View Full Version : New O2 sensor, and a strange burning smell
Hi all, so in the hunt for the cause of my high idle(2000-2500 RPM), I replaced the O2 sensor, as this was mentioned as a possible cause. I had a fairly easy go at the job, however, when I started the car upon completion, I got a strange burning smell(it also did not solve the idle issue). It was definitely not the usual smell to me; it was similar to burning plastic, as far as smells go. I do think I could notice a bit of gray fumes or whatnot that is not usual. Any thoughts on that? Is there a normal "burnoff" period or anything that I may not be aware of? As an aside, I have a hunch that the mixture is set rich. Thanks for any thoughts!
DMC-81
02-28-2018, 07:30 PM
I didn't notice a burning smell when I replaced mine. Mind you, I had some smoke and smell as I was curing Hi Temp POR -15 exhaust paint at the same time. But definitely not a smell of plastic.
Was there perhaps a plastic cap on the sensor end that didn't get removed or was the wire routed too close to the exhaust pipe?
I didn't notice a burning smell when I replaced mine. Mind you, I had some smoke and smell as I was curing Hi Temp POR -15 exhaust paint at the same time. But definitely not a smell of plastic.
Was there perhaps a plastic cap on the sensor end that didn't get removed or was the wire routed too close to the exhaust pipe?
Thanks for the response. There was not a plastic cap, and i know the wire was threaded through the frame properly, and isn't laying against the exhaust. The car only ran for about 30-60 seconds, so I doubt there'd be enough heat generated at the exhaust to melt anything....
Jonathan
02-28-2018, 10:44 PM
The smell is probably related to the mixture being too rich.
The original idle problem being up around 2,500 RPM is probably a bad electrical connection on the thermistor, which is located under the intake in the valley (which admittedly sucks to need to go in there to correct it).
The smell is probably related to the mixture being too rich.
The original idle problem being up around 2,500 RPM is probably a bad electrical connection on the thermistor, which is located under the intake in the valley (which admittedly sucks to need to go in there to correct it).
This would, unfortunately, make sense as that problem introduced itself immediately after the thermister job. I distinctly remember doing my best to make sure the harness on the actual thermister secure before re-assembly. Is there a physical way to check that without doing a full disassembly, down to the valley? I keep wondering if there's a way to get a go pro camera down in there or something with my arm, just to check that harness. The idle jumps around erratically; if I let it sit, for a minute, it will occasionally just jump from 2000-2200 up to 3000-immediately-without any physical touch to the throttle at all. I think the prudent thing is for me to go back and double check all my connections....sigh...
Thanks for the help!
Assuming that the hammer on the throttle presses the idle microswitch, remove the white multi plug that has 4 wires on it. Use a multi meter, set the resistance k ohms and put the multi meter probes on the black/yellow and black/slate (grey) wires at the multi plug. If you get no reading ie open circuit, then you have a wiring problem to the thermistor. In this case, you could refit the white multi plug and put a 10 k ohm resistor across these wires effectively fooling the Idle ecu that it is referencing the thermistor.
Let's hope that the idle motor doesn't have a short or that the ecu had not blown one of the switching transistors that operates the motor etc
Good luck
Assuming that the hammer on the throttle presses the idle microswitch, remove the white multi plug that has 4 wires on it. Use a multi meter, set the resistance k ohms and put the multi meter probes on the black/yellow and black/slate (grey) wires at the multi plug. If you get no reading ie open circuit, then you have a wiring problem to the thermistor. In this case, you could refit the white multi plug and put a 10 k ohm resistor across these wires effectively fooling the Idle ecu that it is referencing the thermistor.
Let's hope that the idle motor doesn't have a short or that the ecu had not blown one of the switching transistors that operates the motor etc
Good luck
Brilliant suggestion! Thank you so much. While I wouldn't put it past myself to make a mistake, seeing as how I did the entire VoD job just to replace the thermister, I distinctly remember being in there, and making absolutely sure the wiring harness was securely fastened to it after it was replaced. Will try this test, and report back. Thanks!
NckT- the white 4 pin harness, is it the one shown in this picture? If so, the colors on my harnesses are black/green on the left side, and black/black on the right side(the side that looks to connect to the idle microswitch). I get 0 ohms/full continuity across the left side(green/black), and open across the right side(black/black).
Am I reading the wrong white harness? I've got a later car, FYI, it's apparently an '84 that was titled as an '83. If I'm reading the correct harness, what does this mean? Thanks.
