Write-Up: Omega Intake Air Temperature Sensor (IAT)

Edit: this was originally posted in the AC Compressor clutch thread, before the P0113 code was detected.

I'm having an odd symptom with my AC. This started all of a sudden today, which happened to be a really hot one (94º). This morning I was driving and the AC worked as usual. About 10 mins into the trip, I notice the air smells musty and it is not as cool. Soon it is blowing warm, damp air as if the vent were on.

I went home and looked under the hood. The AC clutch appeared not to be engaged. Switched the AC off and on - no change. Turned the car off and then back on, the clutch grabs and I have cold air again. Thinking it was a fluke, I started on another trip this evening and the AC works fine. There there was a moment when the air smelled musty, but it recovered quickly. It would appear the clutch is disengaging at random times on its own.

My first thought is a wiring or connector issue. I plan to wire a lamp at the clutch connectors to test voltage next time it happens. But in case it turns out I'm getting voltage, does this sound typical of how the clutch would start to fail?

AA wrote:

I'm having an odd symptom with my AC. This started all of a sudden today, which happened to be a really hot one (94º). This morning I was driving and the AC worked as usual. About 10 mins into the trip, I notice the air smells musty and it is not as cool. Soon it is blowing warm, damp air as if the vent were on.

I went home and looked under the hood. The AC clutch appeared not to be engaged. Switched the AC off and on - no change. Turned the car off and then back on, the clutch grabs and I have cold air again. Thinking it was a fluke, I started on another trip this evening and the AC works fine. There there was a moment when the air smelled musty, but it recovered quickly. It would appear the clutch is disengaging at random times on its own.

My first thought is a wiring or connector issue. I plan to wire a lamp at the clutch connectors to test voltage next time it happens. But in case it turns out I'm getting voltage, does this sound typical of how the clutch would start to fail?

Very quickly -

- wonder if you have a freon leak. Thermal expansion/contraction sometimes losens fittings enough so they leak, over time. With a slow leak the refrigerant level will get marginal and cuse to lo pressure cutoff switch to wink the AC off; when pressure recovers it winks back on.

An AC shop can find such a leak very quickly with a "sniffer," tighten the connections, evac and recharge and you'll be on your way.

Albertj

UPDATE: AC has been running perfectly for 3 days. Weird.

UPDATE II: I found the cause of the AC issue. I think my IAT sensor failed at about the same time the AC problem started, but I was I unaware of it. I had recently pulled the IAT sensor for cleaning, and it didn't look so good to be honest, but I cleaned it with CRC spray and reinstalled. Possibly I damaged it when reinstalling, or it finally just gave up - for many months I've noticed it was reading colder and colder below ambient. A few days after the AC episode I got the SES light for P0113 code (IAT Sensor Circuit - High Voltage), confirmed with the scanner (IAT now reads -38ºF, the value for 0 ohms).

A new Omega thermistor is on the way. I'm going to build a new (better) IAT sensor.

Glad that's over with!

AA wrote:

UPDATE: AC has been running perfectly for 3 days. Weird.

UPDATE II: I found the cause of the AC issue. I think my IAT sensor failed at about the same time the AC problem started, but I was I unaware of it. I had recently pulled the IAT sensor for cleaning, and it didn't look so good to be honest, but I cleaned it with CRC spray and reinstalled. Possibly I damaged it when reinstalling, or it finally just gave up - for many months I've noticed it was reading colder and colder below ambient. A few days after the AC episode I got the SES light for P0113 code (IAT Sensor Circuit - High Voltage), confirmed with the scanner (IAT now reads -38ºF, the value for 0 ohms).

A new Omega thermistor is on the way. I'm going to build a new (better) IAT sensor.

Glad that's over with!

you mind posting pics/writeup of the rebuild of the IAT when you do it?!?

Your IAT being shorted - which CRC spray did you use?

Do tell, do tell.

IAT wasn't shorted, the circuit is open (sorry I said "0 ohms", meant open circuit). I think a solder joint gave out in the sensor. Like I said, it's been showing a lower and lower temp value for a long time. That could be a sign of a weakened connection.

The CRC spray was electronic parts cleaner, same I always use for cleaning MAF and IAT. It wasn't the spray. The sensor was old and just failed. I think when I torqued on it to remove and install, it helped to give up the ghost.

Yest, write-up, no problem.

AA wrote:

IAT wasn't shorted, the circuit is open (sorry I said "0 ohms", meant open circuit). I think a solder joint gave out in the sensor. Like I said, it's been showing a lower and lower temp value for a long time. That could be a sign of a weakened connection.

The CRC spray was electronic parts cleaner, same I always use for cleaning MAF and IAT. It wasn't the spray. The sensor was old and just failed. I think when I torqued on it to remove and install, it helped to give up the ghost.

Yest, write-up, no problem.

The CRC electronic pats cleaner sprays good and cold and has CO2 propellant I think. A marginal solder joint would freeze, contract and crack through; weakened bonds within the thermistor could be a problem too.

If it was really in such a fragile state, then I'm glad it gave out. It's $15 and some minor fabrication. I'm all right with that.

