Need some input -> Problem with 3.2 transplant going lean and misfiring after two min

[Wayne 962]

Okay, today was a complete failure to find the problem. Steps taken today:

- Filled the car up with oil

- Replaced the ignition wires (no change)

- Replaced the coil (no change)

- Replaced the DME relay

- Emptied all the gas out and refilled it with fresh gas (no change)

- Checked the fuel injection wire harness *completely*:
- Resistance of each fuel injector coil, measured at the injector plug
1- 2.5 ohm
2- 2.5 ohm
3- 2.5 ohm
4- 2.5 ohm
5- 2.5 ohm
6- 2.5 ohm

- Resistance of each injector coil as measured through the harness to pin 14, 15 and 87 on the DME relay
1- 3.2 ohm
2- 3.0 ohm
3- 3.0 ohm
4- 3.8 ohm
5- 9.1 ohm
6- 11.6 ohm

- Harness resistance to pin 87 from injector plug
1- 0.4 ohm
2- 0.4 ohm
3- 0.3 ohm
4- 0.6 ohm
5- 1.0 ohm
6- 2.0 ohm

- Harness resistance to pin 14/15 from injector plug
1- 0.5 ohm
2- 0.5 ohm
3- 0.5 ohm
4- 1.0 ohm
5- 2.0 ohm
6- 3.8 ohm

- Injector plug harness resistance to 4-pronged plug (left / right pins in connector)
1- 0.5 ohm / 0.5 ohm
2- 0.5 ohm / 0.3 ohm
3- 0.5 ohm / 0.3 ohm
4- 0.1 ohm / 0.8 ohm
5- 1.8 ohm / 1.0 ohm
6- 1.3 ohm / 1.3 ohm

- Four-pronged plug to ECU & DME relay
pin 1 -> 87 DME Relay - 1.9 ohm
pin 2 -> 87 DME Relay - 1.4 ohm
pin 5 -> 14 DME Connector - 0.5 ohm
pin 6 -> 87 DME Connector - 0.6 ohm

- After cleaning with isopropyl alcohol, placed a 9V batter on pins 87 of DME relay and DME connector pins 14 and 15. Measured the voltage (reference voltage at battery of 9.43 volts)
1- 9.43 volts
2- 9.43 volts
3- 9.43 volts
4- 9.43 volts
5- 9.43 volts
6- 9.43 volts

Conclusion? There was a small increase in resistance in the harness on the side of the motor containing cylinders 4-6, but the voltage drop test didn't seem to indicate that it had any effect. This was with very little current, of course, but my gut is that that these values are not enough to cause issues. Still, some clues seem to point to there being an issue on this side of the engine, and perhaps also with respect to cylinder number 6 (where the resistance in the harness seems to be highest). I still don't think that it's high enough to be causing problems. The connectors are available, I can create a new harness, if even to just test to see if it's an issue.

So, I'm beginning to think that I my initial diagnosis of the issue was incorrect. I think the car was missing all along, and that because the idle was initially set to be around 1,100, I didn't recognize that the car was missing when running and/or starting up. Now, it starts missing at startup, but you have to listen very carefully for it. I think lowering the idle down a bit made it much more obvious? Maybe. I can't recall now exactly how it was running when we first put it back on the road in February. I honestly don't remember it running / idling really terribly - it does feel like it's gotten worse.

When I did the compression check on the engine, I pulled all of the plugs that had about an hour's worth of "idling time" on them (installed new ones today to see what they look like next time I pull them). For the most part, they did not look very good. They actually looked like they had bad fuel gummed up on them (hence why I replaced the fuel). There's no smoke coming out of the tailpipe (neither white or black), although today, there was a small, slight wisp of white smoke for one second on start up - nothing remotely significant.

I'll post the pics of the plugs - the worst one is #6, but they all have brown discoloration on them. Again, these have about an hour on them total, in this poor running engine.


Next possible steps:

- Replace cylinder head temp sensor and/or use paperclip to short it out to test
- Perform leak down test (warm car, if possible)
- Pull injectors one-by-one very carefully and monitor to see if the (seemingly random) misfire is associated with one of the cylinders in particular (already did this, but will do it again)
- Start the car from complete cold and check with the thermal camera / thermometer 30 seconds after starting for header temps. If there is a subtle misfire, then one of the cylinder temps will be slightly less - letting the car run for a length of time seems to merge the temps together.


