[FIXED] Running rich; leaky injector?

Hello everyone!

So, I have a running rich code on 1.4, 4cyl VW gasoline engine which has no MAF. Checked the o2 (narrowband) response be simulating rich/lean condition, looks good. Fuel trims are not consistent but are mostly indicating rich (25-40%). My suspicion falls on leaking fuel injector. I have checked the spark plugs and the one on the 2nd cylinder is visually quite different (dirty and black). I made a test by unplugging each of fuel injectors and monitoring STFT. Trim numbers were still bad except when unplugging 2nd injector when the STFT drops around +-10% occasionally outside. I would let the car stabilize for few min before each injector testing, and also I switched the "bad" spark plug. Each spark has its own ign. coil. Weak ign. coil would cause some misfire right? This engine runs smooth at all revs.

I would appreciate any opinions on this problem.

Thank you,
and It's good to be here, my first posting!

In these times of accurately controlled fueling and difficult to access plugs reading plug colours is a bit of a lost art. My money would be on a leaky injector but I would investigate further, although I don't think it will make any difference but swap the coils over and check the injectors for balance and leakdown or you could also try pulling the manifold off the head and looking for obvious fuel traces. However before doing anything major it might be worth trying some injector cleaner in the fuel.

Over fueling tends to put black fluffy soot deposits on the plugs and pistons, if you see glossy deposits it is oil fouling.
Other problems such as lean air-fuel ratio, or leaking valves tend to give the plugs a bleached white appearance, head gasket failure tends to make the plug and combustion chamber either look very clean or white powdery or with modern OAT coolants a distinctive red/orange stain.

Hi, thank you for your replay!

I have tried to swap the ign. coils with no change in behavior. Removing the intake would be to much work to visually inspect and unfortunately I don't have the tools to do the balance check (fuel pressure gauge and a injector tester), though I have tried to activate injectors through diagnostics(VCDS) via output activation with no success in revealing anything. I forgot to mention that car runs on LPG as well, but during testing it was off. i.e. running only on gas. This should explain nice white color (except the 2nd) of spark plugs, same as when running lean as you have mentioned.

So, in the end I have put my money on injector, changed it and the car is fixed! Went for a nice test drive and trims are back within limits.

Do you thing that this "method" I used (unplugging injectors separately and observing fuel trims) would work in other cases where a suspition is on injector(s)?
I think that by disconnecting the bad injector ECM gained back the control of fueling and compensated with the remaining 3 to keep the exhaust where it needs to be, though engine was running on 3cyl.

When I was running LPG I always added a little injector cleaner to the petrol about 50ml per tank full. I don't if it did the injectors and valves any good but I never had any issues.

In my experience isolating the fuel rail and monitoring pressure drop is the surefire way of seeing a leaking injector. One would think that a leaking injector would also have different flow rates then a non leaking one, so using a tool such as this may also be a viable option in diagnosing it, but I haven't tried it on a leaking injector situation yet.





As for the method you used, it could be worth experimenting with further. One could speculate that disabling an injector has a unique effect on short term fuel trim immediately, and by comparing the fuel trim reaction of each injector it could highlight one that reacts dramatically differently. I like it! I'll have to give it a try on the next leaking injector problem I see, and I'll try and report my findings.

Thans for your replay SniperDan!

The tool on the photo is the right tool for the job.

As you say, i was comparing FT reaction for individual injectors. There was a clear difference and engine was running much nicer in terms of trim numbers and 02, with 2nd injector disconnected. Computer could not control the mixture. Without it it has adopted as much as it could to keep the lambda at 1.
I must mention that i had slight problems getting good readings. By disconnecting inj. you get a fault code and readings stop. The way i got around it; cleared the code, start the engine. It worked for me.
More on this: rewind at 9:45 --

I would like to know what Mr. Danner would comment if the car didnt go into opened loop.

I'll try this on some other cars an see if it can be used, hoping for no data reading problems.

Cheers

I'm sure to combat the problem of the engine entering a default loop status, etc. you could just rig a test light in series to substitute the load and avoid fault codes. The only thing I have used my injector timer on thus far is checking for restricted injectors, but I am keen to give it a whirl on a known leaking injector to see what results I collect.

That's nice trick to fool the computer. I'll keep that in mind, though it was not necessary on this car.
Good advice, thanks!

This is the one I use. It was a cheap-o ($25 I think on Amazon).
The leads are short, and it didn't come with any adapters, but I clip injector connectors off junk engine harness and keep them in the case for when I need them.

I'm kind of curious to see if it would help ID a leaky injector or not. The problem injector would always be leaking, so whatever pressure loss is coming from that injector would be observed no matter which injector you were testing. Also, I think the results would be hard to interpret with the gauge needle constantly falling.

Or maybe I'm over thinking it