2.1L Digifant - Diagnosing a Rough Cold Idle

[john_lessard]

Hello Samba!



It has been a long Winter filled Van maintenance and fixes. We pulled the engine from my friend's 1991 Tin Top (manual) to do some various work to it. Engine has an unknown amount of miles (we are guessing under 100k) Nothing too invasive was done. Upon reinstalling it, it seemed to run well, or so we thought! We are stuck troubleshooting a strange issue with the engine. The problem is as described:


When the engine starts (either it is cold or warm) it idles right around 950 RPMs as you'd expect. After a couple seconds, the engine begins to stumble. It will struggle to stay running around 400-500 RPMs. Opening the throttle would cause a backfire or stall. If we allow the van to stay running like this for a couple minutes, right around when the thermostat housing sees 180F, the engine smooths out to 950 RPMs. At this point, the van runs and drives fantastically! Power seems great and there is virtually no hesitation when accelerating. I will note it seems a tad bit sluggish when accelerating off idle.


Here is a list of parts we replaced / added while the engine was out:

- Clutch Kit
- Rear Main Seal
- GoWesty Stainless Exhaust
- Hollow Catalytic Converter
- New Bosch O2 Sensor
- New Bosch Fuel Pump
- New Bosch Fuel Filter
- CB Performance Oil Pump (Maxi Pump 1)
- GoWesty Billet Aluminum Throttle Body
- New Temp II Sensor
- New Thermostat and Thermostat Housing
- GoWesty rebuilt Idle Stability Control Unit
- Rebuilt Distributor with Bosch Hall Sensor
- Marco Mansi Fuel Injectors
- Van Cafe Aluminum Fuel Rails
- K&N Air Filter
- GoWesty Alternator Bracket supports
- GoWesty HD Alternator Harness
- Rebuilt Alternator


Parameters:

- O2 Sensor Plugged in
- Timing set to 37 Degrees at 3000 RPMs
- AFM signal conditioner harness being used
- Amsoil European Car Formula 5W-40, Mann filter


Diagnostics I've performed:

- Tried other known working ECU - no change
- Tried other known working AFM - seems to change the characteristics of the problem, but no significant improvement
- Cleaned and checked continuity of carbon track on current AFM
- Tried other ISCU - no noticeable change
- Tried other ICV and cleaned current ICV - no noticeable change
- Used a homemade smoke machine to check for vacuum leaks, nothing found
- Tried plugging Brake Booster vacuum line, no change


When I first started attempting to tune this engine, I had an issue where the van seemed to run well cold, and stumbled and bucked (while accelerating) when hot. What seemed to get me where I am now was by tuning the AFM when it was warm. I would primarily adjust the spring tension of the wiper and wiper arm position to hone in on the sweet spots of the carbon track that minimized the bucking. I would drive down the road, row through the gears, and adjust the spring tension (moving the black plastic gear a few "clicks") to see if the bucking would subside. I got to the point where the van ran fantastically, hot idle was smooth, and power was good. But, once I got here, the issue of the rough cold idle arose.


The big red flag to me is that the van will not idle and hardly run at any temperature with the O2 Sensor disconnected! I believe the van should run fine with the O2 sensor disconnected, so I am suspecting when it is hooked up, it is severely compensating for something.


Another interesting point is that during the stumbling cold idle, if I physically move the wiper arm counterclockwise, the engine begins to smooth out (still rough, not a perfect idle). But, once the engine is at temperature is idling well, moving the wiper to that same position causes the motor to nearly die (I believe it causes it to run way too rich).


I apologize for the overload of information. I am really hoping a Digifant Wizard could chime in here. I feel like I have tried so many things and just end up moving the problem around.

_________________
1986 Tin-Top Vanagon, 2.0L Zetec, Peloquins TBD 2WD

[dabaron]

have you confirmed the correct temp sensor values, the TPS, and the O2 coax and grounds are good?

sit down with this lovely gem and you'll find the problem:

http://oldbluesblog.com/files/DigifantProTrainingManual_SingleSided.pdf

pages 17 and 30 are good places to start
_________________
1991 Vanagon GL Camper
i had no idea i wanted to be a mechanic
"burnin oil and cookin coils" -- Destructo

BiWerks Design, LLC

[DigiMatrix]

Sounds to me like a fueling issue. Have you happened to check your fuel pressure regulator pressures? The engine should run just fine with the O2 sensor disconnected (open loop). If your pressure regulator is messed up, maybe in closed loop the ECU is able to compensate.

Also, the O2 sensor signal is very sensitive, so you should measure the signal wire resistance between the O2 sensor and the ECU to ensure there is very little resistance (under 2 ohms) and no shorts to ground or the O2 heater circuit (the other 2 wires in the O2 harness)

[john_lessard]

I did confirm the O2 Sensor is not grounding out to the shielding wire. I will certainly test the resistance of the signal wire to the ECU.

I confirmed the TPS is working correctly and adjusted it according to the Bently.

Testing the fuel pressure is next on my list. That's a good point the ECU may be compensating for bad fuel pressure. The Fuel Pressure Regulator is about 1 year old, and the Fuel Pump and filter were just replaced.

I will also test the resistance values of the new Temp II sensor at different temperatures. Perhaps the new sensor is a dud.
_________________
1986 Tin-Top Vanagon, 2.0L Zetec, Peloquins TBD 2WD

[dabaron]

[zoti]

It sounds like it's lean. It's exactly the same issue I had.