Any suggestions would be helpful, I'm really scrapping the bottom of the barrel for ideas.

-Wayne

[Wayne 962]

More photos. #6 is the first one, then 1,2,3,4,5

[Wayne 962]

Some better photos. #6 is the 2nd one below. I do not think that the base ring should look like it's covered with bad, gummy fuel. If you touch the base ring, it's sticky / wet.

[Wayne 962]

I thought that perhaps the fuel injection cleaning shop made a mistake on the injectors, but I checked the flow rates and they seem okay. Here's a neat site that contains all of the resistance values and flow rates for the various injectors:

https://www.injectorrx.com/fuel-injector-cleaning-and-flow-testing-service/fuel-injectors/fuel-injector-data/bosch-fuel-injectors/

Another one: https://www.robietherobot.com/storm/fuelinjectorguide2.htm



Here's the output from the shop again:



Then again, I swapped around injectors and it didn't seem to make any difference whatsoever...

-Wayne

[Wayne 962]

I also took several photos with the FLIR infrared camera. Hmm, I didn't see these numbers until I just uploaded them here. The temps on #6 appear to be a little bit lower, but that *could* be caused by header design. I have no "pre-problem" data on what they should be under normal operation.

When the car misses, I do hear a very light / faint "slapping noise" on the left side of the engine from within the engine bay, and today, I also heard it quite a bit when I had my ear near the left rear wheel. It almost sounded like it was coming from inside the oil tank.

Hmm. So, we seem to have the following - higher resistance values in the harness for injectors 5 and 6, and lower exhaust temps on #6. That is just a bit way too coincidental for me.

Still, the compression check on #6 came back very strong:



I wonder what the leakdown would say...

Tomorrow I'll be getting the replacement / test connectors for the injector harness, and I can quickly build a test harness to see if this problem goes away.



-Wayne

Cylinder #1:



Cylinder #2:



Cylinder #3:



Cylinder #4:



Cylinder #5:



Cylinder #6:

[creaturecat]

A+ for perseverance. i gotta say.

[dannobee]

If you have an automotive stethoscope, you can hear each injector clicking when it's running. If it's not firing, no clicking. If it's even partially plugged, it'll go "thud thud thud" instead of "tink tink tink." If it's intermittently missing and you think it's an injector harness problem, try the stethoscope trick. Play with it until you get a sense of what you're hearing.

Those old cars were sometimes tough to diagnose. The newer cars point you in the right direction with the trouble codes they give.

[mysocal911]

Originally Posted by dannobee

The newer cars point you in the right direction with the trouble codes they give.

Not always!!! A good example is a mis-fire code.

[mysocal911]

Originally Posted by Wayne 962

- Resistance of each injector coil as measured through the harness to pin 14, 15 and 87 on the DME relay
1- 3.2 ohm
2- 3.0 ohm
3- 3.0 ohm
4- 3.8 ohm
5- 9.1 ohm
6- 11.6 ohm

Injectors 5/6 aren't being driven properly thru the harness!

[Wayne 962]

Holy cow! Would you believe that a replacement harness (that I made this afternoon) did *not* work!?! After all that, I was quite sure that would work. To recap, we had the following:

- Lower temps on #6
- Higher resistance on the #6 injector
- Spark plug on #6 was looking poor

I mean, what more could you ask for. Once again (it's now going on 25+ year), this car is kicking my ass. Just to recap, this is the car that I bought several years prior to me starting Pelican (bought the car in 1995, started Pelican in 1997). I've probably spent 1,500 hours working on this thing over the year (including tracking it down when it was stolen). NOT WORKING YET!!!

Okay, so the replacement harness had no effect. It does not appear to be the harness. However, I did unplug each injector and slowly / carefully listened for the miss. It's definitely coming from cylinder #6. I managed to do a leakdown test on #6 tonight (right before my leakdown tester hose sprung a leak), and #6 is showing about 3% leak down - no problems there.

I also spent a chunk of time looking at the engine with the timing light today. I did not see a single variation or blip in the timing lamp frequency / strobe. Nothing, nada - looks perfect. This does not appear to be an ignition problem.

I took a look at the exhaust temps again. Lower temps on #6 and lower temps on #3.

