Engine misses, surges or bucking (FI related only)


vw-traveller

New member
Okay. Thank you for the info. I will once again go through the FI system tests, although I just did that shortly ago, and it all checked out. Btw. The new emmisions tests we have in California are done under load, on a dyno at idle, 15mph and 25mph. Sure there's no wind resistance but i feel that passing that test warrents the engine is running at it's most effient level, esspecially since they now test HC's, CO, and NOX's.
 

Capt. Mike

Moderator
But that's still relatively steady-state and AFTER the cat and O² sensor adjustments so is artifical. It's a real picture, but not the picture in transisition where you seem to be having most of your problems. One area to look over is your throttle switch -- a throttle switch failure can give those symptoms because it's trying to adjust the FI for transition from idle to run to full throttle, and since it is moved directly by the throttle, it's always changing. There's a similar device on the Digifant FI and I lost mine. It would start & idle OK and run at full throttle but couldn't get past the transistion at part load. Bad AFM can do the same.
 

Trout_Teaser

New member
Hi All, I hope this is the best area for this post. I have an '84.5 watercooled Westy automatic. A short while back the Westy started to experience cold running problems and was a bit down on power when warmed up and run for a little while. The extra load of selecting drive when cold would usually cause a stall unless my foot was working the throttle (air pedal) at the same time. Initial start-up seemed okay for the first 30 seconds then it would start to run very rich and miss a bunch with a rough idle. There was also a nice brown cloud coming out of the exhaust when cold and the engine was reved. When warm there was still a slight cloud that could be seen out the rear when driving. These extra emissions also caused a good headache while trouble shooting.
It had been a while so I did a tune-up. New cap, rotor, plugs, air filter and fuel filters. I was hoping for a clue on the distributor side of things or maybe a huge gap on the plugs. Everything looked okay just signs of rich running on the plugs. Timing was also spot on.

No luck so I checking this site for some clues and found a couple of leads.
First I replaced the 02 sensor. It had been a while for that part and on some cars when that goes the ecu default is to rich and saving the engine. As I kinda guessed, before hand, there was no change. Next to the Bentley Manual.
I checked the Temp. Sensors 1 and 2 at coolant manifold and AFM both read correctly.
I checked the AFM. It worked correctly. When moving the flapper door I did expect to see the resistance on the tester change in a more linear fashion as the door opened. I confirmed resistance to a similar flapper door afm and it read the same way.
I checked the injector spray pattern and it was fine. I procedded with all the other tests and checked all the values at the ecu harness and they all read correctly.
Then I read about the powersteering pressure sensor and sending a load signal to the computer. I looked around for anything that looked electrical on that system and I guess my van does not care about that.
Next I checked the Fuel Pressure at the "T" fitting. I did not have a pressure gauge but I did have an air pressure gauge off of a floor pump for a bicycle. A little fuel hose and I had something that would work. I started the van and the pressure measured just as Bentley had suggested. I removed the vac. hose from the manifold and the pressure changed correctly. Then after about 30 seconds of running the gauge jumped to 90+ psi and the van started missing and putting out the vapor cloud. I shut it off and duplicated the test noting that the fuel lines would get noticeably hard when the pressure went up. I checked the fuel return line to the tank and it is moving plenty of fuel restriction free when the ignition key is cycled.
So it looks like what started out as a temperature related problem is most likely a fuel pressure regulator problem. Unless I am missing something
icon_frown.gif
I am still not sure why there is a delay before the sharp rise in fuel pressure.
 

Capt. Mike

Moderator
Ah, someone who has actually read Guideline #3. Excellent job of preliminary diagnosis.

First, I'm going to suggest you determine if you are a "early version" Digijet per Bentley 24.24 or "late version per 24.241-24.24b. You'll see the late version has the idle speed boost valves for A/C & P/S. Is best I can figure out, these are NOT electric, but change airflow/vacuum that affects the aux. air regulator.

Another area you might check is the vcuum valve for the charcoal cannister system and that system itself. It could be trying to draw from the charcoal cannister and not getting flow due to a defective valve or the cannister itself deteriorating (they turn into a gooey mess inside).

Keep isolating; things that test positive are just as valuable as the 'not working' items.
 

Trout_Teaser

New member
Thanks for the tips Capt. Mike. I am sure glad my 84.5 Automatic Westy came without A/C. Not that there's anything wrong with A/C
icon_smile.gif
. My poor van just needs all the power it can get.

