking in too much heat? I do have a greddy intake.... not really a cold intake, could that be the problem? or do i have a bad sensor? I think the sensor is located under the exhaust manifold (correct me if i'm wrong). Anyways, what should be my next step? should i be worried at all? thanks for all ur help in advace
-Kevin

 

First make sure its connected securely.

2nd. IF its not connected in the manifold, make sure its grounded correctly.

3rd. To take the actual sensor out, its best to take off the manifold and then just unscrew it from the manifold. Could be that its come slightly unscrewed and it needs to be tightened.

Either way, id be real careful on doing heavy driving with a mal functioning IAS.

-E

 

It's located under the Upper Intake Manifold (UIM)

 

Get it out of there while you're there.

1. Plug hole on UIM with 10 x 1.25 pitch bolt (shortened to length and use teflon tape).

2. Relocate new or current IAS to IC outlet tank or extension elbow by drilling and tapping a 10 x 1.25 pitched hole. Also use teflon tape.

3. Cut off factory harness, extend wiring with additional 18 gauge, solder, and heat-shrink.

 

Have you noticed any differences in how the motor performs after relocating the IAS?

 

i too have done this and love it. no more heatsoaked sensor problems. much quicker reactions to air temp changes. not going to give you huge gains in performance, but definately worth doing if you have a lot of mods.

 

clayne,
I'm with Kento, can you notice any real change in engine operation, hot starting, etc. following the relocation?

 

with the sensor located in the elbow it gets more accurate readings as the the air actually coming into the motor. with it under the manifold, it reads a little warmer than the ambient air, causing the ecu to make false corrections

 

anyone know of a "how-to" procedure page for the relocation>? if i am to do this, which it seems like a good idea, i would like to do it right
-kevin

 

Cool, I was looking for this exact information.

 

You relocating guys should mention you're using PFC's, if you are. Im not sure if it's a good idea to do with a stock ECU, since it's mapped with the stock location, heat soak and all, in mind.

 

ok, well since i have a stock ecu, i will not relocate lol. i hope its only bc my sensor wasn't connected properly...... if its not that what else could it be?

 

Not relocating it could blow your engine.

Relocating it will NOT blow your engine.

Situation 1:

a. Heat-soaked IAT sensor providing 60C false intake temps. to ECU.
b. Post-IC charged air temps. actually 35C.

Cold air is more dense than hot air. Fuel and temperature correction maps inside aftermarket and stock ECU take this into account. ECU provides fuel for 65C instead of 35C actual IAT. ECU is effectively providing not enough fuel for the actual intake charge temps. as it thinks the air is hotter and less dense than it actually is -> lean condition -> possible detonation.

Situation 2:

a. Non heat-soaked IAT sensor providing 20C intake temps. to ECU.
b. Post-IC charged air temps. actually 20C.
c. By some massive bizarre hypothetical situation, someone routes your intake airpath through a refrigeration core, sending intake temps. up to 70C.

Hot air is less dense than cold air. Fuel and temperature correction maps inside aftermarket and stock ECU take this into account. ECU provides fuel for 70C instead of 20C actual IAT. ECU is effectively providing too much fuel for the actual intake charge temps. as it thinks the air is colder and denser than it actually is -> rich condition -> no detonation, maybe some backfiring and temporary low power.

As you can see, Situation 1, which is present in stock configurations from the beginning has a lot worse possible outcomes than Situation 2 (relocated sensor).

Not only is rate of change on the sensor increased with it relocated, it could save your engine.

 

That's a very good point Gid.. I have a PFC.

 

dub

 

Clayne I hate to argue with you, but never once have I seen a blown motor with the cause being a heat-soaked IAT. There are 6 and 700 hp cars with the sensor in the stock location, I think it works just fine where it is.

Every bit of software written for the FD was written with the sensor in the stock location, and hundreds of people are using it without and issue. Moving it is just a useless mod.

 

My son had a bad thermosensor in his FD. It was getting about 8 miles to the gallon and pushing a bunch of unburnt fuel into the oil and out the exhaust. I would get it fixed first than worry about relocating it.
BTW - first we did try to tighten it first(as recomended above), but we ended up replacing it. THe part is around $75 if I remember correctly.

 

"oh no" sounds like mine.... my FD leaves little oil/fuel specks on the drivway floor right near the exhaust (spits it out while its warming up) ......... does this sound related?