So, I did an interesting experiment today. I unplugged all three injectors from 4-6 and ran the car on only 3 cylinders (1-3). The car ran and didn't skip, miss or stall (didn't run well, but ran consistently). Then I plugged the other three back in and the car started to randomly miss. Then I unplugged 1-3 and ran the car on only 4-6. It too ran fine with no missing. Plugging them all back together results in a miss (looks like the miss is coming from #6). I have a video of this very odd behavior.

Cylinder number six was recording lower temps today, but then when I started unplugging injectors, it started to not miss and record normal exhaust temps. Odd. I also saw something similar with cylinder #3, although less pronounced.

So, we have the motor running consistently on 4-6 only, and also on 1-3 only, but 1-6 make it miss. How confusing is that? The curious thing is that the fuel dampener is located *right* next to cylinder number six. The pressure regulator is located next to #3 (which also seems to be recording lower temps).

Tomorrow I will swap out the injector on #6 with the injector on #1 and see if a problem follows the injector or stays with #6. It very well could be a bad injector, *or* it could be something with the fuel pressure regulator and/or the fuel dampener. Either way, this problem has been *very* difficult to track down. I thought for sure it was the injector harness, but that doesn't appear to be the case. I made an alternative harness for both #5 and #6 and wired them directly into the breaker box ===> NO CHANGE.

So, I have a video, where I unplug #1-3 and then plug them back in and then unplug 4-6 and then plug them back in. The engine runs consistently on 3 cylinders, but when plugging in all six, it misses and runs poorly. What does this mean? Who the heck knows, but now my theory is the pressure regulator and/or the pressure dampener. Either that, or a bad injector (we'll figure that out when we swap out injector #6 with #1 tomorrow). My gut is that the problem will stay with cylinder #6, as I just had all of the injectors cleaned and tested okay.

Video (poorly shot) of the car with me unplugging a whole side of injectors and then plugging them in on the other side:



Timeline:

- Start -> Injectors 1-3 are unplugged, car runs on 3 cylinders, but evenly.
- Then I go an plug them in, car runs poorly on 6 cylinders (with missing) - Video at 0:25 - 0:46
- Then I go and unplug 4-6 injectors, car runs on 3 cylinders, but evenly. - Video at 1 min - 1:25
- Then I go and plug back in 4-6 injectors -> car runs poorly on 6 cylinders - Video at 1:50 - 2:10
- Then I go and unplug injectors 1-3 again -> car runs on 3 cylinders, but evenly. - Video at 2:40
- Then I go and plug everything back in again, car runs poorly - Video at 3:00

Bottomline, the car doesn't miss on 1-3 only, and doesn't miss on 4-6 only. But when all six are plugged in, the car runs and misses. What could be causing this? At this point, there's only one thing I can think of - the pressure regulator and the fuel dampener. These would appear to be the only things that could cause this behavior?

Summary:
- I believe that the problem with this missing is focused around cylinder #6
- I do not believe it's caused by an ignition issue
- It does not appear to be a wiring harness issue
- It's probably not a mechanical issue (compression test and leakdown on this cylinder looked good)

Could be a fuel injector issue or a fuel supply issue. We'll swap injectors tomorrow, and I'll see if I can borrow a fuel pressure regulator and a dampener from someone. Or perhaps there is some off-the-shelf BMW or VW part that may be relatively cheap that I can swap in and use.

Thoughts?

-Wayne

-Wayne

[Solamar]

I may have missed it, but have you checked fuel volume delivery? Should be something like 850 ml in thirty seconds.

[mysocal911]

Originally Posted by Wayne 962

I made an alternative harness for both #5 and #6 and wired them directly into the breaker box ===> NO CHANGE.
Summary:
- I believe that the problem with this missing is focused around cylinder #6
- I do not believe it's caused by an ignition issue
- It does not appear to be a wiring harness issue
- It's probably not a mechanical issue (compression test and leakdown on this cylinder looked good)

Could be a fuel injector issue or a fuel supply issue.

Again:

1- 3.2 ohm
2- 3.0 ohm
3- 3.0 ohm
4- 3.8 ohm
5- 9.1 ohm
6- 11.6 ohm

Injectors 5/6 now indicate the same resistance as 1-4, right? If not, why not?
If OK, switch injectors 3 & 6.