The rough running, missing and running rich problem has been resolved. I just installed a new Fuel Pressure Regulator, same part number as the old. She is now back to running smooth and clean out the back end.

I am sure the charcoal cannister is a gooey mess just due to age. A while back I checked part of the system because fueling at full flow of the gas pump when the tank was finally full would result in gas spilling out the filler neck. I will save working on that system for another weekend.
 

reddeer

New member
My 84 westfalia, automatic, a/c (not working), has been plagued with very rich running conditions for both myself and previous owner. PO wound up leaning up the mixture to compensate for the real problem to get about 13mpg. I replaced the head gaskets, water pump, o2 sensor, plugs/cap/rotor/wires last year. Halfway through that job I noticed the coolant temp sensor plug was poorly connected - I reconnected properly and voila, when the job was done I was back to 18mpg! Ran like a top for a year. This spring it's back to the old running rich - now about 8mpg. Van surges slightly but no bucking, some black smoke when you hit the gas hard. I've gone through every test I can find in Bentley and on the site - fuel pressure is good, the pin test for FI system shows everything within spec - including temp sensors and airflow meter. Replaced coolant temp sensor, aux air regulator, and 02 sensor as they seemed likely candidates. The next culprit to check is the charcoal canister system, but is this able to affect mileage to this extent? Any help would be very much appreciated. Thanks in advance.
 

Capt. Mike

Moderator
You haven't given the CO & HO readings so it's difficult to tell if your problem is air/fuel ratio or a fuel problem, either excess or poor injection.

Light surge indicates there is instability in the system. O² sensors cause 'hunt' but not surge during running. The most likely culprits are the intake air sensor, auxiliary air regulator and throttle valve/decel/idel switch or full throttle enrichment switch.

The intake air sensor sends a varying signal depending on load and throttle position to the ECU. Most testing is done at idle & full throttle; but the signal must be smoothly progressive and stable in the mid-range. It's a resistor -- it could be dirty or burned in the mid-range. It has it's own topic.

The auxiliary air regulator's primary job is to fake the ECU during cold start. It provides excess air (the opposite of choking) so that the FI brain sends even more excess fuel -- thus giving the higher fuel/air ratio for cold starts. (Choking reduces the air; this replaces the extra cold-start fuel injector of old systems.) However, note that it is in a vacuum flow that includes the idle speed boost valves, one of which is for your defunct A/C. If it is being told the A/C or P/S are "on", it adjusts accordingly. It has its own topic.

The two throttle switches serve different purposes but have the same possible effects. If the idle switch is sending a wrong or unstable signal, the FI will constantly adjust, shutting off & restarting. Full throttle enrichment increases fuel at full throttle. Understandably, if incorrect or unstable, it causes the FI system to adjust fuel at the wrong times. This test does require CO measurement, thus is often ignored by mechanics. Please note both of these switches are not expected to be exposed to water -- and 'engine cleaning' can take them out. They have their own topic.

I rather doubt the charcoal system is the culprit, unless the valve is shot and leaking. Testing it is in the Bentley in the fuel delivery system section 20. It's function is to open the purge line when the engine vacuum indicates the engine is able to handle the unburned hydrocarbons. There is a tiny restrictor in the control side (.006"). However, failure would usually either shut the charcoal filter supply off or leave it on all the time. Off should have no effect; on should purge the filter and then no longer vary the air supply. The charcoal system up to the valve is in the FUEL SUPPLY forum.

All of these presume you have no vacuum leaks; there are a lot of Digijet FI controls based on vacuum. It also presumes the ECU itself is OK; most of the tests monitor input to the ECU -- not its accurate processing of the info. I'll also ask if you did the actual tests of the injectors themselves. (Own topic) A leaking injector, faulty electric valve, or improper spray will reek havoc on FI running. This is a tough test and another many mechanics want to skip (engine runs; must be working mentality).
 

