Sudden rough running afr jumps to very lean

[arhoch]

This is a ms2 V3 board running 2.905 fuel only with idle air control and a dual MAP sensors. In the last few days while tuning, the engine runs rough and misses at idle and under load. Attached a date log which indicates that the AFR goes very lean when the rough running occurs at idle. Yesterday, the engine quit running and would not restart. Fuel pump sounded odd. Fuel pressure showed very low. Began a long coat home with key on and transmission in 5th gear. Sometime during the decent, the engine began running and I made it home, but the rough running is almost constant now. I have checked the fuses and fuel pressure and all is normal. Do I have a fuel pump failing? Any input would be appreciated.

[Bluehaze]

Hi Arhoch
You running a swirl pot or header tank in your system? To take the simplest cause, I'd reckon you've sucked some air or maybe some contaminant into your system. This may also explain why your pressure recovered on the down-hill. Also the "odd" sound from your pump. (see my post-'Something new'') the present rough idle might be a clagged injector.

Richard.

[arhoch]

Richard,

Thanks for the reply. I'm not running a swirl pot or header tank. The fuel system other than the injectors, fuel rail, and pressure regulator is stock Porsche 911 SC with a large filter and accumulator in the system. The tank was flushed a couple of years back and the fuel filter is almost new. Doesn't mean I didn't pick up some trash from the gas station. I will check the filter. Thanks for the suggestions.

Randy

[arhoch]

The fuel filter is clean, and the problem is intermitent. It runs fine then doesn't while idling. It is as if only half the injectos are firing. I have checked the wiring to the injectors and all relevant connections I can think of. Thoughts?

[Bluehaze]

Perhaps you could test your suspicion of having half your injectors firing by pulling one of your injector fuses while it's running well at idle, see if this simulates the condition, or conversely pull a fuse while it's running bad, If it changes nothing, then you will be onto something, and can further investigate the non-firing injector bank.

[arhoch]

Ok. Disconnecting bank 1 makes it run slightly worse, and disconnecting bank 2 kills the engine. Would appear that the issue is with injector bank 1. If I disconnect that bank one at a time, 2 injectors have a neglible effect on running and one causes it to run worse. I either have 2 failing injectors or two clogged injectors. I'm leaning toward failing since there are two injectors operating intermitently.The injectors I'm using only come in 6 packs. A little disapointing if they have failed since they are only 2 years old and have maybe 3000 miles on them. Thanks for the trouble shooting advice.

[Bluehaze]

To be sure it's the injectors themselves rather than the wiring, connectors, etc is you could switch the injectors around to see if the problem follows the injectors or stays located with the associated cylinders. If its the latter, then it's more likely to be wiring and not the injectors themselves. As for buying new injectors, maybe you are able to get them reconditioned and tested? Must be cheaper than buying new ones.

[arhoch]

That is a thought. I need to pull them to check to see if one or two are clogged once that is done assuming I don't find any thing, I will swap the non performing one with one that appears to be performing. I checked the impediance and they are all within spec and very close to each other. The connector to each one check out with a multimeter as well.

[arhoch]

The injectors are clean an appear almost new. I swapped 2 injectors one working one not both on the same bank. There was no change. It would appear to be at the connector or where the trigger wire for that cylinder's injector is spliced in. Otherwise, all the injectors on that bank would have the problem. I have power to the injector in question, and engine off, there is a small amount of resistence on the ground side. Engine running, it shows an open line and it may be that the ground connection is made and unmade faster than the multimeter can respond. Thoughts on further tests anyone? I could simply replace the connector to that injector, but I would like to have a positive test for a failure before replacing parts.

[trakkies]

An LED with series resistor wired across the injector will show the injector signal switching at low revs. Or a noid light.

[arhoch]

Thanks for the tip. I have a 12v led so shouldn't need a resistor. I'll try it and report back.

[arhoch]

The little LED blinks like champ when its running. I'm going to look elsewhere for issue. Will report back as it is resolved.

[trakkies]

The little LED blinks like champ when its running. I'm going to look elsewhere for issue. Will report back as it is resolved.

Obviously that only shows they are being fed correctly. It wouldn't show up a faulty injector. But my experience says this is pretty rare, and more than one even rarer.

[arhoch]

Agreed. The injectors and injector drivers seem to be OK. I tried richening the mixture and get a much improved idle. Vigorous acceleration but not WOT is quite brisk, but when backing out of hte throttle and gearing up to cruise speed and throttle position then appling light throttle yeilds a stumble. It is as if the richer mixture is masking the true problem.

[arhoch]

All,

I believe Ihave found at least part of the problem. I had the AFR fuel clac setting under other fuel st to use a separate AFR/VE tables. I changed the setting to use a combined VE/AFR table, and it imediately settled down. I have since leaned out the idle and will work on the reso of it tomorrow on the road. Thanks for all the suggestion and ways to look at the problem. It was invaluable in finally tracking down the soulution.