85 Vanagon Digijet

[Amskeptic]

I am waiting to see if satchmo's experimental stall kills the pump after the engine has been running.

This morning's report directly from the garage:

The pump runs at key on (accessory position) prior to engaging starter. After stalling the engine, with the key in the same position, the pump runs for a fraction of a second, then stops. The only thing I hear running after that is the idle control valve, which I don't believe the Digijet system has.

Helpful?

Tim

Very.
That's a nice tidy improvement over the bays' use of AFM contacts and wires and nonsense.
Colin

[Lanval]

OK,

I worked through the Digijet Protraining manual steps (with one exception I'll check a bit later ~ it needs two people) and it appears that the ECU is bad.

I still need to do several things before I can comfortably put a new ECU in and try to run the van.

I'm still not sure about the alternator, so I'm going to pull it and have it checked out.

There are several wires that are not connected to anything, but clearly should be. I want to track those down and get them connected.

I want to solder in my jumped wires and clean grounds/connections. Right now I have everything twisted together and wrapped with tape to just check things out ~ might as well do a good job in that arena.

****************

There is a rebuilt harness available for about $400; it's tempting, and I'm thinking about it. Would make things simpler to start from ground zero with decent wiring.

Best,

L

[Amskeptic]

OK,

I worked through the Digijet Protraining manual steps (with one exception I'll check a bit later ~ it needs two people) and it appears that the ECU is bad.

There is a rebuilt harness available for about $400; it's tempting, and I'm thinking about it. Would make things simpler to start from ground zero with decent wiring.

I am loathe to ever agree with you here. My one and only claim/assertion/prognosis of a bad ECU was stopped at the 11th hour, as I was preparing to pull the ECU, I spotted a little bitty connector in the plug that had been pushed backwards and was not making contact. It was visually evident only after I started looking, All of those little fingers should be a clean row on that big connector, the one I was dealing with was the ground signal from the coil and it was just a little pushed in.

A nerw harness sounds like a good idea.

[Lanval]

Well, I'm still wondering if the ECU was fried by:

The alternator, which I still think is suspect.

or

a bad ground/shorted wire that I knocked loose when working over there.

In any case:

Putting in a new ECU won't solve the problem if either the wiring/alt issues aren't dealt with.

That said: The continuity checks out ~ unless I'm using the multimeter wrong, then there aren't any continuity problems. I tested continuity on the ground pin out out from the ECU to the ring on the head. It tested correctly, showing some small amount of resistance, but with continuity.

Everything else tested OK too, and the ECU/Fuel relays are working properly and getting power. The only thing that seems to not be happening is the signal out from the ECU switch on the ECU relay to power the fuel relay.

I'm open to suggestions. The harness isn't cheap, and work is slowing down...

L.

[Amskeptic]

The continuity checks out ~

The only thing that seems to not be happening is the signal out from the ECU switch on the ECU relay to power the fuel relay.

Signal from the ECU "switch" on the ECU relay?

Heck, you can make your own fuel pump relay circuit. 86 is the control voltage yes? Run it off the coil #15 Just to get you around for now.
Colin

[Lanval]

The continuity checks out ~

The only thing that seems to not be happening is the signal out from the ECU switch on the ECU relay to power the fuel relay.

Signal from the ECU "switch" on the ECU relay?

Heck, you can make your own fuel pump relay circuit. 86 is the control voltage yes? Run it off the coil #15 Just to get you around for now.
Colin

I agree, the ECU switch comment doesn't make any sense. I'll go back on Wed AM and work through the checks. It looks to me like everything is powered, and everything is connected (i.e. continuity check) up to the ECU, so I am assuming that the white wire in from the #30 on the ECU relay is hitting a dead end inside the ECU.

As for the fuel relay, once I realized the system was powered by jumping the #30 on the fuel relay, I figured that I could easily add in a power/kill switch if I wanted to just run it the way it is, as well as mess with thieves. But since I have 4 weeks vacation coming up starting next week, I thought I'd work through it diligently and slowly so that I:

1. Actually solve the problem without throwing money/parts at it will-he nil-he
2. Develop a detailed understanding of the electronics of this beautiful yellow Rattivan so that I can solve issues on the fly later.

Trust me, when I realized that I could drive the van with a couple of ground wires, ignition electronics and some 12v to the pump, I was pretty impressed.

Yeah, I can make it run, but can I make it run right?

After all, if I don't master this beast, how can I ever start the Peripatetic Water-Cooled forum?!

L.

[Lanval]

After studying the diagrams in the digijet manual and the Bentley (which are slightly different, I might add) I think there may be a missing ground from the ECU.

The digijet manual tells you to check terminal 7 to the cylinder head for continuity. There is however another ground wire shown in both the Digijet manual and B. It is potentially missing, and I'm wondering if that's a/the problem.

As far as I can tell, everything around the relays is working, except the fuel relay.

When ignition is on, everything shows voltage with two exceptions:

On the ECU relay, the ground wire which completes the opening of the relay shows no voltage. But I don't know if it should or not... it seems like it would, since it's sending 12v from the ignition through the relay to open the switch, and then back to ground.

