Idle Mixture Adjustment

[Jonty]

Ok thanks - the other issue is that I keep somehow damaging plugs. I seem to have to remove them and clean them up regularly otherwise I get a hard start and some misfiring in occassion

[Ron]

I would expect the plug issues with it set "very rich" or with the readings you now have...

Why raising the RMP with the CO screw doesn't change the dwell is what bugs me.

How did you make adjustments when using a gas analyzer?

[Jonty]

Thanks Ron - it’s all very weird I agree. No I’m not using a gar analyser. I did find this am that the three brass screws needed a half turn each to fully seat. Also it was quite hard to start this am. I plan to pull the plugs tonight and clean them up then there is a mech close by that knows Kjet and has a gas analyser so will get it check by that as that might then tell if my O2 sensor or the lambda system is playing up.

[Bertone80]

Thanks Ron - itÂ’s all very weird I agree. No IÂ’m not using a gar analyser. I did find this am that the three brass screws needed a half turn each to fully seat. Also it was quite hard to start this am. I plan to pull the plugs tonight and clean them up then there is a mech close by that knows Kjet and has a gas analyser so will get it check by that as that might then tell if my O2 sensor or the lambda system is playing up.

I don't know what a gas analyzer is but there is another way to adjust the mixture/idle for the PRV engine. I did it 16 years ago to my B28F before the engine developed a leak of coolant into oil. It not only worked I also took the car to a smog check (California) and passed the check with excellent measurements. The tool I used was an analog voltage meter with an input impedance of 10 meg ohms. I guess you could use a cheap digital meter but it will be harder because digital numbers tend to jump up and down too quickly.

What I did was to measure the voltage of the O2 sensor while I did the adjustment. The objective was to adjust the CO screw to obtain 0.45 volt from the O2 sensor. The voltage will drift between 0.1 and 0.9 volt if the mixture is too lean or too rich. It will be necessary to also adjust the brass idle screw because the idle RPM will fluctuate when the CO screw is tweaked. I was able to obtain a fairly stable 0.45 volt and an idle RPM around 950 for my B28F. I was able to rev up the engine and the voltage would stabilize to 0.45 volt as well. The smog check machine checks CO at idle and at 2500 RPM as well. As long as the voltage stabilizes to 0.45 volt at idle and 2500 RPM the mixture will be optimal and the Lambda system will yield a CO reading near zero.

It is easy to do but the trick is to use a voltage meter with 10 meg ohm impedance at the range of 1.0 volt. The one I used was a HP brand VTVM (solid state although called VTVM). I pick it up from a flea market long long time ago.

[Ron]

...No I’m not using a gar analyser..

I was asking about when you said...

I had the system tuned with an gas analyser so I was fairly sure the mixture was right.

...I was wondering about using the brass screws and/or CO screw, etc.

Anyway, I (we?) was assuming the smoke test eliminated vacuum leaks. But since the screws were open, you found one... So at this point, it looks like: leave them shut, clean the plugs, check the timing, and try setting it with a dwell meter. (You can disable the idle speed control system and set it up with a gas analyser like a Volvo (also in the manual) instead of a D, but why?)

[David T]

Large swings indicate 1 or more cylinders not firing properly. Could be spark plugs, bad injector, that kind of thing. To minimize dwell swings you must get every cylinder as balanced (matched) as closely as possible to each other. It starts with a compression test and often involves checking and adjusting the valves.

[Ron]

+1 ...I'm betting on plugs.

[Jonty]

Thanks all - it had a bit of a stutter this morning so you guys may be on to something with the plugs. This will be the third time I’ve had to clean them so something keep messing with them (and it’s not rich which is the usual culprit). Maybe they got damaged at some point

[Jonty]

Ron just picking up on a comment you made about setting up with a gas analyser. Would if not be better to disconnect the FV and set the mixture with a modern gas analyser then simply plug the FV back in again? Would that not set the perfect baseline for the 02 sensor and FV to do its thing? Would that not rule out a dodgy narrow band 02 sensor?

[Bertone80]

If the mixture is already too lean removing the CO adjustment plug can make it even leaner then the idle can no longer hold up but to die. If the mixture is too rich to begin with then removing the plug will cause the idle to surge up and down which can cause the idle to die too when the RPM swings to low and when that is too low. When the mixture is too lean or too rich the engine RPM will drop. This is why removing the plug will cause the engine RPM to fluctuate. usually you will see the idle revs up and down or simply dies.

The correct mixture needs to be set properly before setting the idle by turning the brass screw. You can sniff the exhaust to get an idea if the mixture is right. If the exhaust stinks the mixture is off. Measuring the O2 sensor voltage is the most accurate way to set the mixture properly. Sniffing the exhaust gas is not a good thing to do.

Make sure the vacuum lines are correctly connected and no leak before adjusting the mixture. The mixture has to be right for the idle to be meaningful. The adjustment should only be done when the engine has reached its operating temperature. You can set the idle a little higher then adjust the mixture. Once the mixture is right then turn down the idle to whatever it is supposed to be. I set my B28F with auto transmission to 950 RPM. I set the idle a little higher to prevent overheating when driving in congested city streets.