Apologies(it's early for me!). Of course you weren't talking about that harness; you were referring to the 4 pin white harness at the idle ECU. Duh. I checked that; yes, it's open across black/yellow and black/slate. One follow up question: is it possible that this would be due to anything other than the electrical harness at the thermister? I will of course go back in there if need be, I just want to see if protocol would call for any other diagnosis, before stripping down to go back to the thermister location. Thanks.
Bitsyncmaster
04-08-2018, 10:58 AM
Apologies(it's early for me!). Of course you weren't talking about that harness; you were referring to the 4 pin white harness at the idle ECU. Duh. I checked that; yes, it's open across black/yellow and black/slate. One follow up question: is it possible that this would be due to anything other than the electrical harness at the thermister? I will of course go back in there if need be, I just want to see if protocol would call for any other diagnosis, before stripping down to go back to the thermister location. Thanks.
The two connections also go through the bulkhead connectors so you may want to check if a connection is bad there.
Apologies(it's early for me!). Of course you weren't talking about that harness; you were referring to the 4 pin white harness at the idle ECU. Duh. I checked that; yes, it's open across black/yellow and black/slate. One follow up question: is it possible that this would be due to anything other than the electrical harness at the thermister? I will of course go back in there if need be, I just want to see if protocol would call for any other diagnosis, before stripping down to go back to the thermister location. Thanks.
I live in the north of England so I'm in a different time zone. ..
As what Bitsyncmaster said, the wiring from the ecu goes to the church window shaped multi plugs that are located behind the ignition coil cover in the engine bay. Look for the black/yellow and black/skate (grey) wires on the bottom right multi plug connector that has 9 cables going to it and then put the multi meter problem on those wires (without taking the multi plug off). If you get a resistance reading then the wiring problem is from the multi plug to the engine. If not then it's likely to be at the thermistor itself.
I would recommend to do a continuity test of the wiring from the multi plug to the ecu first to confirm the integrity of these cables before re-stripping the inlet manifold off again.
Just one tip, I've only seen it on 2 cars where the wiring plug for the thermistor and frequency valve were put on in each others place, so if your frequency valve is operating then at least your connectors are on the correct way. Double check the wiring colours at the thermistor if you need to take the inlet manifold again based on your diagnosis.
Good luck with it
Ps, the 10k resistor is a temp fix. I've done this at tech events by making short make to female wires to extend the multi plug at the ecu to the ecu for the non thermistor cables, then used 2 female blade connectors on a 10k resistor and connected this direct to the pins of the ecu where the black/yellow and black/slate (grey) cables were to go, as at the event there wasn't time to step the engine down to fix the thermistor connection.
thanks much-if I get a resistance reading of 0 ohms across black/yellow and black/slate at the white 9 pin connector next to the ignition coil, when the harness is seated, what does that tell me? Cause that's what I get.
I live in the north of England so I'm in a different time zone. ..
As what Bitsyncmaster said, the wiring from the ecu goes to the church window shaped multi plugs that are located behind the ignition coil cover in the engine bay. Look for the black/yellow and black/skate (grey) wires on the bottom right multi plug connector that has 9 cables going to it and then put the multi meter problem on those wires (without taking the multi plug off). If you get a resistance reading then the wiring problem is from the multi plug to the engine. If not then it's likely to be at the thermistor itself.
I would recommend to do a continuity test of the wiring from the multi plug to the ecu first to confirm the integrity of these cables before re-stripping the inlet manifold off again.
Just one tip, I've only seen it on 2 cars where the wiring plug for the thermistor and frequency valve were put on in each others place, so if your frequency valve is operating then at least your connectors are on the correct way. Double check the wiring colours at the thermistor if you need to take the inlet manifold again based on your diagnosis.
Good luck with it
Ps, the 10k resistor is a temp fix. I've done this at tech events by making short make to female wires to extend the multi plug at the ecu to the ecu for the non thermistor cables, then used 2 female blade connectors on a 10k resistor and connected this direct to the pins of the ecu where the black/yellow and black/slate (grey) cables were to go, as at the event there wasn't time to step the engine down to fix the thermistor connection.
In that case there could be a bad connection on the thermistor or bad wiring from the coil church window multi plug to the thermistor.
Shove in a 10k ohm resistor at the church window multi plug and see if the car now idles correct
In that case there could be a bad connection on the thermistor or bad wiring from the coil church window multi plug to the thermistor.
Shove in a 10k ohm resistor at the church window multi plug and see if the car now idles correct
Thanks, NckT. so, to be clear, I want to jam a 10k resistor across the black/yellow and black/slate pins at the white church window harness, and keep that harness seated, right? is 1/4W a problem? I have about every resistor value known to man here, but they are all for audio electronics, so most are 1/4W or 1/2W 10k's.
