81 vanagon missing under load over ~2200rpm

Find/Fix/Report Back

Moderators: Sluggo, Amskeptic

Firstclassdumbo
I'm New!
Status: Offline

81 vanagon missing under load over ~2200rpm

Post by Firstclassdumbo » Tue Sep 19, 2017 3:20 pm

This is going to be a bit lengthy in description, so bear with me.

Two weeks ago, I took the van into a shop to get the pilot bearing replaced. They replaced the pilot bearing, and in the process hooked up the full throttle enrichment switch. I had disconnected that switch because for some strange reason, whenever the van reached that 3000 rpm full throttle spot, it would run rough and cause the van to slow down if I were on a steep grade. Anyways.

So I disconnected that switch again, and began a drive from Bishop to Sacramento. About fifty miles outside of Placerville, the van started acting like the full throttle enrichment switch was still hooked up. Except it didn't start at 3000 rpms. It would start at lower rpms than that. I can't give specifics, but I'm guessing somewhere around 2200. If I pulled out of full throttle the slightest bit, it would drive normally again. If I was at a low rpm, it wouldn't have an issue at all. I managed to drive the rest of the way to Sacramento. At idle, the engine runs fine most of the time. SOmetimes after holding the throttle open, it'll idle around 650 rpms for no reason.

I reached Sacramento. The next day, I drove the van again. The problem became more apparent. This time, I could only go half throttle above 2200 rpms. Anything more than half throttle and it would begin stuttering again.

When it is idling and no load, if I slam the throttle to full, it gets up to 2400 rpms and starts hunting, blowing black smoke in a pulsing rhythm with the engine. if I slowly back the throttle off from full at that point, it'l;l slowly increase the rpms as I back the throttle off. It is as if the full throttle is dumping too much fuel. I tried adjusting the clock spring in the afm six clicks CW with no noticeable difference.

Here is what I have eliminated so far. I have checked for proper valve adjustment. I timed the engine, removing the idle stabilizer and setting timing to 5ATDC at 875 rpm. I can't efficiently time it at 3400 due to this issue. It will not hold a steady rpm there.

I changed the fuel in the tank after noticing that it smelled incredibly alcoholic. It now has a full tank of shell 87 in it. I checked the fuel filter to see if it was plugged and found nothing. I checked idle fuel pressure at 29psi.

I tried running the engine without the O2 sensor (ca model). I tried cleaning the O2 sensor and reinstalling. The O2 sensor gives a proper reading when testing.

I cleaned all the grounds that connect to the engine under the intake plenum. I bypassed the cold start injector. Also, the egr system is bypassed and closed off.

I tested the temp 2 sensor by the test listed in the manual and it passed. I can't remember specific numbers for that one.

I tested the AFM box per specs in the Bentley manual. The resistance across the board jumps between 100 and 700 ohms when moving the arm across the board. I reset the black strip to a new spot by moving the circuit board. The static terminal tests (from 6-9 and 6-27 and so on) all passed.

I can't test any of the wiring to the ecu because the plug on my van doesn't match up to the plug shown in the bentley manual. This may not matter anymore, but the engine was rebuilt at 90000 miles. The only thing that wasn't rebuilt was the case. Everything up from that is new. The electronics are all the age of the van, I'm assuming.


So that's where I'm at. Any thoughts or ideas are welcome. I'm about ready to throw a warehouse of parts at it and hope for the best.
Dumbo- 1981 vanagon/house/daily driver

User avatar
Amskeptic
IAC "Help Desk"
IAC "Help Desk"
Status: Offline

Re: 81 vanagon missing under load over ~2200rpm

Post by Amskeptic » Wed Sep 20, 2017 7:23 am

Firstclassdumbo wrote:
Tue Sep 19, 2017 3:20 pm
This is going to be a bit lengthy in description, so bear with me.
So that's where I'm at.
CURRENT STATE
It is running too rich. The full throttle switch only exacerbates the problem, yes? The oxygen sensor tries to keep everything under control, but it is bypassed at full throttle.


EXPERIMENT
Disconnect oxygen sensor for the entirety of this experiment. Please count the number of turns it takes to bottom out the mixture screw on the AFM, Return screw back to where you found it and let us know how many turns.