AA wrote:

If it was really in such a fragile state, then I'm glad it gave out. It's $15 and some minor fabrication. I'm all right with that.

Yep... my guess is GM did not require a mil-spec thermistor from the OEMs in that application.

Having driven the car with a failed IAT sensor for about a week, the AC problem seemed to have worked itself out, but the SES lamp remained illuminated. I scanned and noticed another DTC in addition to the previous one: P0111 - IAT Sensor Circuit Intermittent High Voltage. I assume this is directly related to the failed sensor as well.

Driving the car in this state probably won't hurt anything, but performance will be sacrificed, as the timing advance will be thrown off. I assume the PCM knows the sensor is bad and so probably defaults to a set value for such situations. Think of it as a "limp" timing advance for safe operation but not optimum.

Yesterday I removed the IAT sensor and measured its resistance at several mega-ohms (open circuit), which is far above the correct range of operation. This indicates the sensor is in fact bad, so it is likely not wiring or connection related.

Since my replacement part is on the way, and I intent to drive the car in the meantime, I decided on a temporary fix. The idea is to use a resistor to simulate the intake air temp sensor, which is the same idea employed by those "chips" that wire inline with the IAT sensor, claiming to increase HP. All they do is add slight resistance to the existing IAT sensor, tricking the PCM into advancing timing a certain amount. In theory, this can increase HP, but it also increases the tendency to knock.

A somewhat better solution is called a "happy knob", or a variable resistor that allows the driver to adjust timing to any amount they choose. It can be wired inline with the IAT sensor, or it can replace the IAT sensor altogether. Obviously, you must have a way to monitor timing advance and KR levels, so you can fine tune the timing, keeping knock at a safe level. This method is more appealing to racers and tuners who really want control over the engine, but it might not be good for someone just looking for a HP boost, and there is potential for pushing the timing advance too hot.

So I made a temporary happy knob to substitute for the IAT sensor until I get the new one in. I used a small potentiometer trimmer to set the value, but you could also use a fixed resistor if you have the correct value. For our cars, the resistance value of the IAT sensor is 2,796 ohms @25*C (77*F) according to this link I found in a GTO forum. Here is the trimmer I used, a 25k ohm unit from Radio Shack. I think I paid $2 for a pack of 10 assorted pieces:



Using a small screw driver to adjust, I achieved a value close to the stock reference spec:



The trimmer's pins are conveniently sized and spaced to plug into the stock connector:



Confirming the value by scanning IAT in AutoTap (it's very close!):



Just a little electrical tape and then tie or clip the connector out of the way:



This set-up may not provide the most optimum timing for the PCM, but it should give an acceptable value, which should cure the SES lamp. To be continued when the new thermistor arrives...

The thermistor arrived today. I took some pics for the write-up, but first a little about the modification.

A couple of years ago I noticed the guys at the GTO board were promoting a little mod using a precision thermistor to replace the sensing element in the OEM intake air temp sensor. The idea is that the OEM sensor is slow reacting, and not in an ideal location to read actual intake air temps at the air filter. The problem is, a slowish or inaccurate IAT sensor will not command ideal ignition timing from the PCM, which bases the value on the IAT reading.

One example of when this could cause a loss of performance would be on a hot day, when the car is sitting idle. Intake air is quite warm around the car, to which the PCM responds my retarding timing a bit. This can be significant when the car is stopped for a minute or two. If the car is launched suddenly, the intake air temp will drop quickly as the car pushes forward into fresher, cooler air. This should allow timing to advance, but only if the PCM knows the incoming air is colder. If the OEM IAT sensor is slow to react, timing will remain retarded until the PCM catches up. In theory, a quicker sensor will let the engine adjust timing sooner, restoring power in shorter time.

Also, heat soak from the engine bay can cause the IAT reading to be higher than the air flowing through the intake, because the air box warms the sensor to some degree. The solution: mount the IAT sensor as far away from the engine, and as close to the air filter as possible. The GTO folks seem to prefer snaking the sensor wire into the cold air intake and laying it right on the filter, not physically mounting it to anything.

How much does all of this really matter? It really depends on who you are. If you run an open cone straight off the TB, this mod won't do you much good. But if you're the type who likes to log scans and compare numbers, or you're the person who believes the little things add up to big results, this $20 mod and 1/2 hour install time is a no-brainer. A company called Vector Motorsports sells this modification plug & play for $99. Vector explains:

"Precious timing is being pulled due to false temperature readings which not only slows your vehicle down but can also cause a slight decrease in fuel economy. Couple this with the fact that all OEM GM IAT sensors react pretty slowly to temperature changes and you've got a real mess on your hands!

...we've done away with those slow GM OEM temperature sensors in favor of one that we designed that responds nearly ten times faster! This has proven to not only make vehicles more consistent but also obviously faster due to the lack of timing retard being pulled from false readings.

One user with the LS2 version of this kit tested it back to back on a dyno. He recovered 8 RWHP with the kit plugged in verses not!"