Thanks, NckT. so, to be clear, I want to jam a 10k resistor across the black/yellow and black/slate pins at the white church window harness, and keep that harness seated, right? is 1/4W a problem? I have about every resistor value known to man here, but they are all for audio electronics, so most are 1/4W or 1/2W 10k's.
Yes that's what I mean. 1/2 w would be fine and probably easier because of the thickness of the resistor legs.
It'd be better if you could fit it between the black/yellow and black/slate (grey) called at the idle ecu. It's the same difference.
Sorry, I somehow may have gotten a "misread" on my previous DVM reading. When I first checked that resistance reading, I got 0 ohms. I just went back to the white church harness at the bulkhead, unseated it, and reseated it, and took the readings again; this time I got 8.95k Ohms across black/yellow and black/slate pins.
You've probably had a bad electrical connection at the church window multi plugs then that's causing intermittent problems. Use electrical switch cleaner to clean the pins and be careful reconnecting the plug as sometimes the male pins can get pushed back if forced.
Recheck the resistance reading at the idle evu plug
Good luck
Jonathan
04-08-2018, 04:56 PM
55897
This is approximately what the jumper being described to you is going to look like once installed. I did not get the 10-Ohm resistor part of the advice when I needed to do this, so you'll just see a short jumper wire in my set-up.
The jumper fools the ECU to think the thermistor signal is present and good to go and the other short wires basically connect the other connections in that white connector to also convince the ECU that they are plugged in normally. This didn't work perfectly for me (perhaps the lack of resistor?) but it did get me to the end of the driving season and then I had the justification I needed to tear into the valley for the first time in my ownership of the car. I had other projects to do in there but sorting out the bad thermistor connection was the main one.
Another test of course if you find your car idling more or less ok, is to disconnect that white plastic connector on the ECU and it should jump to exactly 2,500 RPM and sit there. It's sort of the fail position for that part of the idle system and a good way to narrow down where your problems are.
Man, you guys are the best! THANKS SO MUCH! This community is a huge resource for a guy like me in Central Ohio. Kudos!
Ok, I will clean those harness pins(they are coated in a conductive grease; must have been from the previous owner, as I generally go for clean, dry metal to metal connections).
I will read at ECU and report back.
Also, I know it's not perfect, but I did manage to get some images of the temp sensor, and the thermister, using a GoPro camera taped to a stick. While not pretty, I would like you guys to take a look at it if you have a moment, just to possibly confirm what I'm seeing. As best I can tell, the harness at the thermister looks to be seated.
Ok, thanks again-will report back soon; here's the link to the video:
https://wetransfer.com/downloads/917e817fce1d4cbae61b97aed9f0296620180408210643/ddfdc69b4e622d0d7440f7b9c1cf34ed20180408210643/c6517c?utm_campaign=WT_email_tracking&utm_content=general&utm_medium=download_button&utm_source=notify_recipient_email
Alrighty, last post for the day, I'm gonna wrap this up and dig in tomorrow after this:
I thoroughly cleaned and reseated the white 9 pin bulkhead harness. Upon re-reading at the ECU, i got 18.7k Ohms across black/yellow and black/grey.
The car idles solidly at roughly 2300-ish RPMs. Hope this piece of data is relevant: if I unseat the white 4 pin harness at the ECU while its running, I get NO change in behavior of the idle. It sits bang on at 2300-ish; it did not even slightly nudge up or down when I unseated it.
So...does it look like I am revisiting the VoD?
Jonathan
04-08-2018, 05:59 PM
Hey, don't mention it! I asked many of the same questions over the years since getting my car in 2007 and so if I can now help answer some of those questions for a newer wave of owners, then giddyup.
http://dmctalk.org/showthread.php?11945-Unsure-of-dwell-reading&p=181425&viewfull=1#post181425
Now, this thread I referenced is admittedly confusing and a lot to read. I wanted to include it because the post I linked to in it mentions an O2 sensor ground reference wire than can be a problem. It is on the passenger side of the manifold, towards the front of the car and sort of above the frequency valve location. It normally has a solid red coloured wire to help identify it. I found a problem with my own and it helped, but it wasn't the only thing going on.
This thread was in late 2015 and I did my thermistor and valley work in the winter of 2014 (before this thread's season). So while I got the thermistor issue sorted out, I still had a mixture setting problem. A new oxygen sensor and a mixture setting by a local mechanic who had experience in knowing how to do it for me solved my problems of poor idle.
Just some extra reading material in case the next thing or two you try or test isn't the entire solution.
