ASE doc

I thought I would post this just for informational purposes.

I have an '88 IROC, VIN code F(5.0L TPI), in the shop for a number of repairs. One customer concern was the CEL is illuminated. I checked codes and found codes 33 "MAF signal high", 36 "MAF burnoff failure", and 32 "EGR system failure". Codes 33 and 36 were both caused by an open MAF signal circuit. I repaired the MAF system but was left with the EGR system fault. This vehicle is equipped with the long discontinued EGR temp switch, otherwise known as the EGR diagnostic switch. This is a normally open temperature switch that is meant to close when hot, or in other words when the EGR is open. Code 33 will set if this switch is closed on engine start up or at any other time when the EGR should not function, or if it fails to change state when the engine is operating within the range where the EGR should function. In the case of this vehicle, the switch is shorted internally and always reads closed, even when fully cold.

Some GM vehicles of this period may be equipped with a diagnostic switch that is built into the EGR control solenoid and simply closes when vacumm is present on the EGR valve vacuum circuit. That system is identifiable by the three wire solenoid connector.

On the temp switch system, there are two basic criteria that the ECM is looking for to set code 32. First, the switch or circuit may be shorted, or the EGR may be stuck partially open and the switch read closed at all times or when the engine is at idle, full throttle or other times when the EGR should be closed . Second, the switch may be open and not ever read closed, or the EGR valve or control circuit may not be functional and the switch will not close. According to this basic rule, what we need to provide for this code not to set is a diagnostic switch signal that is open when the engine starts and all other times the EGR valve is closed, and that closes when the EGR valve is open. The EGR control solenoid is a 12 volt solenoid that is controlled either by a straight on/ off ground signal from the ECM, or by a pulse width modulated ground signal from the ECM. The diagnostic switch circuit is supplied 12 volts through a diagnostic resistor from the ECM. The switch is internally grounded and when it closes, it grounds the 12 volt signal, pulling it to about zero volts. The ECM sees this as the switch closing.

My idea was to use the EGR ground control signal as a grounding signal for the diagnostic switch circuit. My only concern was that this vehicle uses a pulse width modulated EGR solenoid. Would the pulse width modulated ground pull the 12 volt biased switch circuit close enough to zero volts for the ECM to read the switch "closed". I first verified that the EGR system was functional. The vacuum supply to the solenoid is good, the solenoid functions properly and the EGR valve(negative pressure modulated) worked as designed. The factory service manual has a very good trouble tree for the EGR system that covers this test sequence. Then, I experimented with my proposed fix for the switch signal using a test lead to connect the diag switch circuit to the EGR solenoid control side. I ran the engine until the EGR system became functional and worked the throttle to activate the EGR, while watching both EGR duty cycle and diag switch state on ECM data. Whalla! Whenever the EGR solenoid is commanded 50% or more, the switch state changes from off to on. I made the modification more permanent by splicing and soldering a jumper between the two circuits in the ECM harness. I am confident that code 32 will now stay gone.

I know that this fault code, or the test for it, can be deleted from MemCal using tuner pro, assuming that your ECM is programmable. Otherwise, you would need to burn or purchase a modified PROM. The above fix is something that you can do at home with no need to reprogram. I urge anyone looking to perform this fix to be sure and verify that your EGR does actually work. The EGR, injecting inert exhaust gas into the intake air stream, is important on these lean burning emmission controlled vehicles to prevent detonation or other means of engine damage due to cylinder overheating.

If your engine is modified for increased power, and you have the ability to tune your fuel and timing curves, that's a different story.

Happy motoring!

eldy

Sounds like you are smarter than the average bear,but for those of us mental midgets , a new REPO switch is available for the 1985 Corvette that worked for me. see my post on EGR Switch dated 06-18-2014

Damon

ASE Doc- that's freakin' awesome. I don't know how much traffic you're going to get in this thread, but somebody should mark this for archive somehow (I'd submit it to the board masters for inclusion in the Tech Articles section). There's gotta be half a million 3rd gens running around out there with this problem, even though nothing is mechanically wrong with the EGR valve.

Kudos. Well done. Home-grown hotrodding at it's best.

ASE doc

Thanks guys for the feedback. Eldy, It's neat to hear that someone is offering a replacement for the temp switch. I wasn't aware of that. Thank you for the info.

Warlocksirix

ok, been having this code #32 issue myself. Replaced about every sensor/valve/line in the front end and the only code to show now is code #32 while driving for about 15-20min on the highway at about 65. I replaced the EGR and it still throws the code under these circumstances. Is there any precaution or measures I should take to test my EGR valve is operational before I do this fix?