Test drive. Get a baseline of engine operation with no oxygen sensor, what rpm does it stumble now?

Now, make an initial mark on the black cog so you can go back to where you started. Turn the black cog 6 clicks CLOCKwise. Test drive. Better? Worse? Report back.
Colin
BobD - 78 Bus . . . 112,730 miles
Chloe - 70 bus . . . 217,593 miles
Naranja - 77 Westy . . . 142,970 miles
Pluck - 1973 Squareback . . . . . . 55,600 miles
Alexus - 91 Lexus LS400 . . . 96,675 miles

User avatar
SlowLane
IAC Addict!
Location: Livermore, CA
Status: Offline

Re: 81 vanagon missing under load over ~2200rpm

Post by SlowLane » Wed Sep 20, 2017 8:28 am

Also don't assume that it is an electrical issue. " False air" is the bane of l-jet, and the self-correcting nature of the lambda feedback loop can mask the effects of a pretty serious vacuum leak, up to a point. Be prepared to do a thorough survey of the usual sources of air not being measured by the AFM.

It's pretty odd that your ECU plug doesn't match the Benteley. How does it differ? Number of pins? Pin layout?
'81 Canadian Westfalia (2.0L, manual), now Californiated

"They say a little knowledge is a dangerous thing, but it is not one half so bad as a lot of ignorance."
- Terry Pratchett

Firstclassdumbo
I'm New!
Status: Offline

Re: 81 vanagon missing under load over ~2200rpm

Post by Firstclassdumbo » Wed Sep 20, 2017 2:34 pm

Colin. I agree with your diagnosis.

Disconnected O2 sensor. There is no mixture screw on the afm housing. There is a spot where it should be, but instead of a screw, it is a recessed flat surface. Don't know if this is common. It's a seven pin afm though. It runs exactly the same with O2 disconnected.

I moved black cog six clicks clockwise. Engine would not start. Moved cog back to original position. Engine started.

Slowlane, I inspected all of the vacuum lines. The aux air elbow is new, the two distributor lines are new, the fuel reg lines are new, and the other ones are old but passed the carb cleaner spraydown test.

In response to the ECU plug: I don't have a connector in pin 18. Which could explain full throttle enrichment (fte), but I don't know. Either way, one of the wires that plugs into the fte switch runs to pin three. The other doesn't show up in any pin when I do a continuity check. Every other check that the Bentley manual suggests seems to be okay.
Dumbo- 1981 vanagon/house/daily driver

User avatar
SlowLane
IAC Addict!
Location: Livermore, CA
Status: Offline

Re: 81 vanagon missing under load over ~2200rpm

Post by SlowLane » Wed Sep 20, 2017 6:09 pm

Firstclassdumbo wrote:
Wed Sep 20, 2017 2:34 pm
There is no mixture screw on the afm housing. There is a spot where it should be, but instead of a screw, it is a recessed flat surface. Don't know if this is common.
See if there's a removable plug in the recessed flat surface.
Slowlane, I inspected all of the vacuum lines. The aux air elbow is new, the two distributor lines are new, the fuel reg lines are new, and the other ones are old but passed the carb cleaner spraydown test.
I got burned by a new AAR elbow once. It was less than a month old when it failed. The common aftermarket ones are crap. Source a genuine VW one if you haven't already.
In response to the ECU plug: I don't have a connector in pin 18. Which could explain full throttle enrichment (fte), but I don't know. Either way, one of the wires that plugs into the fte switch runs to pin three. The other doesn't show up in any pin when I do a continuity check. Every other check that the Bentley manual suggests seems to be okay.
The "other" wire from the WOT/FTE switch goes to one of the terminals on your OXS relay (they're wired in series), which is why you wouldn't have seen it at the ECU plug.

Also, something that might lead you astray is if your vacuum retard isn't working. The 5 degree ATDC timing is assuming that the vacuum retard is in good order. If you time it to that value with a leaking retard can, then the timing will be about 13 degrees too retarded.