Alrighty, last post for the day, I'm gonna wrap this up and dig in tomorrow after this:
I thoroughly cleaned and reseated the white 9 pin bulkhead harness. Upon re-reading at the ECU, i got 18.7k Ohms across black/yellow and black/grey.
The car idles solidly at roughly 2300-ish RPMs. Hope this piece of data is relevant: if I unseat the white 4 pin harness at the ECU while its running, I get NO change in behavior of the idle. It sits bang on at 2300-ish; it did not even slightly nudge up or down when I unseated it.
So...does it look like I am revisiting the VoD?
Morning. As you now get a resistance reading at the ecu it shows that you now have a connection at the thermistor. No point in stripping the engine down to something you've just fixed!
I'd start diagnosing the wiring and operation of the idle ecu:
First check that the black switch at the throttle is closing with a positive click when the throttle is closed. This negative switching goes through a diode located in the wiring loom just being the idle ecu in a small battery sized cylindrical black plastic container before going to the ecu.
For testing purposes get a link wire and with the white plugs still connected to the ecu, connect the link wire from earth (ground) to the black/light green cable. If the idle goes back down to normal idle speed then there's a problem with the switch or diode or its wiring. If it doesn't then your problem is elsewhere. At this point pull the black plug from the idle motor and check to see if you get 12v at the centre pin of the multi plug with the ignition key in position 2.
I'll assume you've checked the fuses first etc
NckT-Thanks much!!!! WHEW!!! Great news, re: engine strip down. I really didn't want to do that again(I also believe that the video I posted a link to shows a visual on the harness being seated properly).
I will do this test at the ECU today. That's great. Before testing, I do want to jump in on the fuse box: a problem introduced itself after I did the thermister job; I seem to be getting several intermittent connections at the fuse box; I pulled all the fuses and tested their continuity, and can recreate the problem sometimes by picking up the fuse box and physically shifting it with the key at position 2. It's a problem I don't really know how to fix, as it seems to be local to the actual contact points of the fuse box.
Ok, back later today with more info. Thanks!
Morning. As you now get a resistance reading at the ecu it shows that you now have a connection at the thermistor. No point in stripping the engine down to something you've just fixed!
I'd start diagnosing the wiring and operation of the idle ecu:
First check that the black switch at the throttle is closing with a positive click when the throttle is closed. This negative switching goes through a diode located in the wiring loom just being the idle ecu in a small battery sized cylindrical black plastic container before going to the ecu.
For testing purposes get a link wire and with the white plugs still connected to the ecu, connect the link wire from earth (ground) to the black/light green cable. If the idle goes back down to normal idle speed then there's a problem with the switch or diode or its wiring. If it doesn't then your problem is elsewhere. At this point pull the black plug from the idle motor and check to see if you get 12v at the centre pin of the multi plug with the ignition key in position 2.
I'll assume you've checked the fuses first etc
More data:
1-with an alligator clip coming off the negative terminal of the battery, while the car was running, I tapped the black/green pin of the white 4 pin harness at the ECU. no change in idle speed.
2-with the key at position 2, i am getting 10.5V at the center pin of the idle motor harness. I can hear it buzz with the key in position 2. I am almost sure at one point I got a visual on the little rotating motor thing opening and closing properly last year during some diagnosis.
Hmmm.....I do know for sure there are some gremlins at the fuse box. Are there any tests to be done there, or elsewhere? thanks again, gonna get this sooner or later!
Remove your idle motor and see if it's physically stuck inside. The valve should be free floating.
Use a multi meter and measure the resistance between the centre pin of the idle motor and one of the outer pins. Take a note of the reading and then do resistance test between the centre pin and the other outer pin.
Idle motor: by taking the hose clamp off the "rear" side, i can see the cylinder swings open when putting the key into position 2. should I pull the idle motor and inspect it when removed from the car?
the resistance readings across middle and outer pins of the idle motor harness: 160k, and 33k.
Not the harness, the motor windings themselves.
If you apply 12v to the centre pin of the idle motor and temporarily earth (ground) one of the outer pins (for no more than 1 second), the valve should fully open or fully close. Repeat the test bit then earth the other outer pin in the same way and it should move to the complete opposite way ie fully open or fully close
Ok, gotcha. So at the motor, I get 10.8 ohms from center to right pin, and 9.1 ohms from center to left pin.
If i disconnect, and bridge the center pin, and tap either right or left pin at the motor to ground, I get a whirring sound. It's really hard to catch a visual at the actual motor, but it sure sounds like something is moving.
That help?
