'99 2.8 Roadster not running right (details inside)

[colb]

Thats annoying not being able to find a used working Maf as you say it would be nice to confirm yours is bad before investing in a oem replacement. As you have tried cleaning it a few times I would think its definitely bad. Mafs don't always set a specific code indicating a fault with it more likely to be other sensors reporting codes because its gone bad. A bad maf will affect the exhaust sensors as the Maf is not reporting the correct air flow so that causes the ecu to richen or lean the fuel settings dependent on what the Maf is reporting. Wrong fuel settings then cause codes to be set by the exhaust sensors. If you can view live data with a diagnostic system you should see what the Maf is doing, it should be registering how many gallons of air are flowing through the maf at different throttle openings it will at least show you that its working. Revving the engine you should see in excess of 20 galls being reported by the Maf as far as I remember with mine. Someone on another forum has similar problems with his Maf, his barely reads 3 as his Maf is dead. With live data diagnostics you would also be able to see the exhaust sensors and if they are cycling their voltages, you should see them rising and falling as they do their job.
Diagnostics are all well and good but some lateral thinking is required when trying to work out why codes have been triggered. What you see is not necessarily the root cause. Carbs are simple to diagnose compared to electronic systems.

[deni2s]

Thats annoying not being able to find a used working Maf as you say it would be nice to confirm yours is bad before investing in a oem replacement. As you have tried cleaning it a few times I would think its definitely bad. Mafs don't always set a specific code indicating a fault with it more likely to be other sensors reporting codes because its gone bad. A bad maf will affect the exhaust sensors as the Maf is not reporting the correct air flow so that causes the ecu to richen or lean the fuel settings dependent on what the Maf is reporting. Wrong fuel settings then cause codes to be set by the exhaust sensors. If you can view live data with a diagnostic system you should see what the Maf is doing, it should be registering how many gallons of air are flowing through the maf at different throttle openings it will at least show you that its working. Revving the engine you should see in excess of 20 galls being reported by the Maf as far as I remember with mine. Someone on another forum has similar problems with his Maf, his barely reads 3 as his Maf is dead. With live data diagnostics you would also be able to see the exhaust sensors and if they are cycling their voltages, you should see them rising and falling as they do their job.
Diagnostics are all well and good but some lateral thinking is required when trying to work out why codes have been triggered. What you see is not necessarily the root cause. Carbs are simple to diagnose compared to electronic systems.

Failed MAF shouldn't throw camshaft sensor DTC which might be the root cause.

[328 Power 04]

I'll see when I can do it then.

Just keep in mind that 1) I'm used to carbs, not all these computers, and 2) I'm hesitant to just replace a presumably expensive sensor if one bad sensor can cause others to show as faulty. So essentially I don't fully trust the diagnostic printouts.

It's all good. The codes don't lie, they always indicate a problem. They are signs and symptoms to be used to formulate a proof based diagnosis. Rarely, they even tell you exactly what needs to be changed. A camshaft position sensor signal code is one of those. Start there.

[LannVouivre]

Leave the MAF alone. Listen to Abel. Don't talk to that mechanic you usually go to. A bad cam sensor can cause driveability issues besides hard or no start. I still wouldn't rule out a stretched timing chain, but I'd look at and probably replace the cam sensor before I touched the chain just because the timing chains are the opposite of fun.

[JKuhn]

I'm planning to do the cam sensor. I just want to test the circuit first, I already looked at a video on how to do that (and how to replace the sensor).

Speaking of which, I have a couple questions:
1. How expensive is a cam sensor?
2. Which makes can I go for?
3. If there's a significant price difference (and it makes sense financially due to the price of a new one), should I just get a sensor from a scrapyard M52/M54 (I'll obviously do a quick test before buying)?

[deni2s]

I'm planning to do the cam sensor. I just want to test the circuit first, I already looked at a video on how to do that (and how to replace the sensor).

Speaking of which, I have a couple questions:
1. How expensive is a cam sensor?
2. Which makes can I go for?
3. If there's a significant price difference (and it makes sense financially due to the price of a new one), should I just get a sensor from a scrapyard M52/M54 (I'll obviously do a quick test before buying)?

1. expensive
2. oem
3. yes, sounds like a wise thing to do.

Testing the circuit can tell you if it's bad for sure, but it will not tell you, if it's good for sure. And, if I remember right, Bentley manual has wrong Ohms measurement written causing lots of misguidance on internet.

- - - Updated - - -

And, if I remember right, Bentley manual has wrong Ohms measurement written causing lots of misguidance on internet.

Oh, checked that was for crankshaft sensor. camshaft sensor resistance should be 1280 +/-10% ohms.

[JKuhn]

But what make? I see the one in there is a Topran.

And if I check on from a scrapyard, what should I check? As I inderstand if it works I'll see the reading change on a multimeter if a wrench or something similar passes before it. Unless the scrapyard people are nice enough to help (I don't currently have a 32mm for the VANOS solenoid), then it can even be installed on the spot.

[328 Power 04]

You should use BMW original or Siemens/VDO sensors only. Aftermarket can cause problems.


On the sensor itself, you should pull back the rubber insulation/cover on the plug end, to make sure wires underneath are not stripped of their insulation and shorting wires underneath. I have seen this a few times in our era cars.

