My poor zed (2.0i) dropped into limp mode :(

[Jason Goldschmidt]

Hello all,
My zed has been giving me some trouble lately. The story goes...

-On a couple of occasions now, it has aggressively lost power and gone into limp mode. The engine idles VERY lumpy until restarted. I pulled the codes from it last time (with my cheapie code reader) and got P1170.
-Pulling coil packs one by one indicates that it appears to be running on two cylinders at that point (cylinders 2 and 3)! Some sort of problem with bank 2 cylinders?
-Reading the codes today with INPA showed nothing.
-If i switch the engine off and start again the issue always clears itself but will return almost immediately when the engine is revved.
-Unplugging the MAF sensor allows me to drive home.
-When I then plug the MAF in again and drive the car as usual, the issue might not return for weeks.

Whats going on here folks!!
Could it be a cracked manifold? I looked today but am not sure where they usually crack?

Thanks in advance all.

 

[t-tony]

You really need to thoroughly check the large rubber intake “boots” for splits, you have to part the lumps with your fingers or even remove the boots.

Tony.

 

[Jason Goldschmidt]

You really need to thoroughly check the large rubber intake “boots” for splits, you have to part the lumps with your fingers or even remove the boots.

Tony.

Thank you for the reply Tony.

I forgot to mention that the boot has been removed and thoroughly inspected.
I intend to smoke test it soon anyway, just to be sure.

It has also had new pre-cat O2 sensors recently.

The car is regularly loosing power, hesitating and the fuel trims are all over the place!
There is quite a mismatch between banks 1 and 2.

 

[NZ00Z3]

By going into limp mode, do you mean that the yellow transmission warning light "cog of death" has come on, you're limited to 3rd gear etc? If so, read the trouble codes in the transmission computer (EGS).

Where the leads on the pre-cat O2's plugged into the correct sockets? If they get switched, you get crazy fuel trim readings and misfires etc. One bank will goes full lean while the other goes full rich.

Couldn't find P1170 in my Z3 DME reference, so can't help you with the code number.

One way to check for vacuum leaks is to do this test.
With a hot engine at idle, look at your fuel trims, 2 long term and 2 short term. If any are >8% then you likely have a vacuum leak. Rev the engine to 3,000 rpm for 30 seconds and watch the short term fuel trims. If they reduce significantly, might even go negative, then you have vacuum leaks to find.

 

[Eddie Zedder]

Sounds similar to what I had once after replacing my vanos seals.

The inlet vanos solenoid shuttle valve was sticking, this caused all sorts of misfire issues.

Easy fix, remove the solenoid and behind it you will find the valve and a spring. Give it a good clean with fine Emery cloth, lubricate and replace making sure it travels freely in the bore.

 

[Jason Goldschmidt]

By going into limp mode, do you mean that the yellow transmission warning light "cog of death" has come on, you're limited to 3rd gear etc?

Thanks for the detailed reply Murray.
This is a manual gearbox so I assume that does not apply? The limp mode I am experiencing is limiting the engine speed to 2500ish rpm and greatly reducing power.
For some reason if seems to only be firing on two cylinders in this state.

Where the leads on the pre-cat O2's plugged into the correct sockets?

I did unplug them one at a time when swapping them so they couldn't have been switched. I will check the connectors are seated and plugged into the correct sockets.

Couldn't find P1170 in my Z3 DME reference,

I know it is a very generic code but I think I have heard some other Z4 owners experiencing it. Normally due to a cracked exhaust manifold.
Maybe it does not apply to the z3?
I did check over the manifold and it doesn't appear to be blowing anywhere.

 

[Jason Goldschmidt]

Easy fix, remove the solenoid and behind it you will find the valve and a spring. Give it a good clean with fine Emery cloth, lubricate and replace making sure it travels freely in the bore.

Hi Eddie, thanks for the reply.
Does this apply to the N46 engine?
When I removed the solenoids, there didn't seem to be any valve or moving parts behind them.

 

[t-tony]

The Z4 Exhaust manifold cracking issue is usually on the 4 cylinder engines.

Tony.

 

[Jason Goldschmidt]

The Z4 Exhaust manifold cracking issue is usually on the 4 cylinder

Hi Tony.

The N46 in the 2 litre zeds is a 4 pot.

 

[t-tony]

Wasn’t sure of the N46 designation Jason, but yes we had this on 2 of these cars where I used to work. First one the manifold was removed and welded (it cracked again almost straight away) it was then replaced with a brand new one which was very expensive if I recall correctly. The second car owner was recommended to fit a new one first off. Can you hear a “blow” on your car?

Tony.

 

[Eddie Zedder]

Hi Eddie, thanks for the reply.
Does this apply to the N46 engine?
When I removed the solenoids, there didn't seem to be any valve or moving parts behind them.

Sorry, assumed you had a Z3. Don't know owt about Z4's.

 

[Jason Goldschmidt]

Wasn’t sure of the N46 designation Jason, but yes we had this on 2 of these cars where I used to work.

Ah ok, fair enough.
I can't actually hear a blow. I did inspect the welds with a mirror and couldn't see any cracked welds.

I also hooked a shop vac up to the exhaust with the engine off and couldn't hear any hissing when using a piece of tube as a stethoscope around the exhaust manifold.

I gather a new manifold is very expensive. I am no longer suspecting a cracked manifold so much though. I assume it would have been obvious when inspected thoroughly.

The rough running data and fuel trim for bank 2 cylinders seems to point to an issue on that bank.