Oh, also-one other thing to mention. the brass L pipe that goes into the manifold: I had issues with it upon reassembly. the o-ring fell out. i believe i got it in place, but i cannot be sure it stayed when i reassembled. one other thing to consider. thanks!
Ok, gotcha. So at the motor, I get 10.8 ohms from center to right pin, and 9.1 ohms from center to left pin.
If i disconnect, and bridge the center pin, and tap either right or left pin at the motor to ground, I get a whirring sound. It's really hard to catch a visual at the actual motor, but it sure sounds like something is moving.
That help?
Oh, also-one other thing to mention. the brass L pipe that goes into the manifold: I had issues with it upon reassembly. the o-ring fell out. i believe i got it in place, but i cannot be sure it stayed when i reassembled. one other thing to consider. thanks!
The motor shouldn't 'whirr', as it moves fully open or fully closed, you should be able to hear/ feel the valve give a slight clunk. Remove it from the car to test and look
The motor shouldn't 'whirr', as it moves fully open or fully closed, you should be able to hear/ feel the valve give a slight clunk. Remove it from the car to test and look
"a slight clunk" is a much better description; it wasn't a slow whirr. it was kind of loud; it sort of startled me. i tapped it once on the right, and it made a metallic clunk. i tapped right again, and it didnt do it. i tapped left, and it made the same clunk. i tapped right, the same clunk. i'll pull it and inspect.
the brass L pipe that seats into the bottom of the fuel distribution unit, if that leaked air, could it cause this high idle problem?
Yes, it will draw in air right and cause a high idle if there was an air leak in there.
You could try adding 2 bolts/ wine corks/ bungs etc in the houses where the idle motor goes, and if there is still a high idle then the problem is not the idle motor or the ecu
ok-i removed the idle motor. i can confirm it moves freely.
where to next?(can i assume we are moving downstream of the ECU?)
......You could try adding 2 bolts/ wine corks/ bungs etc in the houses where the idle motor goes, and if there is still a high idle then the problem is not the idle motor or the ecu
ok, think i may have found the culprit, or at least made some progress: i took voltage measurements at the ECU. it's working properly. i choked off the idle motor hoses. same deal. checked all electrical harnesses. replaced o-ring at rear of mixture unit. all with no change.
when i pull the rubber pin at the fuel distributor, the idle drops down to "normal". after speaking to Steven at Delorean, I now think that my mixture may be off, with the car resulting in a high idle. i've tried several times to get a good dwell reading, and have never been very successful; any tips on how to get the mixture bang on? thanks for the help!
A simple method to get it to an approximate setting (assuming that the CPR/WUR is working within spec, vacuum hoses on correct and the vacuum system operating correctly) disconnect the oxygen sensor electrical connection so that the frequency valve operates at fixed dwell, on a warm normal operating temperature, use a 3mm Allen key turn screw right until the engine dies, then turn screw left until the engine dies, then turn it back to the middle of those 2 points, and insert the bung where the tamper proof plug was.
Obviously, count the number of turns you've done so you can go back to your start point if need be
well, i think i'm back at what i now remember to be my starting place in all of this; the reason i did the thermister job in the first place:
the car won't hold an idle once it warms up. it will also often not start once it is warmed up, upon stall out; i can crank all i want, it wont start. i have to wait til it cools down to start it again.
i can tune the mixture so that the car will idle close to 850-ish(more like 1000 often times) at start up. as soon as it warms up, the idle will drop slowy, and then stall out. i think it's possible that the car was at one point tuned rich to compensate for a vacuum leak. when i overhauled the intake seals/thermister/O2 sensor/WUR, it's possible i solved the vacuum leak, and was left with a rich running car; i no longer get the 2000-2500 idle upon cold start.
any ideas as to what could cause that behavior? i have a new warm up regulator, 02 sensor, thermister, stainless steel fuel lines, o-rings, intake seals, distributor in place. i have verified the vacuum advance, idle motor, ECU, are all working properly. kinda stumped. could this just be a matter of getting the mixture set right? this is one thing that i'll admit i'm not a pro at. thanks!!!
Think I have made progress! As of my last shakedown session, I was able to get the car to hold an idle under 1000RPM. This may sound silly, but it appears that the rubber plug at the mixture unit has hardened or dried out; if I force it down with an alan wrench, it would stay well below 2000RPM. While I'm not doing an entirely scientific mixture adjustment(basically going lean til it stalls out, then going clockwise till it holds an idle), it seems to work for now. Thanks for all your help, you guys!!
Powered by vBulletin® Version 4.2.5 Copyright © 2024 vBulletin Solutions Inc. All rights reserved.