'86 e-check problems - Any help appreciated

Aha! I understand why it stayed in closed loop. The engine wasn't hot yet - the thermostat was still closed even though it had idled for a while. When I took it on the freeway I watched the temp climb and as soon as the temp dropped due to the thermostat opening (I assume) then the dwell started jumping to 27 degrees again.

I had done all prior test with the engine hot. So it seems that getting hot is what takes it out of closed loop.

I decided th check the temp switch and thermo sensor while driving. I disconnected the ECU and put ohmmeters across each of them. The engine was probably around 80 to 100 degrees when I started. The temp switch was closed and the thermo sensor was at 2K Ohms which is about right by the book.

As soon as I started driving the thermo sensor started coming down. In the first mile of driving it came down to about 0.5K Ohms even though the temp gage on the dash wasn't rising yet. When the temp gage started rising the thermo sensor was down between 0.3K and 0.4K Ohms. As soon as the thermostat opened the dash temp gage dropped to about 1/4 and the thermo sensor slowly rose to between 0.4K and 0.5K Ohms indicating a small drop in coolant temp.

The temp switch stayed closed this whole time.

I don't see anything sudden or significant in the changes in these two sensors that would kick it out of closed loop operation when the thermostat opened. BTW, I put a new thermostat in a couple of months ago.

post deleted

post deleted

I did this and the behavior was the same at all mixture settings. Once it jumped to 27 degrees there was no response to moving the mixture screw. At any setting I could give it a very short rev of a couple hundred rpm and the dwell would go down to less than 10 and then start walking up several degrees per second until around 50 and then it would jump to 27 and stick there.

I know you don't like to change too many things at once, but I did try changing the ECU since I had another one and the behavior was exactly the same.

I did check the #1 and #2 ACV valves. The #1 was bad and I replaced it ($50), the #2 passed the check procedure in Mitchel's manual. I have not checked the reed valves in the front of the air cleaner. I did open them and clean them when I first got the truck, but haven't looked at them since then.

Edit: I cleaned out the reed valve again - they weren't too bad. I also rechecked the operation of #2 ACV valve - it is exactly on spec.

post deleted

post deleted

I can't do the O2 sensor test because I am in the midst of pulling the head - In my frustration I decided to eliminate the possibility of deposits being the problem.

However, I did do the O2 sensor tests you recommended in another thread as well as the tests given by Mitchel, and the sensor seemed OK. I do have the chart for back probing the connectors at the ECU. The only check I did of the O2 sensor at the ECU was to verify I was getting a signal. I did check all the voltages at the ECU with ignition on but engine not running per the table in the Mitchel data and they were all within spec.

When I put it together I will set the valve clearances and put the proper tension on the timing belt (which is about 20,000 miles old).

Re: cam: I think everything on this truck is stock. The lady who sold it to me about 6 years ago said the engine had been "rebuilt" but I don't know what was done or when. It had about 125,000 miles when I bought it.

Last week I checked the distributor and found the internals were bound up and the vacuum advance actuator was inop (it didn't leak, it was just frozen). I pulled the internals out of the distributor, cleaned everything and got a new actuator. The centrifugal advance moves freely and the vacuum advance works properly now. There are no leaks and there is good vacuum when I rev the engine. Using a timing light I can see the timing advance smartly when I rev the engine. Neither advance was working when I had the last e-check. The idle timing is right on.

MazdaTweaker,

I changed my mind on pulling the head for now and put the stuff I had taken off back on. I did the O2 sensor voltage tests at the ECU that you wanted:

Engine off - hot...... 0.090V

Engine at idle......... 0.8V to 0.9V (mostly around 0.87V)

Quick rev to 2000... drops to less than 0.1V

Hold at 2300 rpm.... stays at 0.00V

Idle w/ shot of cleaner in carb: quick jump from 0.87V to 0.95V and then comes back down over the next 5 seconds.

Idle and pull a vacuum line: drops slowly from 0.87V to 0.02V.

Because today is the last day that I have to fix this before I get a big fine, I took it in for a test. The results were basically the same. It passes everythins except NO at 15 mph. However, it passes NO at 25 mpg. The tests here are done on a dyno which puts a pretty good load on the engine during each test. Here are the latest results:

Speed_RPM__%CO2_%O2_HC(ppm)__CO(%)____NO(ppm)__RESULT
__15__1877___10.0___6.7____37______0.00_______1635_____FAIL
__25__2947___10.0___5.7____36______0.00________985_____PASS

Since it passes at 25 and fails at 15 what can it be other than the ECU isn't activating the EGR valve at the lower speed?

I'm going to tee into the vacuum line to the EGR and try to see what is happening at various speeds/loads.