'86 535i: Pulsing rich idle, poor gas mileage, runs fine under load.

[code]

I'm the second owner, purchased in 2018. The car idled and ran fine but was lacking power, and an inspection of the cam while adjusting the valves showed the cam was worn. Approximately two years ago I replaced the worn stock cam with a 276 from dbilas, along with HD rockers from IE, resurfaced the head and had new valve seals installed. During this refresh I replaced any vacuum and cooling system hoses that looked worn, and also replaced the timing chain guides and tensioner. After getting the car back together, the idle has been rich and lopey/pulsing, and shakes the car intermittently. Gas mileage is awful, the exhaust smells of sulfur and there are drops of gas exiting the tailpipe. I triple-checked the timing before sealing everything back up, and the car runs fine under load, so I don't think that it is a timing issue (but I could be wrong).

Thinking it could be a fuel delivery issue, I replaced the injectors with flow matched Bosch, installed Bosch WR9LS plugs, new Bremi plug wires, new cap & rotor, new FPR, adjusted the throttle body and TPS, and even replaced the oil filler cap. No change. I thought "this is just how the cam must be" and decided to live with it.

A recent conversation with my brother led me to think the car may be running in open loop, so I decided to try to tackle the issue from a different angle. Within the last few days I replaced the O2 sensor, tested the CTS, and checked/cleaned the battery to chassis ground, the engine to frame rail ground, the head to firewall ground, and the harness to head ground with no change in the idle. When testing the new O2 sensor with a multimeter, I stuck one lead in the white male connector by the passenger firewall and the other lead on a ground, which caused the engine idle to even out somewhat immediately (while measuring .462v). I tested this several times with the same result, before and after checking the grounds. This leads me to believe that there could be an electrical issue, but I've checked and confirmed all grounds in the engine bay are clean and secure. The only time the idle will somewhat even out is if I do the O2 sensor test above.

The cluster board has a short which causes the "inspection" light to stay on and the speedometer to jump occasionally, but it has had this issue since I purchased the car from the original owner and did not cause any running issues pre-cam replacement. There are no other electrical issues I know of. The car drives fine under load, and reminds me why I don't like to drive it when I pull up to a stoplight and it idles rough with the shakes.

I've also installed a Mark D chip in hopes it would help with the idle, without success. Video of startup and idle is below (I rev it for a few seconds from the throttle body). Any help or insight would be greatly appreciated.

https://www.youtube.com/watch?v=k2eMe_RqRdE

[Mike W.]

While not denying anything you said, this does not make sense,

When testing the new O2 sensor with a multimeter, I stuck one lead in the white male connector by the passenger firewall and the other lead on a ground, which caused the engine idle to even out somewhat immediately (while measuring .462v). I tested this several times with the same result, before and after checking the grounds... The only time the idle will somewhat even out is if I do the O2 sensor test above.

Even a 70's, $10 analog meter should not load the circuit enough to make any change in how it runs, let alone any digital meter with Ohm/volt in the megohms range. Electrically it just doesn't work that way. That's electrically. It's possible mechanically though. I had an L jet E24 long ago that would cut out at anything over half pedal. Revs didn't matter, it was pedal position. It turned out to be a broken wire going to the CTS that would make contact, until the engine torqued to a point where the harness moved enough not to make contact. Not saying that's your problem, just that weird shit happens that doesn't make sense, and don't assume all the wiring is perfectly intact on a 40 year old car.

But you said .462V. That's fine, but the O2 sensor readout should be "dithering" which is a fancy word for always changing. It should be, .4, .6, .7, .1, .8. .4, ,4. ,7 and so on. Always changing, up, down, typically from 0 to .9, I don't think I've ever seen a full 1.0V. Disconnect the O2 sensor and see what you get, on the sensor side, not the body side. High is rich, low is lean. The body side should be about .45 disconnected, the ECU feeds a signal to it. But that doesn't matter. On the other hand, it should actually run smoother, open loop. Not necessarily better, but smoother, but either way it should run ok and reasonably smooth. But warmed up, really probably even cold, it should be about .5 volts from the O2 sensor, disconnected.

Now, since this all happened after you worked on it and put in the 276 cam, which isn't all that radical, I'd go back and look at anything even vaguely related to that. Check the cam timing, I know you said you did, but check it again, even if you have to pull the upper timing cover. Then if everything is still good, look into the AFM. IIRC, and search, don't take my word, it should be 5 clicks of pretension. Now I'm really, really reaching here, but it's remotely possible that it's been tweaked, and the combo of it and the cam are interacting in such a way as to cause your problem.

[code]

Thanks Mike. It might be possible I was testing the body side of the sensor wiring. I’ll go back and test the other side and will report back. I ran the car with the O2 sensor disconnected for a couple of days recently and didn’t notice a change in the idle at all, even fully warmed up after a few miles of driving in town.

I’ll check the cam timing again. The previous owner had a mechanic who adjusted the AFM mixture, I’ll also go through my binder and confirm when and what he did.

Thanks for your help, I’ll report back with updated info after I check the above.

[code]

Update on the subject of AFM: Air flow meter was replaced about 15k miles ago (4/27/2015). No invoices or notes about the AFM being adjusted since then.

[Chimi-Changa]

You might need a "tune" with the higher duration cam from stock? Yes, the 276 cam is not crazy, however it is higher than stock ( and should be a lumpier idle) and would not be surprised if it needed a tuned chip in the ECU to adjust the fuel/ air to optimize the upgrade in the motor. Installing a 290 cam in my head, even with a tuned chip the idle was way to low and car shakes at idle, etc. ( my idle is set to 1000 rpm now and better) Unfortunately I don't think you can just throw a hotter cam in it and expect everything else stock to run the same way. I would look to find a chipped ECU and give that a try.

Are you sure that's fuel out of the exhaust and not condensation?