ECU and electrical FI malfunctions


icarus

Moderator
I've done mine several times, battery connected with no ill effects. The bail/clip plug keeper can be a pain, and the plug is a bit funky, but it takes all of 5 minutes. Good luck.

Icarus
 

icarus

Moderator
After years of owning water boxers it seems that I should know the answer to this... How much air circulation do you need around the ecu? Is what's provided by the guard enough to keep it cool? I tend to stuff under the seat with all manner of junk, using every available cubic inch. I guess since I have never had problems before I shouldn't worry. (On the one hand it has cooling fins for a reason, but on the other hand it is enclosed under the seat, and near the heater). What's a guy to do?

Any thoughts?

Icarus
 

Capt. Mike

Moderator
Don't put anything in contct with it. Most came with either boxes or metal guards around them. As long as there's a couple of inches for air-flow and nothing is contacting it to act as an insulator, you should be OK. Heat needs to escape but it doesn't need much of a channel. Heat rises so keep a small channel clear above -- not sideways.
 

icarus

Moderator
Thanks Mike,


Since it is guarded, I figured that if I keep the finned end open, and nothing on top it should be fine. Never had a problem yet,,, but looking for every available inch.

Icarus
 

mystery11

New member
I have an '84 Westy that shorts out the ECU. The Bosch technicians say that there is a positive spike on one of the two grounds. Consequently, we're getting two of these units rebuilt and can't go forward for a few weeks. When it shorts, it fills the pistons with gas on all four cylinders. It does this after a taxing trip up hills or around town. Some months before I had trouble with the engine not starting after getting very hot in the engine compartment and after wiggling wires near the driver's side fuel-injection hoses would get it to fire. One day it quit completely, and as it was up in years and running a little rough from a weak lifter, I had it rebuilt. This behavior has started since the rebuilt and we have searched for bad grounds for over two months. The Bosch mech says that all the grounds in the engine compartment are good and is baffled. He also checked the throttle switches. Where should we start after the ECU rebuilds are done? I hate to do these again! They are expensive.
 

Capt. Mike

Moderator
Testing of the Digijet Ground connection/control unit is done between terminals 25 & 7, per Bentley 24.20 & 21. If you notice, 7 is involved in the vast majority of tests, including all of the injectors. It's also involved in starter enrichment. Thus I'd presume 7 is the primary ground for ECU functions and the likely place to start. It grounds at the left cylinder head. 25 is direct ground of the ECU. Notice in the wiring diagram that all of the ECU grounds go to that same point and a place to look for your transient spikes.

Since these are grounding to the cylinder head (engine) then there must be a continued ground of engine-to-chassis so that the current can complete circuit to the battery negative pole. The ONLY place in the entire electrical system that is over 12v (if that's what you mean by positive spike" is from the coil to the plugs. They ground through the same engine ground strap and since electricity always takes the path of least resistance, shouldn't back up into any other circuit. You state that has been verified as not the problem. Are you sure?

I don't proclaim to know the inner workings and voltages of the ECU. But to 'short out' infers that the current that was supposed to be going through some resistance (light, sensor, valve) is now going direct to ground. Almost all the components feeding information to the ECU are resistors of one form or another. Any of those tests that indicate an ohm reading, could be your short.

I assume the dire warnings in the Bentley, sufficient to call for a box & bold type aren't there to fill the page. "Caution: Do not disconnect terminal 1 at ignition coil . . ."; "To prevent damage to control unit, do not short-circuit connector contacts; "Ignition must be switched OFF before connecting tester; "Do not connect any test light to terminal 86 of adaptor C . . . if control unit is connected; and "Do not connect test light to throttle valve switch connectors if control unit is connected." The bold type is VW's. This is NOT to say your mechanic is deliberately doing those items, but there may be a condition or act that duplicates them. For example, when testing injectors, the warning box says not to short-circuit the connector contacts. But what if an injector/connector defect causes them to short internally? Doesn't mean the injector's valve is not working, but it may be overloading the ECU or have an intermittent arc fault. Some of the others may disrupt a captive circuit and trigger 12v where it isn't supposed to go. Test lights are just short circuits with a bulb in them. "Do not use test light" is, in essence, 'do not ground' or 'do not change resistance outside parameters by adding the bulb resistance'. But those circuits could be grounding through some other fault and replicating the very thing you're being warned against.

Electrical trouble-shooting is slow, tedious and there are NO short cuts. Break down each circuit into it's components and individual legs and test each independently.