If you're in the Livermore area this weekend, I can give you a hand troubleshooting this.
'81 Canadian Westfalia (2.0L, manual), now Californiated

"They say a little knowledge is a dangerous thing, but it is not one half so bad as a lot of ignorance."
- Terry Pratchett

User avatar
sgkent
Addicted!
Location: Citrus Heights CA (near Sacramento)
Status: Offline

Re: 81 vanagon missing under load over ~2200rpm

Post by sgkent » Thu Sep 21, 2017 8:57 am

To qualify myself, Lyle and Colin are much more informed on the AC Vanagon FI than I am so their opinion has much more weight. One thing that occurred to me is that if any of the late AC motors have an electronic idle adjustment valve that bleeds in air, or an electronic EGR, the throttle enrichment switch might be involved in their logic - hence the switch might be the trigger for the actual error but it is in another component that is changing the mixture.
TBone208 wrote: "You ppl are such windbags. Go use your crystal ball to get rich & predict something meaningful. Nobody knows what's going to happen. How are we supposed to take ppl who don't know the definition of a recession & "woman" seriously?"

Merlin The Wrench

User avatar
SlowLane
IAC Addict!
Location: Livermore, CA
Status: Offline

Re: 81 vanagon missing under load over ~2200rpm

Post by SlowLane » Thu Sep 21, 2017 11:51 am

sgkent wrote:
Thu Sep 21, 2017 8:57 am
To qualify myself, Lyle and Colin are much more informed on the AC Vanagon FI than I am so their opinion has much more weight. One thing that occurred to me is that if any of the late AC motors have an electronic idle adjustment valve that bleeds in air, or an electronic EGR, the throttle enrichment switch might be involved in their logic - hence the switch might be the trigger for the actual error but it is in another component that is changing the mixture.
Thanks for the props, Steve.

The Vanagon aircooled FI system isn't much advanced over the late bus system. The EGR is the same mechanical one as on the late bus (except, oddly, the 1980 California Vanagon, which had the EGR omitted altogether, only to see it make a comeback in the 1981 model). The Vanagon also used the same bimetallic-spring AAR as did the bus systems, so no Idle Air Control valve to worry about.

To the best of my knowledge the WOT/FTE switch does exactly the same job as on the bus systems as well, only on the CA-emission spec vehcles it's wired in series with the OXS relay to only provide full enrichment over 3000 RPM (to protect the cat from too much richness). I can attest to this behavior from watching my AFR gauge whilst driving with foot to floor.

It does sound as if this is a largely unmolested original FI system, though, if the idle mixture adjustment screw on the AFM had never been fiddled with under its anti-tampering plug.

Mr. Firstclass (clearly you ain't no Dumbo): have you eliminated basic things like spark plugs/wires from the equation?
'81 Canadian Westfalia (2.0L, manual), now Californiated

"They say a little knowledge is a dangerous thing, but it is not one half so bad as a lot of ignorance."
- Terry Pratchett

User avatar
sgkent
Addicted!
Location: Citrus Heights CA (near Sacramento)
Status: Offline

Re: 81 vanagon missing under load over ~2200rpm

Post by sgkent » Thu Sep 21, 2017 12:02 pm

thinking about that and confessing I have not read all the posts thoroughly, Bosch logic would be to use the O2 sensor to hold 14.7:1 except at WOT. On California bays the ECU cuts off fuel at a set RPM (5400) not to cause lean misfires that might overheat the cat. The injectors would only handle up to a set flow and above that RPM the mixture would lean. Following logic the WOT sensor switch would probably be a lean enrichment below 14.7:1 to prevent overheating and detonation. The enrichment could be accomplished two fold - tell the ECU to ignore the 02 sensor and go to a preset flow, or it could be listen to the O2 sensor but add extra a preset value of fuel in addition to what ever it says. Either way if the FPR was over pressure the additional flow could be more than the fuel mapping set because the fuel mapping would assume a preset pressure. Therefore I would suggest the owner thoroughly test his FPR to be sure it is still within spec, and that the vacuum lines to it and the decal are intact. I would also include the decal diaphragm in that test. Over pressure on the fuel system might be compensated by the O2 sensor enough at cruise that the bus would run reasonably well, and only show up when the enrichment is triggered. Just a thought.

Also there have been a few bay posts on TS over the years where the FPR internals degraded in a way that let fuel be pulled into the TB thru the lines. That would be exacerbated on WOT.
TBone208 wrote: "You ppl are such windbags. Go use your crystal ball to get rich & predict something meaningful. Nobody knows what's going to happen. How are we supposed to take ppl who don't know the definition of a recession & "woman" seriously?"

