Interesting EGR issue

[sntrym95]

So I'm still trying to figure out why the engine light comes on and throws a code 32 on my 95 Yukon (stock 350, automatic). I replaced the EGR valve and solenoid, and cleaned the passages on the intake when I replace the heads (with TBI heads).

The light will sometimes illuminate when both of these apply:

1. Starting the engine when cold (not freezing, just not running for a while).
2. Driving to the freeway (<2 miles), then setting cruise at 60 mph.

The light will never illuminate when both of these apply:

1. Starting the engine when cold (not freezing, just not running for a while).
2. Driving to the freeway (<2 miles), then setting cruise at 65 mph or greater.

I have gone through all the items here: http://www.chevytalk.org/fusionbb/sh...1a4c04da3ab1a/

Thoughts? Possible vacuum issue at 60 mph rpm?

[ChevyTech]

This one has me thinking, but I don't have a theory.

[sntrym95]

I'm glad I'm not alone in being stumped. My first thought was TPS but it doesn't appear that the ECM uses it to calculate the data, hence why I went towards the vacuum issue at around 1750 rpm.

[ChevyTech]

HERE IS THE CRITERIA THAT MUST BE MET FOR AN EGR SYSTEM TEST

Here is the info for a 1993 C/K TBI 5.0, 5.7 and 4.3L (s/t)
The actual test that sets the code 32 is done when:
-Throttle position is between 10% and 25% angle
-Steady throttle with no more then 2% change in throttle
-Above 32 MPH closed loop operation
-Max short term fuel trim less then 6 counts for 3.2 seconds
-desired EGR command greater then 40%
-MAP above 47.5 kPa or less then 15 kPa.

[ChevyTech]

Does your truck have a stock exhaust system on it or have you put something on it less restrictive to improve flow?

[sntrym95]

The Yukon has an aftermarket exhaust. Gibson stainless shorty headers, Magnaflow stainless Y-pipe and converter, and Gibson cat-back stainless 3".

Here's what make this more interesting. I have almost the exact same exhaust on my truck (which does not have this issue, and has the exact same drivetrain). The only difference is the truck has Edelbrock stainless shorty headers. Otherwise the Y-pipe and cat-back are the exact same parts from Gibson, save for the Yukon's is a tad shorter.

[sntrym95]

I've been thinking about hooking up the computer to the Yukon and running WinALDL to see if I can find something in there when/if a code is set

[ChevyTech]

WinALDL will not work on a 1995 with an automatic because it has the faster baud rate PCM instead of the slower baud rate ECM.

TunerPro will work.

[ChevyTech]

The system is not seeing a bid enough change when it cuts the EGR operation during a self test.

The GM literature warns of exhaust system modification affecting EGR flow but you don't hear a lot of people having trouble after they modify their exhaust.

How much different is the oxygen sensor position between the two trucks?

You might try a different oxygen sensor to see if it takes care of the problem. The sensor could be getting a little lazy. I suggest you stay away from Bosch brand oxygen sensors.

[sntrym95]

Sorry, I meant Datamster 0D. I've used it many times and it records data nicely.

I'll have to look at the O2 sensor position. Despite the P/N being the same, there are some differences in the cat design, specifically the angle at which it's welded. The truck's exhaust is much older than the Yukon's. I can tell you they are in roughly the same spot, although there may be an inch or two difference. I'll check in a little bit.

Both trucks have the exact same sensor - 3 wire made by NTK. The one on the Yukon is brand new.

[sntrym95]

So I suspected earlier that the TPS might be causing this issue, but I didn't just want to throw parts at it. Yesterday, the TPS confirmed my suspicions by not only causing the Yukon to run like crap over the last few days (fluctuating idle, hard shifts, not shifting at the right points, if at all), it finally threw a code 22. New part going in today. Perhaps that will clear everything up. I'll keep you posted.

[sntrym95]

TPS replaced, running normal again. Will check freeway from cold once it cools down from the test ride. Hopefully it's two birds with one stone.

[ChevyTech]

I am still keeping an eye on this thread so let us know how it goes.

[sntrym95]

So far so good, but I'll give it a week or so before I confirm anything.

[sntrym95]

No light since TPS change. I'm blaming it on a faulty TPS (that did not throw a code for months). Case closed for now.

Thanks as always for the replies, ChevyTech.

[ChevyTech]

You're welcome!

I will try to keep this in my memory of strange things that occur with these systems.

Thanks for the updates.

[sntrym95]

Spoke too soon. SES light came on again today. Had been driving for quite some time (well over an hour), so engine was warm. Got on freeway, set cruise at 64. Big headwind here today, plus was following box truck, so lots of wind resistance. Light came on, accelerated to 67 or so, light stayed on for a couple minutes, then went off. Stayed off rest of drive (~20 minutes or so). Code 32. I'm stumped.

MAP sensor maybe?

I really don't want to throw parts or $ at this since I just picked up a Vortec motor to drop in it. Trying to keep the $ for the swap.

One thing I did not mention before is this does have a throttle body spacer on the intake, but stock PCM. Would that matter?

[ChevyTech]

I can't think of a reason the throttle spacer could cause this problem.

Even if the MAP were stuck at a specific value, cutting the EGR should still be seen by the system.

Cutting the EGR is not making a big enough difference to meet criteria.

If you have a vacuum supply problem to the EGR, like a leak in the hoses or a solenoid that bleeds off some vacuum that could cause the code. You replace the solenoid and valve so I would look at the hoses. You could put a vacuum gauge on the supply hose to the solenoid and make sure it has a good supply.

If it were mine I would try swapping the oxygen sensor. It is the part that should be seeing the change. It could have got contaminated which is not very common anymore but in the early days many silicone sealer products were not oxygen sensor safe. If they were used anywhere on the engine it could contaminate the oxygen sensor.

[sntrym95]

I can't think of anywhere that sealer was used when I put the heads back on, except RTV on the block and threads. There should be nothing like that in the air passages, combustion chambers, or exhaust. I'll look at the hoses first, they are original and may not be up to snuff.

Thanks!

[ChevyTech]

In the early 1980s we could wipe out an oxygen sensor by resealing a rocker cover. The PCV would suck the fumes in, and they get burnt. Intake, oil pan, rocker cover ... all had do have O2 sensor safe sealant.

We used up the other sealers on non engine repairs, or older vehicles. The orange GMS we had was a problem when used on engines with an oxygen sensor.

[sntrym95]

The issue may have been occurring when I bought truck out of Texas earlier this year. Since one of the header bolts was broken in the head, I never drove it far before I tore the intake and heads off. After I put another set of TBI heads on, I was able to drive it normally. That's when I noticed the SES light come on for code 32. The old EGR was pretty cruddy, and the intake passage was clogged up. Cleaned all that up and put a new EGR on - same problem. Replaced solenoid just because - same issue. That's the back story.

I'm going to replace the vacuum hoses because it's easy and cheap. If that doesn't clear it up, I'll put another new O2 sensor in when I replace the motor with the Vortec one I picked up today, since I'll have the exhaust apart anyway.

[sntrym95]

Didn't get to the vacuum hoses yet, but I had a thought since you mentioned the O2 sensor. It's a 3 wire, so it's heated. Even though it's brand new, maybe the heater element isn't working all the time?

[ChevyTech]

The three wire oxygen sensor heater and sensor share a ground. I never thought that was a good idea. I prefer the 4 wire heated sensors.

If the heating element did not work it would take longer to get to closed loop and you might notice that, but if it is intermittent, or works poorly, it would be hard to know that, or prove it.