Within the multiconnector and sensor connections, you have all leads terminating close to each other. The leads end in the connector with what should be a secured wire end, but I have seen instances where a defect or damage causes the connector to back out instead of slide on. This could be forcing it into another connector or short. Regardless, one does not replace a burned out fuse (or ECU) until they have determined why.
 

icarus

Moderator
Bruce,

I'm trying to get my head around your problem. Do I read you correctly in that you have blown up two ecu's? If this is the case I would think real hard before I plug in another (rebuild or otherwise) one. Please clarify,

Icarus
 

mystery11

New member
I'm afriad you read correctly. And I also burnt the original unit as well. Thankfully (or maybe not) I live where the labor is cheap so far. There has been so much done to the engine in the last few months that I'm tempted to just bring it home and work it over very closely myself. I wish I didn't have another job to go to. Thanks for your advice!

Bruce
 

icarus

Moderator
Capt. Mike and I have had a brief conversation off site about this, and I confess I don't have any real answer. My first inclination would be to check the expertise of your Bosch guy. Blowing up two ecu does not sound like good procedure. I don't want to cast aspersions on anyone, (hell, I've been know to blow up all kinds of stuff!) but it is suspicious.

The next thing I would do is do a VERY careful tracing and testing of the entire wiring harness that connects to the ecu. I would pay particular attention to the plugs. I suppose it is possible that there is some corrosion on a multi-pin connector that could allow voltage to pass the wrong current path. I'm not pointing out anything here but the obvious, but electrical gremlins can be very hard to find. I am not very familiar with the digijet ecu, but I assume that it plugs into the harness with a multi-pin connector.

If the injection is sending a signal enough to "fill the cylinders with fuel" clearly something is getting a crossed message. I had a '70's vintage saab onetime that exhibited a very similar problem. It poured pure fuel into two cyinders. It turned out that the "brain" (as it was called in those days) was faulty.

The fact that you have burned up two ecu' tells me that the problem is not with the ecu's but lies elsewhere. I suppose that an improper repair to an ecu could cause this however.

Good luck,

Icarus

PS. After re-reading you original post, you give one clue that may be worth following. You mention a problem that was temperarily cured by "wiggling the wires". It seems that if you have had that kind of trouble in the system in the past, I would look at all the connectors.
 

mystery11

New member
Thanks so much for the further information. I sent the two ECU's off for repair today and anticipate them back in about two weeks. We'll start again when they return.

I'll keep you posted and thank-you in advance for your brain-power.

I think I have the best mechs in Naples, Italy looking at this one and even had a mech fly down from Germany to take a look and give a few suggestions. No one has seen this problem on this set up. Of course the fuel injection is not the norm here either in this year (North American VW and Porsche 911). I think it will take a patient soul to find it.
 

mkollerjr

New member
Starting Problem (copied from Vanagon Syndrome thread)

Captain Mike,

I hope I'm in the correct thread now (see below for text copied out of the Vanagon Syndrome thread). I've begun going through the FI test in Bentley 24.61. For test 4, you are supposed to be able to bridge points 3 & 13 on the engine harness with the ignition ON and hear the fuel pump running - but I don't hear it running. However, when I plug back in the ECM and turn the ignition ON I can hear the fuel pump turn on for about a second. My van will turn over fine, but now it won't start, whereas before it would at least start and run roughly if I kept it above 1500 RPM.

Thanks,
Mark


It's NOT Vanagon Syndrome. With Vanagon Syndrome, it ECU will reset, start and run fine for a period. See the factory tech bulletin in the first post in this topic that describes the symptoms. You have other FI problems covered by the other topics in this forum. Have you done any diagnostics per the Bentley or those topics?

Captain Mike,

I have a 1991 Vanagon GL full camper with AT and about 155,000 miles. I've only had the van a few weeks and it has ran flawlessly - until yesterday. I recently went on a 900 mile road trip and returned home with no problems. When I went to start the van the next day it turned over, started, and idled very roughly and sporatically (400-2000 rpm) then it died. I tried starting it a few times with the same result. Sometimes it would start and sound ok - but then when I put it into gear it would die. I've read the posts on the Vanagon syndrom and I installed the replacement harness, but the same problem still exists. Any suggestions?

Thanks,
Mark
 

Capt. Mike

Moderator
I'm sorry, but other than that you replaced the Vanagon Syndrome harness, and therefore presumably eliminated Vanagon Syndrome as a cause, we don't have anything to go on (Guideline #3). You have a Digifant; what other testing, troubleshooting or repairs have you done? :confused: There is a detailed Electrical testing chart on pg 24.61-24.62 in the Bentley to start with that requires little more than a volt-ohmeter and a bridge wire.
 

mkollerjr

New member
Captain Mike,

The current symptoms on my 1991 VW Vanagon GL full camper (AT / 155,000 miles) are that it will turn over fine but will not start.

