Log in

View Full Version : Fuel hunting idle when cold / running rough



bbb
12-21-2015, 09:21 PM
have had the car for a couple of years without a lot of use. it is a auto with turbo fitted.
problem is that it always has a hunting idle when cold. fluctuates from about 500rpm to 2000rpm (when in park) it is better when in drive.
it also runs slightly rough when driving. I have had my mechanic look at he says that it is running lean. trying to avoid paying him for hours of troubleshooting / testing when I can do some of it myself. but am not having much luck.
I have replaced the warm up regulator, and have tried a known working idle control solenoid, and idle ecu
the frequency valve seems to be working and checking co with a dwell meter, I am getting a reading of 45 when cold and about 12 when warmed up (although this warmed up reading could be intermittent - I also got 45 with car warmed up)
looking for any advise or help on what to look at next.

Bitsyncmaster
12-22-2015, 03:41 AM
The engine should run very rich when you first start it cold if your mixture screw is adjusted correctly. Before the O2 sensor has warmed up you should see a steady 45 deg dwell. It should be 55 deg if engine is below 60 deg. F. As the engine warms up the warmup regulator starts leaning the engine toward normal mixture. When the O2 sensor is hot then the dwell will start swinging because the lambda system is holding the mixture for 14.7 AFR.

So the AFR you get as the engine warms up is dependent on the calibration of the WUR, Fuel pump primary pressure and mixture adjustment.

You start adjusting the mixture with a hot engine so your dwell reading is averaging 40 deg. Your hunting may not be caused by improper engine mixture. Everything needs to be working as designed to get these PRVs to idle smoothly.

David T
12-22-2015, 09:22 AM
Do you have 1 or 2 turbos? With all of the extra plumbing there are a lot more chances for vacuum leaks so that is one of the first things to check. Turbos also have pressure switches (Hobbs switches) to modify things when the turbos pressurize the intake. They all need to be hooked up and operating properly. Do you have a boost gauge and an A/F gauge? What are the readings? At idle the car should operate like an ordinary car with a steady idle at the proper A/F ratio.

PJ Grady Inc.
12-22-2015, 09:46 AM
Do you have 1 or 2 turbos? With all of the extra plumbing there are a lot more chances for vacuum leaks so that is one of the first things to check. Turbos also have pressure switches (Hobbs switches) to modify things when the turbos pressurize the intake. They all need to be hooked up and operating properly. Do you have a boost gauge and an A/F gauge? What are the readings? At idle the car should operate like an ordinary car with a steady idle at the proper A/F ratio.
The twin "Island" turbo setup needs the CO set up to 5% instead of the normal 1% on a stock car. This is primarily due to the increased base timing advance. Otherwise you will having excessive surging and a rough idle.
Rob

bbb
12-22-2015, 04:05 PM
thanks for the replies guys,
BAE single turbo. yes do have a boost gauge.
I suspect a vacuum leak, but have been unable to find one yet.
also not discounting that it may be some sort of electrical problem
Don't want to start making mixture adjustments until I am sure that it is not a mechanical problem.

David T
12-23-2015, 08:56 PM
The easiest and fastest way to find vacuum leaks is to use a smoke machine. Besides all of the obvious connections for the turbo, some of the more common leaks are:
Injector seals
"O" ring for the air pipe in the bottom of the mixture unit for the idle motor
Missing plug over the mixture screw
You are right in trying to find and fix any vacuum leaks BEFORE attempting to make any adjustments. No amount of adjustment can compensate for vacuum leaks. Using a smoke machine I have found some very hard to find leaks including where a bolt went into a core plug on the front right side of the intake manifiold (it was TOO long!). A missing "O" ring on the cold start valve, a hose fell off the vacuum canister, Splits in vacuum hoses, the vacuum solenoid hooked up wrong, and bolt holes that were drilled through on the intake manifold and leaking. Trying to find leaks by using Propane or some other volatile fuel to see if the idle changes is difficult and dangerous.

bbb
12-25-2015, 06:09 AM
have changed the air pipe o-ring in the bottom of the mixture unit.
afterwards on cold start-up the idle held steady for about 15 seconds then started hunting again. so I think that I am on the right track.
If I hold the throttle very slightly open, about 900-1000rpm, the idle hold steady. which leads me to think that it is overcompensating from a too low an idle. (idle adjustment? or mixture adjustment?)
when warmed up with a multimeter connected to read the FV duty cycle, I get a reading of between 12-20, which if I understand correctly from other threads, means the the frequency valve is trying to lean a running rich situation. I got change in reading when I adjusted the mixture, so not sure if I am reading it right (returned mixture to original setting - so have not actually changed it yet).

Bitsyncmaster
12-25-2015, 06:20 AM
have changed the air pipe o-ring in the bottom of the mixture unit.
afterwards on cold start-up the idle held steady for about 15 seconds then started hunting again. so I think that I am on the right track.
If I hold the throttle very slightly open, about 900-1000rpm, the idle hold steady. which leads me to think that it is overcompensating from a too low an idle. (idle adjustment? or mixture adjustment?)
when warmed up with a multimeter connected to read the FV duty cycle, I get a reading of between 12-20, which if I understand correctly from other threads, means the the frequency valve is trying to lean a running rich situation. I got change in reading when I adjusted the mixture, so not sure if I am reading it right (returned mixture to original setting - so have not actually changed it yet).

You can verify your meter is reading correctly by pressing the wide open throttle switch. When pressed it should read 60% duty cycle or 55 deg. dwell.

bbb
12-27-2015, 09:49 PM
have manged to make improvements.
warm up is much better, and goes to closed loop and reading duty cycle on multimeter shows about 48. but after about a minute, starts running rougher and duty shows about 13. and no longer hear the frequency valve working.
so suspect that may have a faulty frequency valve. (it is a new oxygen sensor, would it have failed so quickly?) is there a way to clean or bench test the frequency valve itself?

DMCMW Dave
12-27-2015, 10:24 PM
have manged to make improvements.
warm up is much better, and goes to closed loop and reading duty cycle on multimeter shows about 48. but after about a minute, starts running rougher and duty shows about 13. and no longer hear the frequency valve working.
so suspect that may have a faulty frequency valve. (it is a new oxygen sensor, would it have failed so quickly?) is there a way to clean or bench test the frequency valve itself?

It's not the frequency valve. Going silent just means that it's all the way closed. 45-48 is open loop. When it goes to 13 the computer is trying to lean it out all the way. Have you adjusted the CO screw?

Bitsyncmaster
12-28-2015, 03:50 AM
have manged to make improvements.
warm up is much better, and goes to closed loop and reading duty cycle on multimeter shows about 48. but after about a minute, starts running rougher and duty shows about 13. and no longer hear the frequency valve working.
so suspect that may have a faulty frequency valve. (it is a new oxygen sensor, would it have failed so quickly?) is there a way to clean or bench test the frequency valve itself?

You need to adjust the mixture on a warmed up (hot) engine. The dwell or duty cycle should be swinging up and down (hunting) on the warmed up engine. Dwell should be set for 40 deg. average, duty cycle would be set for 44% average.

bbb
12-28-2015, 06:21 AM
put the old oxygen sensor back in.
now duty cycle show shows movement up and down, always seem to correct to a duty cycle reading of about 49
now I will try adjusting mixture. should only need a slight adjustment
looks like a small vacuum leak at the air pipe and a bad 02 sensor have been my problem (would the air leak made the 02 sensor go bad?)
thanks for all the replies and help