Trout_Teaser

New member
A few thoughts. 1. The PO may have done some other things to "fix" problems when all that was being done was masking the real problem. Take a look around for other not by the manual repairs. 2. I am curious how the correct mixture was determined after the temp. sensor lead was fixed. Expensive equipment is needed to get it right. Pulling plugs does not work on unleaded fuel engines. 3. If the engine is running super rich and 8mpg is twice what might normally be used. There most likly is a big drop in power. Did power and acceleration decline when fuel economy dropped? If performance did not change but fuel economy got bad maybe there is a fuel leak somewhere. It can be hard to spot a fuel leak in a hot engine compartment. 4. When "hitting the gas hard" the FI system will richen the mixture and change ignition advance. During this brief period not all the fuel will burn in the combustion chamber and darker exhaust will result. For some vehicles a little bit of "darker/richer" exhaust is normal. Most of the time with a good working catalytic converter a lot of the unburned fuel will be burned at the cat. before exiting the tail pipe and nothing will be seen. An infrared temp gun is a good tool for checking to see if the cat is "hot". Check temp at manifold and head then check Cat.. The Cat. should be much hotter if working. 5. I am not a pro on O2 sensors for these systems. In general the O2 sensor works at idle and part throttle. It sends a signal to the engine computer to richen or lean the mixture a small amount to maintain a ratio that keeps the cat. running correctly. Your small "surge" might be the computer trying to adjust an air fuel ratio that is well beyond what it can correct for.
 

reddeer

New member
Thanks to both for your suggestions. Given that the PO had presumably messed with the whole FI system, I had a shop time, tune, and adjust the mixture after my major head gasket/waterpump job. I'm sure there isn't a fuel leak - power is definitely low, but this is intermitant. The 'surging' seems to be the system coming on and off fully rich, so the loss of power comes and goes maybe every few seconds. This leads me to suspect the idle switches Capt Mike mentioned. I'm going to check leaks at the injectors, replace both of these switches, and failing that, swap the ECU and air sensor from another '84 van. If no luck after that I'll be doing the thelma and louise Canadian syle somewhere in the rockies.
 

reddeer

New member
Well here's my update. I replaced both the idle and throttle switches this week, but no change to my 8mpg. Onward. I switched the airflow meter with another van - drove only a short distance (10 miles) and could tell by the bogging down it was no better or arguably worse. Next, switched the CPU from another van, and again no improvement. Next I'll be testing the injectors for leakage and have the CO measured.
 

reddeer

New member
Well I feel almost defeated. Since last post I tested spray pattern and also switched injectors from the other van - no change to rich running. Also switched valve from charcoal system - nothing. Replaced ground wire that looked a bit ugly beneath the coil. I haven't had a shop measure the CO, nor tested the AC idle adjustment valve. Unplugged the Coolant temp sensor while engine running and hot and clearly ran worse, so safe to say that is not the problem. With new throttle/idle switches, that's safe to rule out as well. I had switched the ECU and Airflow meter and had no change (only ran van for 15 minutes, maybe should have tried for a few days?) The previous fuel pressure test at the regulator showed a constant pressure according to Bentley (30?psi), but I wonder if I should monitor this while driving for a few miles with a buddy. Also wondering whether injector wiring may be mixed up (that doesn't explain how it ran well all of last year though). At this point I'm about ready to fly Cpt Mike up to beautiful Canada to figure this out. Help?!
 

icarus

Moderator
It sounds like a major problem for you. Two thought reading through all your posts. First is might you have a plugged exhaust/cat. I had a cat blow its gut once, and the symptoms were similar. The other thought, although even farther fetched is, is it possible that the cam timing is off? Has the engine been rebuilt and has it ever run right since.

It's been a long time, but in the old days we used vacume gauges as diognostic tools, and in the right hand, they can tell you alot about things, even fancy electronic instruments may miss.

Good luck, Don't do the Thelma and Louis thing, we would have to get Alberta Mot to clean up the mess.

Icarus

PS After re-reading your last post I see that it has run right in the past. Not to state the obvious, but have you done a compression/leak down test, to make sure you don't have other mechanical problems (leaking valves or stuck/broken rings)?
 

Capt. Mike

Moderator
Reddeer, don't think that wouldn't be tempting since we just had two consecutive days of 101° & 103°F, but it would be cheaper to bring the car down to the Jimmy at Southern States (MECHANICS -- NC topic). You can't afford me -- once we old folks retire, we get real expensive for that nasty, 4-letter word, "Work". 'Sides, he's forgot more than about Vanagons & FI than I'll ever know. At least you could camp in my woods and get some decent Southern vittles while you're here. Ain't no grits, country ham or BBQ pork-pig in Canada.

Just what I was going to suggest, icarus. Time to go back to the beginning and basics with a leak-down test and some careful checking of some things like valve timing and distributor type & settings. I've done a couple of dummy things like once set a car's FI pump to time at TDC instead of the FI syncronizing mark on the pulley. Guess how much time I spent trying to adjust the FI to make up for that snafu!

