Vectra C 1.8 petrol XE engine...idle rough as a bears rump!
Discussion
Vectra C 1.8 model year 8/03, 78K drives well, sound engine, no excess blowby or other nasties.
Idle speed was going up and down, almost stalling but recovering.
Removed the throttle body, found some muck around the throttle plate seat to bore and cleaned with carb cleaner manually moving the throttle plate, also unblocked the 1mm bleed screw in the base of the throttle body.
Now the idle speed is consistant but the idle is rough as the proverbial badgers backside!
Hmm....inspected the system to check for accidently removed pipes causing an airleak, nothing found.
Removed the throttle body again to check for split gasket, nothing found, refitted....still a rough idle.
I had a similar problem with a VW Beetle and did a throttle body re-learn and this cured the issue, so I went into the Vectra ECU with a borrowed Snap-On scanner (my own Gendan 25 can't do this) thinking maybe the throttle body needed a re-learn, only to find this function isn't available on this Vectra...maybe only for later Vectra's?
Now the cars drives fine over 1500 rpm, no stutters, flat spots, just drives well, but under 1500 rpm there is misfire and jerkiness to the drive...similar to a massive airleak, not coilpacks or igition related. I've sprayed all around the engine bay for air leaks with carb cleaner and nothing, the rough but consistant idle remains just that.
OK, I then started checking the live data using my Gendan scanner and found the MAF grams/second too high at around 6 grms/s at idle, I'm expecting this to be 1.6 grms/s or there abouts.
Fitted a new MAF and the results were the same....I didn't think the MAF was at fault as I did the old trick of removing the MAF connecter and the nothing changed. I'm guessing she running so rough it needs this additional air to maintain the idle speed? For Sale MAF used for 3 mins....
Long term fuel trim is...wait for it -100!.....but I'm not getting any codes....so the ECU is trying to lean her off when at the same time the MAF is saying the engine is injesting more air than normal.
Lambda sensor is switching.
EGR valve not stuck open, I removed it and checked.
Pulling the lambda connector makes no difference to the idle (as a limp home default map check) nor does pulling the camshaft sensor, nor pulling the throttle body connector, just gives me a change in idle speed but still with a rough idle....even pulling the air inlet from the throttle body so unmetered air is entering the engine still gves me the same rough idle...WTF!
I listened to the injectors with a screwdriver and could all firing, but pulled the injecter connectors anyway one by one just to check if I had a badly spraying injector or worse. All were firing and it appears I dont have a cylinder down on compression as the rev drop was the same pot to pot?
I checked for loss of earth using my DVM for the MAF and ETC earths (engine temperature coolant) but earth is fine. There was a problem with earlier XE engines ECU's loosing their earth......but then the data would of been nonsense I'm guessing for the MAF and ETC wouldn't it?
So that's it...I got a used throttle body coming tomorrow but I can't see that being the issue somehow...but hey I'm getting desperate....ECU? Don't think so, the cars drives fine when the throttle plate is opened/opening.
any ideas gentlemen?
One thing did occur to me do I need OP-COM to so a throttle body re-learn and thats why the Snap-On can't do this feature?
Idle speed was going up and down, almost stalling but recovering.
Removed the throttle body, found some muck around the throttle plate seat to bore and cleaned with carb cleaner manually moving the throttle plate, also unblocked the 1mm bleed screw in the base of the throttle body.
Now the idle speed is consistant but the idle is rough as the proverbial badgers backside!
Hmm....inspected the system to check for accidently removed pipes causing an airleak, nothing found.
Removed the throttle body again to check for split gasket, nothing found, refitted....still a rough idle.
I had a similar problem with a VW Beetle and did a throttle body re-learn and this cured the issue, so I went into the Vectra ECU with a borrowed Snap-On scanner (my own Gendan 25 can't do this) thinking maybe the throttle body needed a re-learn, only to find this function isn't available on this Vectra...maybe only for later Vectra's?
Now the cars drives fine over 1500 rpm, no stutters, flat spots, just drives well, but under 1500 rpm there is misfire and jerkiness to the drive...similar to a massive airleak, not coilpacks or igition related. I've sprayed all around the engine bay for air leaks with carb cleaner and nothing, the rough but consistant idle remains just that.
OK, I then started checking the live data using my Gendan scanner and found the MAF grams/second too high at around 6 grms/s at idle, I'm expecting this to be 1.6 grms/s or there abouts.
Fitted a new MAF and the results were the same....I didn't think the MAF was at fault as I did the old trick of removing the MAF connecter and the nothing changed. I'm guessing she running so rough it needs this additional air to maintain the idle speed? For Sale MAF used for 3 mins....
Long term fuel trim is...wait for it -100!.....but I'm not getting any codes....so the ECU is trying to lean her off when at the same time the MAF is saying the engine is injesting more air than normal.
Lambda sensor is switching.
EGR valve not stuck open, I removed it and checked.