Merlin The Wrench

Firstclassdumbo
I'm New!
Status: Offline

Re: 81 vanagon missing under load over ~2200rpm

Post by Firstclassdumbo » Thu Sep 21, 2017 1:04 pm

Firstly, I'll address Lyle on his reply last evening. I greatly appreciate your offer to help troubleshoot this beast in Livermore, but I really don't think the van will be able to make the trip. It can't do 55mph currently. At least right now I have a semi-running engine located at a friend's backyard shop. This van is my house, so a semi-running engine can at least get me to grocery stores and the like.

I took a risk and drilled the recess on the AFM, then used a slide hammer to remove a 1/2" aluminum plug and expose a screw. So yay for that. The screw turned down two complete turns before bottoming out. I reset the screw back two turns up.

The vacuum on the distributor can did not leak at all on either side during the one minute leak tests.

The AAR elbow is not purchased from a dealer. I actually made one myself using an elbow for a spark plug boot off of a different vehicle and some aluminum tubing to ensure proper sealing on the non-AAR side. I'm positive it isn't leaking there.

When the engine starts cutting out at 2000 rpm, it does it very quickly. I initially believed it was a spark issue. I swapped coils with a known good one and pulled a plug wire to see what happened to the spark at that rpm. The spark was solid throughout the stumbling. When idling, I pulled the plug wire from each cylinder to verify a noticeable negative change in the engine before putting it back.

As for the FTE: In the bentley manual, there is a test for FTE. It says to connect an ohmmeter between pins 3 and 18 on the ecu plug with the pedal pushed to the floor in the van. I didn't push the pedal to the floor in the van, but I did toggle the FTE switch mounted on the throttle body. Same thing, I would think. When doing that test, nothing shows up. That led me to do the continuity tests on both wires that I did prior. Correct me if I'm wrong, but if I should get 0 ohms when toggling the FTE switch at pins 3 and 18, shouldn't I get continuity from FTE switch to two different spots on the ecu plug (normally pins 3 and 18, but I don't have a pin 18 so 3 and something else)?

Sgtkent I assume you're referring to the deceleration valve when you say "decal." I tested the FPR according to the l-jetronic instructions on the ratwell site: https://ratwell.com/mirror/potts/index.html. It passed. I will do a test on the decel valve today just for some giggles.

To give you guys an idea on my mechanical knowledge, I have been doing mechanics for a career fifteen years, although primarily on locomotives. I was the rebuilder of this engine and have driven the entire country on it in the past year. My friend, the place I'm crashing at right now, he was an incredible auto shop teacher for a local high school and has a sweet spot in his heart for the Porsche 914. He is just as stumped as I am. We have been leaning towards a faulty ecu, but have no way to test it. And the van only has 125000 miles, so I'd be really surprised if that was the case.

Thank you guys for your time and help. Words cannot express my appreciation.

-derrick
Dumbo- 1981 vanagon/house/daily driver

User avatar
sgkent
Addicted!
Location: Citrus Heights CA (near Sacramento)
Status: Offline

Re: 81 vanagon missing under load over ~2200rpm

Post by sgkent » Thu Sep 21, 2017 3:09 pm

when is the last time you replaced the fuel filter?
TBone208 wrote: "You ppl are such windbags. Go use your crystal ball to get rich & predict something meaningful. Nobody knows what's going to happen. How are we supposed to take ppl who don't know the definition of a recession & "woman" seriously?"

Merlin The Wrench

Firstclassdumbo
I'm New!
Status: Offline

Re: 81 vanagon missing under load over ~2200rpm

Post by Firstclassdumbo » Thu Sep 21, 2017 4:38 pm

May. I pulled it off two days ago and blew backwards through it in front of a paper towel. Nothing but fuel. Easy to blow through.
Dumbo- 1981 vanagon/house/daily driver

User avatar
Amskeptic
IAC "Help Desk"
IAC "Help Desk"
Status: Offline

Re: 81 vanagon missing under load over ~2200rpm

Post by Amskeptic » Fri Sep 22, 2017 8:11 am

Firstclassdumbo wrote:
Wed Sep 20, 2017 2:34 pm

I moved black cog six clicks clockwise. Engine would not start. Moved cog back to original position. Engine started.
As per your subsequent post, you found the mixture screw and it was at a typical 2 1/2 turns from bottom.