 

Well let's talk about cars that are tuned only on an average scale and not up to the minute. While we don't have any statistics, since they're hard to come by, I wouldn't be surprised to find plenty of regularly modded 13Bs being blown due to inaccurate sensor output. EFI system, might as well take advantage of it and make it as accurate and precise as possible. Someday I'll have a MoTec .

600-700 hp engines are not typically seen sitting in traffic. The owners also watch their IAT like a hawk, as with any other engine reading. I know when I see above 50C in my far from 600 hp car, I tend to take it easy.

Regardless, it doesn't *hurt* and it's sort of a design "flaw" in a way. At the minimum you will see greater rate of change from the sensor and accuracy.

 

relocation doesn't matter on the ecu. pfc has similar temp corrections as the stock ecu, unless you change them. it could potentially damage the motor from heatsoak, that's why i moved it. that and you have quicker fuel adjustment if the air temp does change abruptly. i can just monitor it with the pfc, which i do

 

spoolin,

Exactly, ECU doesn't matter. If the ECU thinks the intake temp is 70C, but the actual temp is 25C, that's never good - PFC, Pettit, M2, Haltech, Mazda, whoever, they all use an IAT of some sort for map correction. You can't sense heat-soak without extensive ECU logic that requires parameters we don't even have access to.

If I stop my car, come back 10 minutes later, start it back up, IAT now showing 70C due to manifold heat soak, ambient air actually around 25C, and start driving around the block - I guarantee you my engine is inhaling 25C (+compressor charge temps) regardless of what the ECU thinks. Are you going to boost to 10 psi with the ECU thinking it's in death valley when you really aren't? Now with a PFC I can see my IAT and I know when it's bunk or not, so technically I don't *have* to relocate the sensor as I can adjust my driving style to suit "IAT bunk factor." But I'd like higher rate of change and external factors to not be an issue. People with stock ECUs should definitely do it as they have no visibility of sensor output.

 

Probably not. That's just condensation from inside the motor getting spit out the exhaust as your engine warms up. Same reason that you see light puffs of steam as your car warms up in the morning.

 

Clayne,

First of all: Damn!

Second of all:
This is great if you're running an aftermarket ECU that can be tuned to "trust" the sensor in a better location.
Mind that I do NOT have an aftermarket ECU when I say this.

As for the stock ECU: True, other stock ECUs I've worked with give a richer mixture for colder air. Other car's I've worked on put the air temp sensor way the hell out in the beginning of the intake tract, just behind the air filter. If you're adding FI to an NA motor with that config, you can patch a boost-tripped switched resistor in the sensor's circuit to fool the ecu into giving more fuel.

However, the ECU may also advance the timing as well for colder air. I don't know what the stock FD ECU does regarding timing... do you, or anyone else know for SURE?

One thing that alarms me about our stock ECUs is the existance of the fuel temp sensor. The usually non-descript FSM says it's part of the Pressure Regulator Control system (to overcome vapor lock). Other OBD-I EFI systems I've seen have nothing of the sort.
Why do we have one? It could be because mazda is paranoid about flooding our engines by unnecessarily increasing fuel pressure.
Why don't they use the intake air thermosensor to estimate fuel temp? The fuel temp should be highly correlated to air temp.

Almost like they dont "trust" the air temp sensor...

The ECU already has data from water temp, fuel temp, and heatsoaked air-temp. Whose to say the ECU isn't performing any correlational analysis-based rate-of-change prediction (based on where the sensor is, stock)?

I'm not saying you're wrong, just playing devil's advocate here. You COULD be right and the ECU is a total dumbass, which nobody, including me would be suprised to know. But why bother with a mod with questionable gains on a stock ECU?

 

I look at a IAT as a input which might "trim" or fine tune a mixture but have clear boundaries as to how much it can add or subtract. I would think a max of 5% of the active mixture/timing map. Just a educated guess, not something I would quote. John

 

A vane-type AFM system can estimate the air flow (which accounts for temp) directly. However, they do use intake temps to anticipate which map to use when the vane is swinging due to the throttle snapping open. The popular boost-switched-resistor mod tends to help throttle response when aftermarket FI is added. I've heard of some getting backfiring after doing this mod.

However, the stock FD ECU is a speed-density system. Air flow is derived from pressure, engine speed (the flow resistance of the manifold is taken as a constant "mapped" VE), and temperature. It's much more reliant on temperature than the above system. How much more, who knows? Maybe it is 5%...