[JKuhn]

On the connecter side? They looked fine to me, but I can check again tomorrow.

[JKuhn]

UPDATE:

The scrapyards were of no help. One said they don't have those engines, the other one said that the sensor is part of the engine, and that I can't buy the sensor seperately. So, since I don't currently have what I need to test the circuit, I decided to buy what I could find (Topran, I know it's not OEM but I don't have lots of cash laying around, and the shop said it's not bad). Anyway, before the car ran consistently bad. Now it's much more inconsistent, so it seems to have brought an improvement. Next will be to test the circuit.

If it's still a bad sensor then at least this one will help a bit while I save for OEM.

[328 Power 04]

With this sensor in, I would have codes cleared and see what other codes come up in a few days driving. Sensor may be fixed, and something else needs looking at.

[JKuhn]

Makes sense. If I can't find an affordable (but not too cheapskate) multimeter (I can't believe how expensive those things got!) I might decide to look for a code scanner. Assuming the shop is open tomorrow. Also, it depends on the budget.

[328 Power 04]

If you have a windows laptop, and are feeling a bit braver, I can guide you to read your own codes, cheaply. It would give you a huge advantage over taking it to the shop all the time.

[JKuhn]

But then I'll need to find and buy the correct cable and adapter (and of course get the software). So it'll probably cost more than a normal code reader (and adapter).

[BimmerBreaker]

Owning and maintaining a BMW costs money

[JKuhn]

I'm aware of that. But at the moment I need to be extra careful.

[ntinkle]

Quickest way to kill the enjoyment of a hobby is to over spend. Good luck.

[BimmerBreaker]

Blindly parts swapping is a lot more expensive than getting a proper diagnostic setup, spend the money upfront for the cables (they really aren't that expensive) and you'll save money in the long-run

What's the saying - there is nothing more expensive than a cheap BMW?

[deni2s]

But then I'll need to find and buy the correct cable and adapter (and of course get the software). So it'll probably cost more than a normal code reader (and adapter).

Bullshit. Cable + adapter costs like 20$, software is free and it’s less limited than “normal code reader”. People should understand that code readers sucks compared to cable+laptop config.

- - - Updated - - -

Seems like you wasted money on that Topran sensor, as you are not 100% sure it’s working perfectly...

[LannVouivre]

He's in South Africa, it might indeed have atrocious shipping costs attached. However, it would be the only reader he ever needs for an oldish BMW.

[JKuhn]

It might be true that the laptop setup will be better, but as stated I'm in South Africa. The products that were previously suggested to me don't even ship here. I'm also currently still trying to get back on my feet after the initial expenses (tyres, a less deafening silencer box, initial service as it had absolutely 0 recent history and of course the expense of buying the car). The thing is though, I haven't had the chance yet. By the way, importing something takes about a month even with expensive shipping. Then there's the exchange rate (for the product and shippping) and import tax. So basically I need to buy local, and for that I need to find it first.


Also, I'm really starting to suspect the MAF now. I did some more testing, and it clearly runs smoother without it (but then it has even less low-end power).

EDIT: Oh, and to top it all off, the VW (in my sig) broke my wallet even before I bought the Z3. I had to drive it a lot (it had a way of not starting if I didn't drive it for 3 days), and it gave me 5.something km/l on a good day. And it had the reliability of a recently brought back 47 year old car.

[deni2s]

I have spare brand new, never fitted camshaft position sensor by Meyle (Germany). If interested I can send it to you for 15 euros + shipping costs. You can also order cable/adapter to my address (Finland - faster and more expensive or Latvia - cheaper but might be not so fast) and I can send all together to you, so you will not have to pay some import taxes. Just PM me if interested before 2018. You might also want to consider ordering OBD2 wifi thing which you can use together with torque app on your mobile.

[JKuhn]

Thanks for the offer. Unfortunately I'm not sure if that's possible. I don't think it's the sensor though, but of course I could be wrong. Also, regarding the WiFi adapter, that will certainly be nice to have, but I use Windows Phone 8.1, so I'm very limited regarding apps (I'm not planning to go back to Android, and I'm not giving money to Apple).

[JKuhn]

UPDATE: I went to a local electronics store, and while I was there they helped me to test the circuit (I had to do the probing, the person there said he doesn't work on cars). I didn't see the readings myself, but the ground wire works (continuity test), and the others were 5V and 12V in that order. So does that mean the circuit is fine?

Oh, and another thing. I assume it won't be needed, but just to be sure, might it help if I reset the DME (battery disconnect)?

[deni2s]

Oh, and another thing. I assume it won't be needed, but just to be sure, might it help if I reset the DME (battery disconnect)?

No, there is no such thing as DME resetting (by battery disconnect).

- - - Updated - - -

UPDATE: I went to a local electronics store, and while I was there they helped me to test the circuit (I had to do the probing, the person there said he doesn't work on cars). I didn't see the readings myself, but the ground wire works (continuity test), and the others were 5V and 12V in that order. So does that mean the circuit is fine?

I have no clue what are you doing and why. Replace camshaft sensor, clear the codes, read the codes. Move to next suspect if car still doesn't run fine.

If you wanted to test camshaft sensor, you should check ohms, not volts.