Pulling the lambda connector makes no difference to the idle (as a limp home default map check) nor does pulling the camshaft sensor, nor pulling the throttle body connector, just gives me a change in idle speed but still with a rough idle....even pulling the air inlet from the throttle body so unmetered air is entering the engine still gves me the same rough idle...WTF!
I listened to the injectors with a screwdriver and could all firing, but pulled the injecter connectors anyway one by one just to check if I had a badly spraying injector or worse. All were firing and it appears I dont have a cylinder down on compression as the rev drop was the same pot to pot?
I checked for loss of earth using my DVM for the MAF and ETC earths (engine temperature coolant) but earth is fine. There was a problem with earlier XE engines ECU's loosing their earth......but then the data would of been nonsense I'm guessing for the MAF and ETC wouldn't it?
So that's it...I got a used throttle body coming tomorrow but I can't see that being the issue somehow...but hey I'm getting desperate....ECU? Don't think so, the cars drives fine when the throttle plate is opened/opening.
any ideas gentlemen?
One thing did occur to me do I need OP-COM to so a throttle body re-learn and thats why the Snap-On can't do this feature?
Edited by iconic on Tuesday 20th October 20:06
Thanks for the reply...this is a tough one.
Yes, those EC'sU have a reputation for breaking down, my only thought on the 5 volt reference voltage for the sensors is that the sensors are giving me data as seen on the scanner...could be corrupted data, but if they lost the 5 volt reference would they not work at all?
I thought the ECU issue was the earth back to the ECU that is shared by the MAF and ETC or this is an issue with Vectra B's....I'm learning here all the time.
I'll check the 5 volt reference whilst the engine is at idle just to be sure the additional vibration at idle isn't causing an issue with the reference line.
cheers for the reply
Yes, those EC'sU have a reputation for breaking down, my only thought on the 5 volt reference voltage for the sensors is that the sensors are giving me data as seen on the scanner...could be corrupted data, but if they lost the 5 volt reference would they not work at all?
I thought the ECU issue was the earth back to the ECU that is shared by the MAF and ETC or this is an issue with Vectra B's....I'm learning here all the time.
I'll check the 5 volt reference whilst the engine is at idle just to be sure the additional vibration at idle isn't causing an issue with the reference line.
cheers for the reply
Ahh, sorry what I meant by Lambda switching is that it is responding correctly to a simulated rich condition.
At idle it shows zero volts or nearly - yes its that lean! - when I inject some propane into the air inlet I see the volts rise in reponse....also same trick with carb cleaner...engine runs smooth for a while too so it really is running lean...also tried a known good lambda sensor but no change to values.
Wierd problem, lambda sensor saying help I'm lean but engine not doing much to help..this is in closed loop. Drives fine wide open throttle or accelerating but then it going to in open loop for some of this.
Last gasp checks today as I can't find any vaccum leaks externally is a to block brake booster inlet to see if there a leak here.
Also fuel filter looks very old maybe restricting fuel flow...but then common sense would say this would effect wide open throttle more than idle?
Strange one....all sensors giving good data back, MAF, ETC, IAT...bar the injection duration as I don't know what to expect here..although it is changing under demand.
I can't see any data value displayed for fuel pressure which is frustrating as this is a good check for fuel system integrity, my system will not show it, if indeed its reported back...should be?
At idle it shows zero volts or nearly - yes its that lean! - when I inject some propane into the air inlet I see the volts rise in reponse....also same trick with carb cleaner...engine runs smooth for a while too so it really is running lean...also tried a known good lambda sensor but no change to values.
Wierd problem, lambda sensor saying help I'm lean but engine not doing much to help..this is in closed loop. Drives fine wide open throttle or accelerating but then it going to in open loop for some of this.
Last gasp checks today as I can't find any vaccum leaks externally is a to block brake booster inlet to see if there a leak here.
Also fuel filter looks very old maybe restricting fuel flow...but then common sense would say this would effect wide open throttle more than idle?
Strange one....all sensors giving good data back, MAF, ETC, IAT...bar the injection duration as I don't know what to expect here..although it is changing under demand.
I can't see any data value displayed for fuel pressure which is frustrating as this is a good check for fuel system integrity, my system will not show it, if indeed its reported back...should be?
Ahh...missed the fact that these cars are non return fuel systems, hence fuel pressure reg' is part of the fuel pump and run at constant fuel pressure...doh!
Blocked off the EGR circuit and brake booster and no difference.
From what I'm seeing my guess is low fuel rail pressure. Route now will be new filter when my filter removing pliers turn up, then test power supply and ground to fuel pump, then current draw, then a new pump.
Blocked off the EGR circuit and brake booster and no difference.
From what I'm seeing my guess is low fuel rail pressure. Route now will be new filter when my filter removing pliers turn up, then test power supply and ground to fuel pump, then current draw, then a new pump.
Edited by iconic on Tuesday 27th October 15:37
Edited by iconic on Tuesday 27th October 15:38
Gassing Station | Home Mechanics | Top of Page | What's New | My Stuff