I ran through the FI tests described in Bentley 24.61 - 24.62 with the following results:

Test 1:
11.3 - 12.2 V between pins 13+14 and 14+19 (OK)
Test 2:
10.60 - 10.65 V between pins 13+25 (OK)
Test 3:
9.6 V between pins 1+13 (OK)
Test 4:
Bridge pins 3+13 with ignition ON and DON'T hear fuel pump running (NOT OK?) (With the ECM plugged back in and the ignition turned on I can hear the fuel pump cycle for about one second and then turn off - is this normal?).
Test 5:
I couldn't get any reading that was close to 15-20 ohms.
Test 6:
The steps in Bentley aren't realy outlined well here, but I ran the test described in Bentley 24.57 with the results OK. Was the the correct procedure?
Test 7:
Throttle valve = Closed 1.2 ohm / 75% open (infinity) / Fully open 1.0 ohm (OK?)
Test 8:
Pin 6+17 = 5.6 ohm (OK) / Pin 17+21, moved sensor, ohms changed (OK)
Test 9:
Pins 6+9 = 2500 ohm at 68F (OK)
Test 10:
Bridged three pins on Hall sender connector. Pins 6+8 = 0 ohm / Pins 6+18 = 0 OHM (OK?)
Test 11:
Pin 2+13 = infinity ohm (OK)

Sorry for the lengthy reply, but I'm trying to narrow things down a bit...
 

Capt. Mike

Moderator
That was an excellent and correct reply -- you did the diagnostics necessary for anyone else to help you. I'm not trying to be facetious, but you also answered your own question. Tests #4 & #5 failed, so that's where to start looking. The fuel pump and current supply relay are very much interdependent so that's the most logical place to start. Test 5 requires these items in test #4 to be operational.

To test a sensor, it's often easiest to remove & bench test by heating in water per the procedure outlined elsewhere on the site -- I think in Cooling System. 3rd in your list of 3 to check.
 

mkollerjr

New member
Captain Mike,

As for test #4, I'm still confused. If I connect the ECM back up, pull the fuel hose off of the regulator and turn the motor over, there is plenty of fuel coming out. Is the purpose of test #4 to see if the fuel pump is getting electricity? Because the pump seems to be functional when I turn the motor over. It seems like the engine is not getting a spark.

Thanks,
Mark
 

Capt. Mike

Moderator
If you've got fuel to the injectors and the plugs aren't getting spark, then that's another forum. You had originally said the fuel pump stopped after a minute.
 
Engine Control Unit (computer)

According to my mechanic, who has tried every test and possible work-around under the sun, my 1995 Eurovan (2.5L, gas, standard) needs a new ECU. :( "Its dumping too much gas," he says.

I have searched high and low, at innumerable wreckers and parts shops to no avail, and VW Canada says the necessary part (which costs $1000+ whether new or rebuilt) is backordered in North America, meaning there is no sure way to get one.

Is there some way to acquire this very specific part that I have overlooked? Are there any reliable / reputable ECU rebuilding services for VWs?

Thanks for your time, and please do move this to a different section of the forum if necessary.
 
Last edited:

luvshak

New member
fuel injectors not firing

I have a 1981 air cooled gas vanagon
I brought the van in to a shop for a state inspection, it was running fine when I brought it in and left it there. I was called two days later and was told there had been a small fire in the engine compartment. I was lucky. It had done very little damage. It happened because the cold start valve had broke and was leaking. The cold start valve, of course, had to be replaced, small amount of wires along with the fi connectors to the cold start valve and a few vacuum hoses. Then I was told that my combo relay might be bad, so I ordered one and replaced that. Now I'm told that the fuel injectors are not getting any pulse. Could anyone give me any advise or should I just have AAA tow it to my regular vw man. I told them to hold off on any more work until I get advice from a VW professional because I do not want to pay for hours of just guessing.
Thanks;
LUVSHAK
 

tigergary

New member
My 91 Vanagon Westfalia has a failed ECM (ECU?) which has a "D" suffix after the number, and we have not been able to find a remanufactured "D", but we can find a non "D". My mechanic has installed a used non "D" and says it runs fine, and that I would not notice the difference. What are the implications of replacing a "D" with a non "D"?
 

Top