Also set my own Vanagon distributor once for #1 TDC to be at the cap indent instead of the hash-mark and it ran! Not well, but ran.
 

reddeer

New member
At this point I'm not afraid of a stupid question. I haven't taken the time yet to check compression as suggested by Icarius and Cpt Mike - this weekend will do. Prior to my problems, my coil went bad this spring. I replaced it and all was good, but noticed that the elec plug to distributor (hall sender?) seemed loose - van would die if you jiggled it. Sounds from other posts that these two often die together. Any chance a bad hall sender is leading to the rich mixture?
 

treejay

New member
1983 water cooled vanagon L

Here's something interesting that I learned today which fixed a bucking/surging, running extremely rich problem I have been having. Given that the prior poster is trying to figure out what's wrong with their van, this is one more thing you can check on the long list of things that you've already checked.

About a month ago, I realized my O2 sensor wasn't conected. I bought a new one, installed it myself. here's the rub: I striped the green wire coming out of the FI multipin connector. I didn't realize that this wire was a shielded wire with an inside core, then insulation, an outside wire, then the outside green wire. It's just like coax cable. When I striped it, I mistakenly braided the inside wire with the outter shielding wire and then spliced both to the new O2 sensor. This eventually caused all these recent problems I've been having, namely running way too rich and bucking/surging. Today, I learned that the outter wire shielding is a ground connection and this shouldn't be braided with the inner core. I peeled the outter shielding wire back, and viola = van runs great again. I find it amazing that only a couple strands of copper improperly connceted or shorting out can cause such dramatic running problems.

I hope this helps someone else somewhere. good luck

ej
 

icarus

Moderator
EJ makes a great point. One of the "Hidden" problems with the O2 sensor/FI is damage/shorting of the O2 sensor wiring. Inspect your wires carefully, paying particular attention to the end of the O2 harness where it plugs into the car harness. Fairly often the shield braid will be bad and it will be shorting, causing intermittent problems. This is the first place that my Guru looks for with poor running issues.

A fairly good test for O2 sensor problems is if the car starts fine, but then runs poorly in 30+- seconds, but runs fine again after quickly turning the key off and on ro reset the O2 sensor default/start mode. You can also unplug the O2 sensor and it will default and the car will run fine. (You shouldn't leave it disconnected permanently however as this is an essential component, but it will allow you to get home.) If you have unplugged the sensor and it runs fine, look for damage to the O2 wiring or replace the sensor.

Icarus
 

reddeer

New member
EJ, Icarius, Cpt Mike: I can't thank you enough. EJ was bang on - the green wire connecting the 02 sensor to the ECU wiring had broken over time and the two layers of wires were making contact. Fixed and running normally for the first time in many years. Given my long winded struggle with this, we resorted to buying a tent trailer and decided to sell the westy. It only took one camping trip to change our minds and try to figure this out. Done deal. Thanks alot.
 

icarus

Moderator
Whoo-hoo!
Congrats of figuring it out. Too bad none of us thought of it sooner (Kudos to EJ!) It's frustrating to spend tons of time and tons of money fixing what turns out to be a simple problem.

Congrats,

Icarus
 

KenS

New member
Hello All,

I was returning from a camping trip the day before yesterday in my 84 Westy and started to experience surging, especially on hills. I was driving on quite rough road and so at first thought I may have compromised the fuel delivery system some how, and so pulled over. I could not see any gas leaking out, and so carried on to try and make it back to civilization. When I made it to a gas station and filled up I found that my gas consumption was roughly double what is usually is. I had no choice (my four year old boy was in the back) and so carried on for another hour and made it home. On the highway the surging was not as bad, but, there was infrequent backfiring on decceleration.

Yesterday, I started to do some trouble shooting with my multimeter (as per Bentley 24.28) and found right away that there was no continuity through the temperature sensor for the intake air sensor. I then was interupted (by the 4 year old and his 2 year old sister) and was unable to do any work on it until this morning when I saw that the temp sensor wire had broken. I fixed it. It checked out with the multimeter.

I am again watching my kids and so am unable to go through the whole system check (as per 24.20) but am suspecting that the temp sensor wire was the problem, as it was fooling the intake air sensor. Does that make sense?

I will hopefully be taking it for a test drive later today, and am planning another camping trip this weekend. Given my time constraints are there any suggestions on anything else I should absolutely make sure I check before I run the van again? I know that it isn't good to be running it when it is running so rich.

Thanks,

Ken
 

KenS

New member
Just a quick follow up.

I got lucky, and the broken wire on the temp sensor was the problem. The van is back to normal, noisy lifters and all. Hope I can get another year out of the ol' 1.9.

Ken
 

Top