Have you checked idle mixture to confirm it is correct?

Idle mixture adjustment:

Bring engine to operating temp. (Fan should turn on once).

Disconnect O2 sensor
Disconnect ICV
Disconnect breather hose AT breather tower and plug (tower is unplugged, hose is plugged).

Turn engine on. Let idle and check voltage on O2 sensor (green wire and ground). Should be about 0.7V.

 Coarse adjustment is by moving wiper arm (loosen Phillips screw on wiper arm and move arm left). Fine adjustment is with 3mm Hex air screw on AFM.

[john_lessard]

zoti,


My only problem at the moment is engine hardly runs with the O2 sensor disconnected (only if I start the engine with it unplugged, so the ECU goes into open loop mode). I did attach my volt meter to the O2 sensor during the stuttering cold start and noticed <0.1V values. You may be onto something.

I could probably dramatically adjust the wiper arm position so it will run when the O2 sensor is disconnected. If I physically move the wiper arm counterclockwise, the engine seems to smooth out.
_________________
1986 Tin-Top Vanagon, 2.0L Zetec, Peloquins TBD 2WD

[zoti]

[john_lessard]

[zoti]

[john_lessard]

I think I'm making some progress here. I need to stop assuming new-ish parts are working as they should and they couldn't possibly be contributing to a problem.

Following the Pro-Training Manual, I performed a fuel pressure test:

First off, at idle, with everything still plugged in, I was seeing a fluctuating needle from 25-30 psi.

I then disconnected the Fuel Pressure Regulator, and watched the needle ride and bounce between 30 - 35 psi.

I tested just the fuel pump by jumping the pump relay. I saw a steady needle at 33 psi.

I then noticed as soon as the pump was off, the pressure would fall quickly (gone in less than 5 seconds). I clamped off the return line and saw the pressure hold continuously.


I have determined my <1 year old Bosch Fuel Pump is not putting out the correct minimum pressure of 36 psi. I will try an old spare pump to see if the pressure returns to where it should be. I will also shop for a good replacement.

I have also determined my <2 year old Fuel Pressure Regulator is no holding pressure. I think that stupid thing cost me like $90 from a major Vanagon supplier! I have a new Beck/Arnley on order now.


Needle 25 - 30 psi at (roughly) 950 rpms idle. FPR connected.

Needle 30 - 35 psi at (roughly) 950 rpms idle. FPR disconnected.

Any ideas what the fluctuation of +/- 5 psi fuel pressure is all about?
_________________
1986 Tin-Top Vanagon, 2.0L Zetec, Peloquins TBD 2WD

[DanHoug]

couple things.... you said it runs better if you push the AFM wiper around. we're it mine, i'd adjust spring tension until it runs nicely, shooting for disconnected O2 sensor values as close to 0.5v as you can get. the sweet spot on my '87 is 90-110 grams spring tension to where it just lifts off the rest stop.

then, how is the fuel pressure gauge needle acting? twitchy fast varying or slow pulsing. any regularity/periodicity to it? the four things affecting pressure are the pump condition, the intake screen in the tank, the fuel filter, and the regulator. do the cheap things first like another filter. then blow air back into the tank. pinch off the return hose and see if the pressure shoots to close to 100psi.
_________________
-dan
60% of what you find on the internet is wrong, including this post.

'87 Westy & '89 Westy both 2.1 4spd

Past projects can be found at--
www.thefixitworkshop.com

[zoti]

[john_lessard]

[MarkWard]

I'll bet if you put a vacuum gauge on the vacuum hose between the intake and the fuel pressure regulator, that needle would also bounce. Sounds like you have a manifold vacuum problem. The changing fuel pressure is not helping matters.

Here is the deal with adjusting on the AFM. It was never intended to be adjusted on. You'll notice there is zero bosch information about adjusting on it.

When you adjust on the AFM, your change is across the entire sweep. So, when you adjust it for hot idle, you are likely ignoring the real problem.

So, I'd concentrate on getting a good smooth idle during start and warm up, then move on to the warm running performance.

You are exactly correct, the O2 sensor is likely working it's butt off to compensate and hide the actual problem. A proper running engine should not even need the OXS sensor. Good luck. You are learning a lot.
_________________
☮️

[john_lessard]

[MarkWard]

[john_lessard]

Happy to report I have made significant progress with this issue. Following the Digifant Pro Training Manual, I was testing the harness from the ECU connector confirming everything was okay. Sure enough, I found a bad reading on pin 19 (ground wire that bolts to the driver's side cylinder head). The wire was partially broken. After repairing this wire the engine ran 1000 time better and smoother! This broken wire must have thrown the ECU through a loop causing strange issues.

I had a successful test drive after minimal changes. There was still a tiny bit of hiccupping when accelerating, but I am hoping to adjusting the AFM or timing slightly will resolve that. I will spend some time dialing in the engine following the fuel/air mixture procedure.

As always, when you fix one problem a new one must pop up! After this test drive, I noticed the van was running a bit hot. I went to measure the temperature at the thermostat housing and after reading 230F, I herd a "pop." The rear heater core sprung a leak and flooded the inside of the van! A New heater core, radiator, and thermostat should solve the hot issue. Should be back on the road soon!!
_________________
1986 Tin-Top Vanagon, 2.0L Zetec, Peloquins TBD 2WD

[MarkWard]

[john_lessard]