TR stumbling and backfiring

I made a 60-mile ride the other day with my 88 TR, stopped for about an hour, started it back up and when I got on the highway it stumbled/hesitated for about a mile, then cleared up. Now it does it also when cold, does not last long (about a minute), but I am worried it may get worse. It even backfired.

It does have the cats removed and I retuned the KE to compensate. Ran fine after that for a couple of months, just started after this 60-mile drive.

Any hints where to start?

 

Thanks for the tips. I checked the spark plug wires and all the connections, everything was OK. I will check the triangular relay next.

Meanwhile I did some troubleshooting by hooking up the O2 sensors to voltmeters. Started the car (started right up by the way), let it get warm, found that the left bank was running right where it should (~ .7 volts closed-loop) but the right bank would drop to .1 when it stumbled. Went open loop and the right bank would drop to 0 during the stumble. It only lasts for about a second, then recovers. Happens maybe once a minute.

This tells me that the problem is the fuel system because if it were ignition the O2 would go way rich.

The car was just sitting when doing this, and I don;t recall it being worse when driving (and I live on a dirt road so I woulkd have guessed the bumps would have made it worse).

Also I did bypass all the high current-draw conectors in the fuse box two years ago.

Another hint or maybe not related at all: the power to the clock went out twice in the last week. I know the battery is OK as it starts great. I can't imagine any circuit that the clock would have in common with only the right bank.

Phil,
Would the relay cause only one bank to go bad?

 

The plot thickens...

Now it won't run: it starts but dies when I let off the starter. Would this relay finally failing altogether cause this no-start problem?

 

I pulled the relay and checked it by hooking Terminal 31 to ground, 15 to + 12v. It clicked fine so i checked the resistance between 81 and 30, was 0.02 ohm. I even shook it a bit to make sure there was not a loose contact inside, stayed at 0.02 ohms the whole time.

I may get a new one anyway becasue it could be a heat thing. And $150 for the peace of mind is worth it. But I do want to find the smoking gun.

I'll check the injectors next

 

Steve,
Thanks for the quick reply, I will check the tachometric relay tonight and also check again for spark.

I am wondering if the stumbling and the new no-start are even related. And the third gremlin of the clock reset (usually an indication of a disconnected battery) could be a third issue.

 

Hooked up my voltmeter to the violet wire, cranked it and it started right up! Not only that but the stumble is gone. Runs perfectly.

I then hooked up my voltmeters to the O2 sensors and they are reading fine. I noticed that the right bank takes about a minute longer then the left bank to stabilize at ~ .75 volts, but I guess that the right O2 sensor just takes longer to heat up. Not related to the other problem.

So here is my theory: the viloet wire may have been loose in the white connector. When I rebuilt the fuse box I did not bypass that one. Just like my right-turn signal wire that jiggles loose once a year, maybe that is what happened. When I put the alligator clip on the violiet wire maybe I shoved it back in place.

Anyway I want to really thank everyone who chimed in. I can't imagine getting this far without your help.

 

The stumbling and backfire are back.

I don't think it is related to the no-start problem, as the car started great. Drove 30 miles, started again no problem.

Does not seem to be temperature related, as it happens just as often cold as warm.

The right O2 sensor goes to .100, left stays fine, so it is the right bank.

Must be fuel as the sensor is going lean (.100), if it were spark it would go rich (> .900).

Seems to be part-load, because if I am > 50% throttle and above 3500 RPM, does not backfire. For this reason I doubt it is fuel flow (like a plugged filter).

I doubt it is a loose conection in the fuel circuit electricals because when I drive on my bumpy dirt road it does not do it any worse. Smooth pavement and even at idle it will happen.

I may change the injectors next but I would have thought that it would not be so intermittent.

Any further thoughts are appreciated.

 

What I could do is put a scope on the violet wire to see dynamically what is happening (better than a voltmeter), and if I see it cutting out at the same time the right-bank O2 goes down then it would be a clue.