I can try swapping O2 sensors between banks but both are new.

Strange thing is the engine seems to run great most of the time and then throw a little fit!

Very confused at this point

 

[Jason Goldschmidt]

Out of interest Tony, do you know of any competent specialists near Northampton that might be able to help if I can't solve this one?

 

[t-tony]

Sorry Jason, can’t help with anyone based down there mate, I’m near Lincoln. Might be worth asking on the open forum or even on Z4forum if you’re on there? I think you would have found a leak with the vacuum test, and yes the manifolds were a lot of money. The original fault sounds just like my old Z3 2.0 did when the inlet camshaft position sensor was playing up, the car would go into limp home mode but not throw the EML light on or record a fault code. After clearing any codes it may be ok for a few weeks. Is the P1170 code the only one that comes back?

Tony.

 

[Jason Goldschmidt]

The original fault sounds just like my old Z3 2.0 did when the inlet camshaft position sensor was playing up, the car would go into limp home mode but not throw the EML light on or record a fault code.

I did actually suspect the cam sensors initially but didn't want to blindly replace them.
I forgot to mention that it did once stall when idling at a junction which is quite symptomatic of cam sensor issues.

I did pull a different code from it the first time it happened but have lost the paper I wrote the code on! I will hopefully be able to get my laptop plugged in if it happens again.
If I find the paper I will post the other code.

I swapped the cam sensors around to see if that made a difference but the issue has occured twice now with them swapped and in their initial locations.

Did you replace the sensor with genuine? They are much cheaper as aftermarket but probably for a reason!

I did actually seek some help on zforum but I didn't really get much response.
I have more information on the issue now so maybe I will try again

 

[t-tony]

To be frank I find that Z4Forum is not as helpful as it could be, which is a shame because we all need some help now and again. I bought a sensor for my Z3 from ECP, branded as Febi I think, but, it was actually a Bosch part with the Bosch logo removed. The original part number was still there and when Googled it came up as a genuine Bosch part. About half the cost from BMW. Dn not buycheap eBay etc. sensors.

Tony.

 

[Jason Goldschmidt]

To be frank I find that Z4Forum is not as helpful as it could be, which is a shame because we all need some help now and again.

Unfortunately that has also been my experience. I find there are more members there who are keen to take their car to a dealership or indy than do their own diagnostics.
Nothing wrong with that. My pockets just aren't that deep, so I need a place to talk mechanics stuff sometimes!

Dn not buycheap eBay etc. sensors.

Good advice.
I will try and get some live data from the cam sensors and consider replacing if the readings are not consistent. I seem to remember that one was reporting intermittent data in INPA but I am not sure my installation of INPA is all that sound.
Maybe I will update my software first.

 

[Jason Goldschmidt]

Right, Im back!

I have had a play with the car today. Got a few bits done that I have been putting off for awhile; replacing the cam cover gasket was the main task for today.
-Whilst I was at it, I tested both camshaft sensors with an oscilloscope. Both seem to be functioning correctly so I think they can be ruled out.
-The timing chain and guides look like they have been replaced by someone in the past, no issues there. Engine looks to be timed correctly.
-I ran diagnostic functions on the vanos system using ISTA. All seems to be working properly. Solenoids are nice and clean.

I went out for a good hoon once the RTV on the cover had gone off and the car hit limp mode about half a mile down the road. Not happy!
I scanned it for codes, realised I hadn't reset the valvetronic positions after removing the motor, thought that must have been the reason.
Apparently not. I have a couple of codes present relating the the DSC unit.
5E19
5E1A
and 6121 which is related the the engine speed (its an EPS code). I have no idea that engine speed really has to do with the electronic power steering!

So I have done some digging and there seem to be suggestions that the codes could relate to rate sensors such as yaw rate, latitudinal acceleration or longitudinal acceleration. I am worried that they actually mean that the DSC unit is on its way out!
I re-calibrated the lat and long sensors using INPA as well as telling the valvetronic motor to learn its limit positions. Hopefully that sorts it.

I did notice the yaw rate and lat. acceleration do not sit at zero when stationary. Yaw is reading 0.3 degrees/second and lat is reading 0.03g (If i remember correctly). I am not sure if there is any tolerance on these values or if they should truly read zero when stationary.

I will cross my finger that the issue doesn't return and post back.
If anyone has any other suggestions, please let me know.

 

[Jason Goldschmidt]

And a quick pic of an N46 valve train.
Lip all cleaned up ready for the new gasket.
Complicated little buggers compared to anything else I have worked on.

 

[Jason Goldschmidt]

Ok, no dice.

Calibrating the sensors did not solve the issue.
It did, however, stop other codes returning, so they did need calibrating after a previous battery dissconection.
The important code that does keep popping up is 5E19.

5E19 relates to a communication error between the DSC and DME modules. Specifically, it is an error that the DSC throws when it asks the engine to reduce torque but the DME does not respond fast enough.

There are a number of other threads floating around with similar problems. Many suggest that the fly-by-wire throttle body has suffered a fault and is no longer controlling engine torque adequately when the DSC (traction control) kicks in (which it does rather a lot when I'm driving ).

So I jumped on eBay and managed to grab an entire spare inlet manifold including throttle body, map sensor and variable intake actuator for 30 quid. Bargain!

Swapped out the TB this morning and have not had the code return after a good few runs. In fact, the car idles better and throttle response has improved.

I have got my fingers crossed! I will report back in a few days (hopefully with no codes!).
If the codes return, next step is to drive it off the tallest cliff I can find.