Intermitent missfire on all cylinders
Print Topic | Close Window

By Andy4k6 - Friday, August 09, 2019 5:56:34 AM
Hi Everyone,

Sorry this post is going to be long and thank you for reading,

I am having trouble with my own car, I'm an ex tech and don't have access to a full workshop any more although I have some time at a friend’s shop this weekend and I am looking for ideas to help me trace the fault.

The car is a 2006 Saab 9-3 2.8T V6 fitted with a Bosch ME9.6 management system.

I had this issue start around 2-3 months ago, the car has a rough idle most but not all of the time, the RPM's are fairly stable and not hunting but it is not smooth, altering the load on the engine sometimes makes a difference to the rough idle (i.e turning the a/c on or off, sometimes being off or on will make it idle better but there is no rhyme or reason to which will make it better if at all).

In addition to the rough idle I have a misfire which appears mostly under moderate load below 2500 RPM, this is most apparent in 3rd gear and above, accelerating normally within 1st and 2nd rarely show the fault although the fault does appear to change daily and it can be mild one day and much worse on the next, somedays the fault is barely there and the car has a smooth idle.

On a cold start when the SAI is active and the rpm's are around 1500, if I allow the engine to idle with no input it runs with a partial miss and 50%of the time will throw a misfire code, the code can be any of the cylinders, no cylinder is more common, other than at this time it almost never throws a code for a misfire even though I can create clear full misfires by accelerating fairly hard between 1500 and 2500 RPM in a higher gear, at full throttle the misfires do not seem as severe and once above 3000 RPM the car almost never misses a beat and when provoked goes like a scalded cat.

So far I have done and checked the following:

Engine was replaced 2 years ago for a separate issue, the engine fitted was a 60k engine which I replaced all the timing chains on at the time, the engine has covered 20k since.

The spark plugs were nearing change, so all of the plugs were replaced - no change.

As the coil packs are common failures on this engine I have recently replaced all 6 coil packs - no change

The air mass was looking suspect at idle and has been replaced with a brand new genuine sensor which is now reading correctly at idle - no change

Up stream wide band lambda was replaced as it had some white contamination on it, replaced with a genuine Bosch - no change

Evap valve replaced, this stuck open when being actuated and was replaced with a genuine

Boost pressure sensor swapped with a spare I had - no change

Intake system smoke tested, no visible leaks found,

Prior to the air mass and lambda replacement an emissions test was done, it was a fraction weak but only a fraction (this may be down to my friends tester that is not in regular use or serviced).

Vacuum seems a touch low reading 18 in-hg at idle with the a/c etc off, live data reports this as being -22 - -24 ish though I am not aware of any sensor for the vacuum, there is only a boost / temp sensor which is located before the throttle flap and not subject to manifold vacuum.

The only fault codes raised are P0300 - p0306 (random multiple miss and all 6 cylinders), live data and fuel trims etc appear to be good (except for note above on vacuum).

Suggestionson what to check next would be appreciated, I know some of you may suggest me to take it to a well-equipped shop who is far more up to speed than I am currently however, I am not in a position to be able to do this at the moment and I have to try and sort it myself, plus I hate being beaten and having to pay someone to do something I should be able to do.

I have plans at the weekend to carry out another smoke check and also try running the engine with the vacuum, evap etc blocked off to check for excessive airflow this way, I'm going to check the wiring for the coil packs and see if I am suffering from a bad wire causing reduced power to the coil packs, I have considered checking the CPS however I only have a basic single channel scope and i'm not too confident with it.

Many thanks for your time


By Robski - Friday, August 09, 2019 6:45:55 AM
scan tool used ?

check TSB's for a known fix to a similar fault !
By K And R - Friday, August 09, 2019 7:46:13 AM


By Andy4k6 - Friday, August 09, 2019 7:49:05 AM
Hi Robski,

I have a Tech 2 with saab 148 card, I also have access to an Autel MS906, neither tool gives me anything to work on Crying I have also checked it via OBD, and also using a snap on solus ultra and op-com as a Vauxhall Vectra, still nothing :-(

I have checked the Saab tsb's but have found nothing, I don't have access to any other TSB's from other manufacturers, the same engine is used in the Vectra VXR, Cadillac BLS and some Holden vehicles which I'm not sure on the models.

Many thanks

By K And R - Friday, August 09, 2019 8:35:11 AM

"live data and fuel trims etc appear to begood (except for note above on vacuum)." !!

sure i would double/ treble checked that before throwing lot of bits at it.

By K And R - Friday, August 09, 2019 8:35:13 AM

"live data and fuel trims etc appear to begood (except for note above on vacuum)." !!

sure i would double/ treble checked that before throwing lot of bits at it.

By Andy4k6 - Friday, August 09, 2019 9:43:06 AM
I can only assume the vacuum reading is calculated as I can see no way the system can measure the pressure inside the inlet, it has no map sensor only a barometric and boost sensors and despite trying to work it out I can't see any reason this calculation should be out, I am guessing it's calculated via maf, intake temp, rpm and evap position etc.

I am confident there are no leaks in the system although I intend to completely go through the evap, brake booster and pcv again this weekend even though I have checked and found nothing before, the issue also happens under boost so it's not entirely vacuum and there are no leaks carrying out a smoke test, I appreciate that it is not as easy to rule out a leak coming in via a one way valve that is closed when carrying out a smoke test.

Many thanks for your time

By K And R - Friday, August 09, 2019 10:03:34 AM

So what is the VAC using a gauge ( barring in mind your thread is not easy to read, far too many missing space bar commands)

By Andy4k6 - Friday, August 09, 2019 10:39:29 AM
Please accept my apology, I didn't check my original post when I posted it and I had no idea it had gone to bits.

Thank you for taking the time to read and reply to that dreadful unedited post (I have now edited it).

I have measured the vacuum with a standard vacuum gauge taken directly off the manifold, the ECU reports vacuum and boost within the same data line, I can only assume this is a calculated figure.

Many thanks
By Timbo - Friday, August 09, 2019 12:55:42 PM
Yeah, and what is actual inlet manifold vacuum at idle, cold and hot, and at 2000rpm or so?

And what are STFT for both banks at idle and fast idle ?
By Andy4k6 - Friday, August 09, 2019 1:51:21 PM
If I remember correctly but the long term fuel trim was +5% at idle but it was about a month ago when I last checked so I'll double check it all tomorrow and confirm.

Actual vacuum at the manifold is approximately 16-18 in-HG at idle with everything off and at 2k it's 19-20 in -hg

Many thanks

By K And R - Friday, August 09, 2019 6:07:38 PM
"Actual vacuum at the manifold is approximately 16-18 in-HG at idle" which is on the low side, is it a stable reading ?


By Timbo - Saturday, August 10, 2019 5:17:29 AM
Get your propane torch or wd40 around the intake manifold gaskets.
By Andy4k6 - Sunday, August 11, 2019 11:49:41 AM
Hi Guys,

Vacuum reading is stable at idle, it's not bouncing round.

Already had brake cleaner around the manifold, no leaks found.

I did another smoke check on the manifold there is a very very tiny leak of 0.010 thou however there is no visable signs of any leaks, I have removed every line to the engine and blocked it off so I can only assume it's making it's way out the exhaust.