As per the above experiment, I KNOW your creative and inquisitive mind went ahead and moved the black cog 6 turns in the opposite direction (counterclockwise) from "0" to see what would happen since we now know that you pretty much fell off the edge of the operating parameter at 6 clicks lean. So what happened when you went 6 clicks in the opposite direction?

EXPERIMENT
You are familiar with hard/soft failure points? Ignition is usually a hard failure (both intermittent and total failure) that is harsh to the engine, fuel is usually soft. You have determined that the ignition system seems to be fully operational through the failure zone, but keep in mind that vacuum advance rotation inside of the distributor has sometimes grounded out the points/Hall Effect trigger only at a specific load (not just rpm like centrifugal advance would) point.

Rev the engine and try to tickle the wiper either rich or lean at the stumble point (ox sensor disconnected). Any improvement/deterioration noticed?
Colin
(be happy that you do not have a 1992 Lexus throwing codes "if code 25 occurs, test the transmission ecu spark retard circuit before assuming sub-sensor failure . . . ")
BobD - 78 Bus . . . 112,730 miles
Chloe - 70 bus . . . 217,593 miles
Naranja - 77 Westy . . . 142,970 miles
Pluck - 1973 Squareback . . . . . . 55,600 miles
Alexus - 91 Lexus LS400 . . . 96,675 miles

Firstclassdumbo
I'm New!
Status: Offline

Re: 81 vanagon missing under load over ~2200rpm

Post by Firstclassdumbo » Fri Sep 22, 2017 12:57 pm

So moving the cog six clicks ccw does nothing. In fact, I can move the cog half a turn ccw from zero and that is the point where it begins to stumble at idle. The stumble is at roughly the same spot regardless. Maybe instead of hunting at that rpm it'll steadily stumble, but it still shoots black smoke and refused to increase rpm.

I'm familiar with hard/soft failure points. The failure is hard.
Today, we hooked up an ohm meter to the hall generator plug on the side of the distributor with the other end grounded to the distributor housing. We then moved the mechanical advance super slowly, testing each of the three pins in the connector with no signs of grounding.

I pinned the throttle wide open to get ~2500 rpms and tickled the wiper seductively in both directions. Tickling it to the left resulted in more stumbling. Tickling it to the right just over 1/4" made the engine rpms skyrocket, like it should do when snapping the throttle wide open. Oxs still disconnected.

Drive that Lexus into a lake and drink a glass of scotch while watching it gurgle it's last breaths.
Dumbo- 1981 vanagon/house/daily driver

User avatar
satchmo
Old School!
Location: Crosby, MN
Status: Offline

Re: 81 vanagon missing under load over ~2200rpm

Post by satchmo » Fri Sep 22, 2017 2:27 pm

Does this have the double relay like the late bus? And the behavior started after the engine was removed and replaced, right?

Check that all the pins in the double relay connector are making positive contact when the plug is inserted, and not being pushed out of the back of the connector block. Failure to make good contact on all the pins can cause intermittent stumbling/missing at certain engine speeds and harmonics, almost like you were turning the ignition off and on rapidly.

If you do find this to be at least a part cure, check the condenser/points, if you have them, because this missing behavior can cause pitting in the points contact surfaces.

Good Luck,

Satchmo
By three methods we may learn wisdom:
First, by reflection, which is noblest;
second, by immitation, which is easiest;
and third, by experience, which is bitterest. -Confucius

Firstclassdumbo
I'm New!
Status: Offline

Re: 81 vanagon missing under load over ~2200rpm

Post by Firstclassdumbo » Fri Sep 22, 2017 4:50 pm

Satchmo, it does have the double relay. I checked all the pins and then followed the method for testing it in the bentley. When touching the test light to one of the connections, I touched it to another one accidentally and fried the double relay.

I did not find the double relay to be part of the cure, but now it is part of the problem. If anyone has a suggestion for purchasing a reliable one of these from someone other than busdepot because they want 85 bucks, I'm all ears.
Dumbo- 1981 vanagon/house/daily driver

Post Reply