'86 e-check problems - Any help appreciated

I have a 1986 B2000 that is stock with 150K miles. I keep failing emissions tests because of high NOX. I have checked the thermo switch and thermo sensor and they are working properly. The EGR valve, position sensor and EGR control solonoids are working OK and the EGR is actuating when the engine is speeded up (but maybe not being controlled properly). I checked the voltage output of the O2 sensor under several conditions and it seemed OK but I changed it anyway (the old one had some black carbon on it when removed). The A/F solonoid seems to be working OK - I can feel it click when voltage is applied. The thermo sensors and O2 sensor are being seen OK by the ECU. I have tried changing the ECU too. The vacuum advance is working and the timing at idle is dead on. I also changed both cats a few months ago. The truck runs great and gets about 22 mpg - but it won't pass the e-test.

When I put a dwell meter on the brown/yellow check connector to adjust the mixture it indicates a varying dwell for 15 seconds or so after starting the engine and then jumps to 27 degrees which I understand means it has failed closed loop operation for some reason and gone into a fixed dwell state. If I then rev the engine for a moment the dwell will drop to a very low value and then gradually creep up to the 40 to 60 range and then suddenly after 20 seconds or so will drop to 27 degrees and stay there.

Anybody have any good tips on how to solve this?

Thanks in advance.

post deleted

post deleted

Hello MazdaTweaker,

I don't have a shop manual on this truck - they seem to be rare - but I do have access to Mitchel info. I have done a number of checks that they list for the emissions system. Following their procedures I have checked:

Voltages at ECU with the ignition on but engine not running - all seem OK.
Vacuum switch and vacuum sensor - they are OK.
Themor sensor and temp switch - switch closes and sensor resistance varies with temp according to spec.
Solonoids on carb are OK and duty solonoids are OK too. Clutch and neutral switches are OK. O2 sensor is new and voltages out seem OK (it's an after market Bosch).

With regard to the EGR valve. I have tested it under vacuum and have removed it and cleaned it (it was pretty clean) and verified that the diaphragm moves under vacuum. I have checked the position sensor resistance and it may be a little out of spec - it is 5K across the resistor as it should be and on one variable leg it vaires from 5K (no vacuum) to .7K (w/ vacuum) like it should, but the other leg goes from 2K (no vacuum) to 6.2K (w/ vacuum) instead of .7K to 5K like it should. I don't know if this is enough to cause a problem.

There is a small vacuum leak in the EGR valve. If I pull vacuum on it, it deflects fully by about 4 inches of vacuum. If I pull 15 inches of vacuum it leaks down at about 1 inch per second until 4 inches and then the leak seals and it stays there. Apparently as soon as it starts to come off full deflection the leak seals (it still feels like it is fully deflected according to my finger).

I have tee'd a vacuum gage into the EGR line and can see vacuum is being applied and feel the EGR move when I rev the engine - but I don't know if it is moving as much as it should.

I went out this morning and drove the truck until it was warm and checked the dwell. Today it is reading 0 degrees and won't show any dwell whether I rev the engine or turn the mixture screw. ??? I checked the temp switch and it was closed (about 0.3 ohms to ground at the ECU). I also checked the idle switch and that does not seem to be working - could that be a problem?

BTW, I can screw the mixture in until it will die if I pull a vacuum line off the carb or screw it out until it speeds up when I pull a vacuum line. I have it set now to where the engine just slows a little when I pull a line.

I'm sorry to be so wordy, but it's hard to condense this stuff.

post deleted

Did it with engine both cold and hot - same result either way. Tests are repeatable.

Idle is degraded with 2 inches of vacuum on EGR valve.

Engine dies with 2.5 inches of vacuum on EGR.

When I rev the engine to 2000 rpm I can feel the EGR valve twitch just a tiny bit. It takes over 3000 rpm to hold it open any steady amount.

post deleted

I figured out why I wasn't getting a dwell reading. When I tested the vacuum switch I didn't get the connector all the way in and it had backed out and wasn't making contact. I have also adjusted the idle switch to spec.

With the engine hot and standing still at idle it is at 27 degrees. If I pull the vacuum line that feeds the duty valve off the carb/manifold, the idle slows by 80 rpm or so and gets a bit rough and the dwell goes to zero. When I plug the vacuum line back in the dwell starts climbing a few degrees a second up to 60 or 70 and then jumps to 27 degrees.

At 2000-2200 rpm the dwell departs from 27 and is jumping between 30 and 50 (now and then going into the 20's). It keeps doing this as long as I keep it at around 2100 rpm. When I bring it back to idle it starts climbing a few degrees per second up to 60 or 70 and then jumps to 27 degrees.

All this is with the truck standing still. BTW, the dwell meter I am using is a Snap-On digital unit I borrowed. So it is just taking snap-shots at a rate of about 2 or 3 per second.

The plugs have close to 20,000 miles on them now and when I changed them the old ones were very clean and looked like the engine was running lean. I don't know how many miles were on them - they were the ones in there when I bought it. I will pull them again, check the brand and do a compression test. Then I will try and do the dwell test with the truck in motion.

Eh, well I would like to help ya by letting ya in on a little secret.

Get the truck to operating temp. and then pour small amounts of water down the carb with the throttle opened a bit. You should see black stuff come out of the exhaust when you do this.

Have someone watch for you until there is nothing more coming out. It cleans the valves and combustion chamber out from all the carbon pretty well I would say!

Well I did the vehicle in motion test with the dwell meter hooked up. One thing I learned is my vehicle tach is right on.

At a steady 2100 rpm at about 50mph the dwell sits at zero for about 20 to 25 seconds and then jumps to 27 degrees. If I accelerate just a bit for a moment the dwell goes back to zero and the cycle repeats itself.

Whenever I am accelerating normally it seems to go to zero. Right after I shift the dwell will be in the teens or 20's for a moment and then go to zero until the next shift.

Next I will pull the plugs and do a compression test.