It's likely the Vector kit uses a thermistor (or similar to) manufactured by Omega Engineering. It can be purchased online for about $20 shipped and install yourself with a little soldering. The GTO board has a write-up here:

http://kb.ls1gto.com/KB/article.aspx?id=10279

If you want to upgrade your OEM sensor, the GTO write-up is pretty straightforward. If you use a FWI tube with threaded IAT sensor, see below for pics of my install.

Order the 44005 thermistor here: http://www.omega.com/ppt/pptsc.asp?ref=44000_THERMIS_ELEMENTS&Nav=temd11

The product took about 4 days to arrive. One thing I really like about USA-made products: the paperwork contains genuinely useful info, written in plain English. How refreshing. The paper contains specs, including a tolerance graph, typical reaction times, and a chart on the back with predicted resistance values at every temperature in the range. It's simple, but very complete and thoughtfully prepared:



My old IAT sensor, literally falling apart as I cut off the plastic tube with a razor knife. Note: this is not a stock sensor. It was an aftermarket used to build my custom FWI:



There was just enough clean lead to solder in the new thermistor. Here I'm attaching the heat sink and aligning the leads. I also cleaned up the old leads by scraping them with a blade:



Here it is with both leads soldered in:



I folded a piece of electrical tape around one lead to prevent contact:



Roll the tape around both leads and slip over a proper length of heat shrink tubing. A quick check shows the sensor is right on the money (it happened to be about 75º ambient):



Here it is heat sealed and ready for install. Note the long length will place the thermistor head directly in the center of the intake air stream, and isolate it from heat soak. This should give the most accurate and current IAT readings:



The new sensor is installed, but I haven't tested it on the car yet, and I don't plan to compare to stock numbers. I remember the OEM sensor was slow, taking several seconds to catch up to the intake air temp. The previous replacement sensor was faster, and I suspect the new Omega will be at least as good, or better. For $20, can't beat it.

Update: I've been monitoring the Omega sensor for the past week. IAT reading is almost spot on (-2/ 0 deg F) with ambient temp when engine is off and cold (this is a good test). While driving, the FWI matches ambient or drops slightly below near 100% of the time. Reaction time is pretty quick, sometimes I see IAT change on the scanner within 5-10 secs of seeing a change on the climate control. Sometimes IAT changes on its own, regardless of ambient.

Interesting is the observed "wind chill" effect. I have been able to repeat under throttle application. This was surprising to me, and really shows how accurate and quick-reacting the Omega sensor is. When cruising at freeway speed, I usually see a consistent IAT value of between 1-3º below ambient, sometimes the value is equal, but almost never above ambient. When throttle is is zero, coasting on decel, IAT will climb 1-2 degrees within a few secs, sometimes in as few as 3 secs. When given moderate to hard throttle, IAT will drop back down to 2-3, even 4º below ambient. I have not been able to exceed -4º ambient with the Omega sensor. All test were performed in warm weather, between 65-90º ambient, with similar results within this range.

When stopped at idle, as expected IAT rapidly rises above ambient. Depending on duration spent idle, IAT will climb 10, 20 or more above ambient, due to engine heat accumulated around the car. As soon as the car moves forward, IAT will drop back to near ambient. A WOT launch results in much quicker IAT drops, as would be expected.

BUT...

The benefit of this mod may not be as significant as I had hoped it would. Looking at the PCM's timing table for IAT adjustment, there appears to be no timing advance for cooler IAT values. However there is some timing retard starting at 122ºF (-1º) and increasing to -3º at 176º. That seems rather high, but I've seen 140º+ intake air temp with my FWI on a 100ºF day with extended idle time. With a stock air box, I can see IAT easily moving above 150ºF in hot summer city driving. In this case, the Omega would help reduce timing retard between idle periods. In mild and colder months, IAT really won't give a timing boost.

However, using DHP's 1.5 BIN program, the timing table is mapped for a +1º spark advance at or below 32ºF, and up to a +2º advance at 14ºF and down, depending on MAF (Mg/cyl air flow). There is no timing retard at hotter temps.

The Omega sensor was an interesting experiment that I'm glad I tried. Also, it will be fun in colder months to try reprogramming the IAT spark table to use hotter timing as temps drop.

Reprogrammed the IAT timing table today. I like that the DHP 1.5 BIN has hotter spark for cooler temps, but I also like what the OEM did as a protective measure: retard timing for hotter intake temps. DHP removed this, keeping base timing even up to 176º. Since I've been pushing my base timing higher lately, I doubt I can get away with it above 122ºF, so I decided to blend the 1.5 BIN with the OEM to create a map that is optimized for about 70-85ºF intake temp.

Here is the stock map:



Here's the DHP 1.5 BIN:



And here's the new modified map:



You can see starting at 86ºF IAT a slight bit of timing will be pulled in the boost range. As IAT increases, the spark retard ramps up, even more aggressively than with the OEM map. Conversely, as IAT drops there will be some hotter timing added to in the non-boosted range, starting at 68º and working toward 2º as it gets colder. I didn't want to add more than +2º spark, as DHP seemed to use that as a limit.

The biggest advantage here is being able to run a hotter base timing table for good weather, and knowing there is some insurance when things get hot. The accuracy and response time of the Omega IAT sensor has enabled me to fine tune the IAT timing table for some functional results.