But wouldn't a failure in that circuit show up on both banks?

 

The problem is getting a lot worse, now it only runs for about a minute and backfires violently. I measured the supply pressure to each KE and it is right where it should be. Tells me that the pumps and filters are OK and that the tachometric relay is OK.

I put a scope on the right-bank coil wire and the trace goes highly unstable at the same time as the backfire.

I researched all the ignition threads and was going to swap the module to the left hand side (in theory then the car should not start due to the tachometric relay), but it looks like the module is potted pretty well to the heat sink. Do I just take a thin knife and pry under the module? Or do I have to cut the seal/heat-transfer paste?

As to sourcing: I studies all the threads and it seems that T. Rutlands was last year one of the best sources. Motorcityreman seemed almost too cheap to be true. Not sure if the market is any different this year

 

Steve,
Thanks again for the help. The backfire had become so violent that I thought it would do some serious engine damage, so I decided to play it safe and replace the right-hand coil/module assembly. That did the trick! It now starts and runs great.

However, the original symptom of the right-hand O2 sensor varying remains. I guess that was a red herring to the misfire.

So I can still use advice. Here is where I am at:
Both O2 sensors are brand new (I know that was not the cause but I thought the originals needed a break after working for 24 years). Also is the fact that right after I shut it down they both read within .003 volts of each other at 0.913 (which I find pretty amazing that they are that close)
Open loop the LH side is pretty steady at .480 - .520 volts. The RH side varies from .200 - .550. Since this is open loop I don't think the EHA is to blame.

I then went closed loop and the LH side is dead steady at about the same values as above, the RH side still varies about the same as above.

Maybe another red herring but when I hook up the O2 sensors the way the WM says (7-12 goes to the black connector) the LH side goes full lean and the RH side goes full rich, telling me that the signal / feedback are reversed and each is trying to over-compensate
each other as they have the false signals.

I do not know where to go next.

 

Steve,
Thanks again for the prompt and detailed reply!

Thanks for the advice on checking to make sure the LH O2 is connected to pin 8, I will do that. Should I check the dual-pin connector as well, or are these fed from a common supply?

I can vary the O2 sensor from full-lean to full-rich by turning the FD mixture screw. LH side seems more linear (ratio of screw-turns to O2 change) than RH side (seems to be more sensitive/step-function).

I had it set to.7 volt before, had leaned it out a bit to ~0.5, I will change it back to 0.7. I am measuring the open-loop voltage between the unplugged wire of the O2 and ground.

I will check to confirm that the water thermoswitch is closed when cold, thanks. Interestingly enough the O2 is ~ 0.8 at start, which seems correct, then goes to 0.2 for ~ 3 -5 minutes, then goes to the values I mention above. I would have thought it would go from rich right to the warm-condition, niot sure what this interim is.

 

Steve,
Thanks once again for the quick reply, I appreciate it. I am confident that with your help I will be driving the TR this weekend.

One thing I realized in rereading my posts (and your replies) was that I may have been negligent to mention that my recent measurements (LH dead steady, RH varying) were all at idle since I have not been able to drive the car due to the backfire. I wanted to get the idle base O2 right before driving. So when I hit the throttle in neutral the LH does vary from lean to rich, as does the RH. Then the LH at idle goes steady while the RH still varies.

Hope that did not cause any confusion.

Thanks also for clearing up the issue of lean while still warming up.

I am grateful for the support you and others have given me; while I don't look forward to problems, this has given me a greater understanding and appreciation for my TR.

 

The new coil/ignition module definitely solved the backfire. Timing light and scope show steady ignition.

I borrowed a couple of Fluke meters for the O2 sensors so I could read the average and both sides are ~ 0.5 v while letting it idle for 10 minutes. I think the lean/rich excursions I saw were basically the KE doing its job.

Went for a 30 mile drive, runs great.

Once again, thanks for your help!