ASE doc

What the ECM is looking for is a "change of state" in the EGR temperature(diagnostic) switch. The switch should close pretty quickly when the EGR valve opens. First, make sure the valve is opening with vacuum applied. Next, be sure that the valve opens when the solenoid is activated by briefly grounding the control circuit. To be really sure, I also T in a vacuum gauge to the line from the solenoid to the EGR valve and watch for vacuum while I drive the car. Once you're really sure that the EGR is actually working when commanded by the ECM, tap into the temp switch circuit and look for 12 volts with the EGR closed, 0 volts with the EGR open. The most common failure of the EGR temp switch is failed open. Meaning that it will read 12 volts at all times. The last one I saw was shorted so that it read 0 volts at all times.

Do you have a compatible scan tool? Something like the Snap On MT2500 "brick" works fine($200 on ebay). I use my Snap On Modis Ultra with the early style 12 pin GM connector. The scan tool will let you see the EGR command from the ECM, and also the state of the EGR switch. It will read open or closed. The key is that the switch changes state when the EGR is commanded on. Code 32 will also set if the switch is closed when the engine starts, but if the switch works normally otherwise, the code will go away.

The GM factory service manual trouble tree is very helpful when diagnosing this code. The service manual has a trouble tree for every fault code. The tests are simple and will get you to the cause of the problem in the most straight forward manner. I think even the Haynes manual has these trouble trees. All Data, which you can access at most public libraries has them as well.

ASE doc

Okay, hold the phone. I just rechecked your vehicle info. Your car is a 91. Completely different animal. GM deleted the temp switch by that time and used manifold pressure (MAP) as a test reference for the EGR system. What the ECM is looking for is a rise in manifold absolute pressure, or a drop in vacuum, when the EGR is opened. What you need to do is be sure that the EGR is actually opening when you drive the car. The vacuum gauge T'd into the line from the solenoid to the valve is the sure way to know that the valve is getting vacuum as it should. At highway cruise, the valve should get at least 5" of vacuum. Use a vacuum pump at the valve and open it while the engine idles. The engine should just about stall with the valve open at idle. This is a test for clogging of the EGR passages. Use a test lead and ground the EGR solenoid control circuit. With the engine running, the valve should open with throttle snap. Check the vacuum lines for good connection and no leaks.

On many vehicles, the EGR will open just with throttle snap normally. Some others need to have vehicle speed input to the ECM before the valve will open.

On this system, code 32 will set if the valve either doesn't open enough when commanded, or if the passages are restricted so that the valve opening doesn't produce a sufficient change in MAP.

Again, a scan tool and the factory trouble tree for this fault code are very helpful in pinpointing the cause of the code.

Warlocksirix

sorry did not know the setup you ran was a different design. But thank you verry much for the heads up on all this, i'll try the vacuum T this afternoon. I really don't want to replace the EGR again...it's a nightmare on this engine lol. I am not entirely sure on this factory trouble tree you speak of, also whats a good priced tool you recommend for these cars?

ASE doc

None of the commonly available newer scan tools will work. They are for ODBII vehicles with 16 pin data link connector. The GM OBDI has a 12 pin assembly line data link(ALDL) connector. You need either a professional scan tool with OBDI capability(like my new Modis Ultra at $7,000), or an older used tool that is compatible with the GM OBDI. Win ALDL is also an option, if you have the electronic savvy to build your own diagnostic connector and don't mind the headache of using a put together tool.
I recommend the Snap On MT2500. Commonly known as the "brick" for its shape and virtually indestructible chassis. These were the scan tool to have back in the 90s and early 00s. I still have one in the bottom drawer of my cabinet. I paid $2,500 for it 15 years ago, now it's practically worthless for trade in. They are available these days on ebay and craigslist for about $200 or so.

Before you buy one, be sure that it has a power cable. The MT2500 has only a small 9v battery for memory and will not power itself for diagnosis. Therefore, you have to have the power cable that comes with the tool for the OBDI cars. You of course also need the GM 12 pin connector and the domestic cartridges. This tool will do everything the factory Tech1 would do.

As far as the diagnostic trouble tree goes, unless you are really familiar with the systems on these GM vehicles, you pretty much need the trouble tree for guidance on what to test and how to test. The trouble trees for these cars are pretty cool in that, aside from a compatible scan tool, they don't require any special tools. Mostly just a 12 volt test light, vacuum pump, maybe a DVOM. I've been a professional auto technician since 1976 and started with GM in 1990 as a runnability diagnostic specialist. I still use the trouble trees to save time and so that I don't overlook anything. Especially since these early systems are not something I see everyday.

If you don't have a service manual for the car, you should find one. I have the same Helms Publishing factory shop manual for my 87 IROC Z28 that we used at the dealer. You can't get better information on the car. I have used the hell out of that book, so much that I've had to tape up the spine to keep it in one piece. You can still get these books and I think they're also available on CD ROM from Helms online.

Also, as I said before, I think that the Haynes manuals have the factory trouble trees. All Data, which I have here at work, has the factory service information, right out of the shop manual. Not as complete but enough to get you where you need to go. I know that the public libraries around here have All Data and Mitchell On Demand on their computers for public use. I'm not sure about the libraries in your area but it's worth a shot. The trouble tree will print out in one or two pages.