02 sensor heating relay clicking and strange fault codes.

[Shane318is]

Hello all...curious to see what you think!

So, I live in Australia, I drive an E30 318is, highish miles (200k), rhd (obviously) and fitted with the factory catalytic converter.

Driving home from work (on a hot day) my engine seemed to lose power when I backed off the throttle, like it was dying or cutting out when I lifted (even slightly) off the gas. It ran great during acceleration.

I pulled over to check all was ok, everything looked ok under the hood. When I tried to restart the car to drive off, it wouldn't fire, not even on one cylinder.

I waited 20 minutes, and it started up perfectly, so I drove off. About 15 minutes of driving later, it started to die again on a lifted gas pedal. I drove it home using lots of throttle!

I checked my engine at home, found some potential vacuum leaks, and fixed up the ones I could see. I also found something very strange. The relay for the o2 sensor heating circuit is clicking constantly, maybe 3 times a second. I swapped the fuel pump relay and the o2 sensor heater relay (they're the same on my car) and the clicking stayed in the o2 sensor relay position. In other works it looks like the relay is fine, it is just getting a signal from the DME to switch on and off very quickly. Any ideas?

Last night I borrowed a diagnostics kit from work (Autel maxidas)...connected it to the obd port and got the following fault codes.

200 DME control unit
76 Idle CO potentiometer

Is my DME damaged?

Has anyone else ever found the Idle CO potentiometer fault?

Finally, this diagnostics tool seems quite useful..live data display of all the engine sensors. So I plan to clear the fault codes and go for a good drive later.

[bmwman91]

Weird thing about the CO fault. The cars that came factory-equipped with catalytic converters & O2 sensors don't use the CO input. Have you at any point worked on the wiring harness up on the firewall and connected a small single-wire connector that was sitting up there, disconnected? If so, disconnect that. It is the CO input and is supposed to be disconnected if you are have the O2 sensor option.

Otherwise, it could be a dying DME, or maybe the high temperature is causing it to switch in and out of open/closed loop operation. I am pretty sure that the DME turns off the O2 relay when the engine is cold, under full throttle and above 4000RPM which are open-loop conditions. That could explain why it runs better when you give it more gas.

Maybe the O2 sensor died and developed a short that is stressing the relay?

[Shane318is]

Thanks for the reply.

So I watched the live data from the car warming up, the diagnostics tool gives the option to display the co potentiometer voltage..and sure enough, there is a voltage displayed and it varies with throttle opening.

The o2 sensor heating circuit is also being switched on and off..though it seems like only when the idle speed drops a little low.

I checked the single wire on the bulkhead near the ignition relay and it is disconnected.

Looks like my dme is kaput!

[Shane318is]

As you can see the idle was hunting a bit when the engine warmed up. What's suspicious is that the co pot and the o2 sensor seem to track each other exactly...in other words when the single co pot wire is disconnected the dme "makes" it's own co signal by scaling the o2 sensor readout.

[Shane318is]

Looks like CO potentiometer voltage is consistently almost exactly 4 times the the oxygen sensor voltage.

[bmwman91]

Strange. Maybe pop open the DME cover and look for obvious damage?

[Shane318is]

will do.

What's confusing is that the live data shows the o2 heater is being switched on and off......but it looks like "o2 sensor control" is actually off.

All this talk about "idle co potentiometer"...it doesn't seem like there is a specific "co potentiometer"...doesn't the dme simply just read the afm door position?

[bmwman91]

The CO pot is buried in the AFM. You can see a round protrusion next to the main plug, and it is plugged. There is a little adjuster screw in there. People claim that messing with it will totally screw up your idle and mixture, but I am not so sure about that since the output of that potentiometer is not connected (on cars with O2 sensors anyway). Anyway, I wouldn't mess with it.

Thinking more about it, it might not be all that weird that the CO voltage mirrors that of the O2 sensor. The CO pot is there to adjust the mixture at idle, and is sort of a "manual" version of the O2 sensor. Is the CO pot plot's Y-axis showing voltage, or some other parameter?

My guess is that either:
- The two signals are tied together via resistors in the DME, and the CO voltage is going to follow the O2 sensor voltage since its input is "floating" (the wire is disconnected)
- If the Y-axis is some software value (not volts) then it is just some internally calculated parameter that isn't actually being used for fuel calc's since the O2 sensor is there.

So the CO signal may not be indicative of anything. When you say, "live data shows the o2 heater is being switched on and off......but it looks like 'o2 sensor control' is actually off" do you have a plot or something? If you are referring to the up/down in the O2 sensor voltage, that is normal. In fact, it should switch between 0.1v and >0.6v at least twice per second when it is warmed up. The stock O2 sensor is ONLY accurate at 14.7 AFR, so the DME will run run more and more rich until the O2 voltage hits a certain threshold and then run it more and more lean until it hits the other voltage, and so on.

It might just be a bad DME, or a bad AFM (I can tell you how to check it). Can you find anyone locally with a good DME to borrow?

[Shane318is]

Thanks again for the replies!

I agree that the two signals must be tied together within the DME.

No, I mean that the live data shows that the "02 sensor control" is permanently off.

It will be clearer when I load a video of the live data on to youtube, I'll post a link here.

[Shane318is]

live data video



Need to watch in reasonable definition!

https://www.youtube.com/watch?v=2VwAG-ug-gE