[creaturecat]

"baby bottle" test. to determine flow/pattern out of injectors? Here I Go: Testing my refreshed 1983 Lambda CIS

[jpnovak]

Have you verified fuel flow volume and/or cleaned the supply line screen in the bottom of the tank?

Have you checked to make sure the supply and return lines are not kinked? Is this using the stock 914 fuel pump mounting location? Does yours have the "window" on the DS of the bulkhead to mount the fuel pump? Is this the stock fuel tank?

I used to have issues with the return loop of the pressure side of the pump getting kinked under the fuel tank when the pump was removed/replaced. careful hose routing is required.

It is possible for the engine to run normally for a short period of time until the fuel pump can not deliver enough fuel to support the overall requirements. Restricted fuel screen would also cause this issue.

[stlrj]

Originally Posted by Wayne 962

Any suggestions would be helpful, I'm really scrapping the bottom of the barrel for ideas.

-Wayne

You might continue scrapping the bottom until you realize the problem is not fuel related but an ignition problem. Plugs do not look that way unless they are missfiring and that points to an ignition problem. Modern cars eliminated that problem when they eliminated the distributor/ignition cables for that very reason.


Cheers,
Joe
87 Carrera that had lots of problems with plugs looking like that until I concentrated my efforts on ignition!

[gtc]

Did the 5/6 injector harness resistances come down to match the rest when you tried your test harness?
Have you had the injector harness off to inspect? I found mine to have worn through the vinyl sleeving and some of the injector wires' insulation when i had my engine apart for rebuild a few years ago. I suspect it could have been shorting to the fuel rail.

[Wayne 962]

Originally Posted by Solamar

I may have missed it, but have you checked fuel volume delivery? Should be something like 850 ml in thirty seconds.

We emptied the tank and refreshed the fuel on Sunday. While I did not exactly measure the fuel flow, we were able to fill a 5-gallon gas can in about 3-4 minutes. This was out of the "fuel pressure test port", so there doesn't appear to be a clog or problem with volumetric flow?

-Wayne

[Wayne 962]

Originally Posted by jpnovak

Have you verified fuel flow volume and/or cleaned the supply line screen in the bottom of the tank?

Have you checked to make sure the supply and return lines are not kinked? Is this using the stock 914 fuel pump mounting location? Does yours have the "window" on the DS of the bulkhead to mount the fuel pump? Is this the stock fuel tank?

I used to have issues with the return loop of the pressure side of the pump getting kinked under the fuel tank when the pump was removed/replaced. careful hose routing is required.

It is possible for the engine to run normally for a short period of time until the fuel pump can not deliver enough fuel to support the overall requirements. Restricted fuel screen would also cause this issue.

Hmm, so emptying the tank (which is what we did) would indeed presumably bypass the return line. But that might show problems across the board - not just on cylinder number six?

-Wayne

[Wayne 962]

Originally Posted by stlrj

You might continue scrapping the bottom until you realize the problem is not fuel related but an ignition problem. Plugs do not look that way unless they are missfiring and that points to an ignition problem. Modern cars eliminated that problem when they eliminated the distributor/ignition cables for that very reason.


Cheers,
Joe
87 Carrera that had lots of problems with plugs looking like that until I concentrated my efforts on ignition!

Thanks Joe. At this time we've replaced 100% of the ignition system:

- Plugs
- Wires
- Cap
- Rotor
- Coil
- ECU (test unit loaned to us by Steve W.)

I guess I haven't replaced the distributor, *but* I took the timing light yesterday and checked (for about 2 min) cylinder number one, and cylinder number two. They were spot-on. If there were an ignition problem, I'm not seeing any clues to that anywhere, and I wouldn't know where to look next either?

-Wayne

[Wayne 962]

So, I missed uploading a video that gives another data point. I noticed that the fuel pressure gauge yesterday started vibrating. I hadn't seen that before, and this seems to hint at the fuel pressure damper and/or pressure regulator not performing their respective tasks?

Some people have said that he damper is used to insulate the injectors from pulses contained in the pump, and some have claimed it is used to even out the pulses that would emanate from the fuel injectors. I have not seen a BOSCH document that definitively indicates the purpose of the fuel damper - I'll have to keep hunting for it...



-Wayne