On the Fuel relay, the dual wire that goes to the fuel injector number 2 and the fuel pump has no power when the ignition is on.

Currently, I have a jumper wire in place connecting the 85 (ignition power in) on the fuel relay to the 87 out (injector/pump).

I check continuity on the wire from the relay to the injector ~ it is connected.

My confusion is this ~ when the ignition is powered with the correct set up, I can verify 12v on the 86 side of the fuel relay. This should indicate that the relay is open.

However, I'm wondering if the the 86 (which goes into pin 13 of the ECU) is grounding out inside the ECU because there is a (potentially) missing ground wire...

In this case, Colin's assessment would be correct.

By the way, the van runs pretty well with the jumper in place. Works for now ~ next week I have vacation, so I'll be cleaning grounds and seeing if I can do some rewiring to fix the problem.

Updates as work occurs....

L.

********************

Question about continuity checking:

It looks like my multimeter has a setting for continuity, but if so, it's not clear how it works. I tested it on a known good relay so I could verify how it works, and continuous or broken showed the same number. Therefore, I've been checking continuity using the Ohms setting. The multimeter shows 1 (I assume this means infinite resistance) when not connected, but then drops to other numbers for resistance settings ~ the new relays, for example showed a resistance value of 79.2 or so between the 85 and 86 poles. The 25 yr old original VW/Audi relays showed a higher value of 82.9 or so. Is using the Ohm setting OK for continuity, or do I need to figure out the correct method for my multimeter (manual is MIA, and Sears doesn't have the correct manual online).

Question about measuring Amps:

In order to verify the ground for the fuel pump, the Digijet manual states I need to:

"Connect ammeter to fuel pump in series with power supply to fuel pump"

I'm not quite sure how that's supposed to work with a multimeter... do I set the meter to amps and then touch the power and ground points on the pump?

Thanks,

L.

[dingo]

Question about measuring Amps:

In order to verify the ground for the fuel pump, the Digijet manual states I need to:

"Connect ammeter to fuel pump in series with power supply to fuel pump"

I'm not quite sure how that's supposed to work with a multimeter... do I set the meter to amps and then touch the power and ground points on the pump?

Put meter in Amp mode (sometimes requires moving the red lead to a different plug'A').... Remove the Pos lead from the pump and connect it to red lead from meter. Then take black lead from Meter and attach to the terminal of the pump where the POs lead had been. Start the pump or engine or whatever...and read the Amps off the meter

[Amskeptic]

After studying the diagrams in the digijet manual and the Bentley (which are slightly different, I might add) I think there may be a missing ground from the ECU.

The digijet manual tells you to check terminal 7 to the cylinder head for continuity. There is however another ground wire shown in both the Digijet manual and B. It is potentially missing, and I'm wondering if that's a/the problem.

As far as I can tell, everything around the relays is working, except the fuel relay.

When ignition is on, everything shows voltage with two exceptions:

On the ECU relay, the ground wire which completes the opening of the relay shows no voltage. But I don't know if it should or not... it seems like it would, since it's sending 12v from the ignition through the relay to open the switch, and then back to ground.

On the Fuel relay, the dual wire that goes to the fuel injector number 2 and the fuel pump has no power when the ignition is on.

Currently, I have a jumper wire in place connecting the 85 (ignition power in) on the fuel relay to the 87 out (injector/pump).
I check continuity on the wire from the relay to the injector ~ it is connected.

My confusion is this ~ when the ignition is powered with the correct set up, I can verify 12v on the 86 side of the fuel relay. This should indicate that the relay is open.

However, I'm wondering if the the 86 (which goes into pin 13 of the ECU) is grounding out inside the ECU because there is a (potentially) missing ground wire...

In this case, Colin's assessment would be correct.

By the way, the van runs pretty well with the jumper in place. Works for now ~ next week I have vacation, so I'll be cleaning grounds and seeing if I can do some rewiring to fix the problem.

Updates as work occurs....

L.

********************

Question about continuity checking:

It looks like my multimeter has a setting for continuity, but if so, it's not clear how it works. I tested it on a known good relay so I could verify how it works, and continuous or broken showed the same number. Therefore, I've been checking continuity using the Ohms setting. The multimeter shows 1 (I assume this means infinite resistance) when not connected, but then drops to other numbers for resistance settings ~ the new relays, for example showed a resistance value of 79.2 or so between the 85 and 86 poles. The 25 yr old original VW/Audi relays showed a higher value of 82.9 or so. Is using the Ohm setting OK for continuity, or do I need to figure out the correct method for my multimeter (manual is MIA, and Sears doesn't have the correct manual online).

Continuity testers have to use their own battery source to determine the resistance of a circuit. That is why there are warnings about not using continuity testers in live circuits or up your ECU. You must always take a wire off any component you are testing, otherwise you will be measuring the circuit instead of the component itself. This includes "wires" which are component elements of a circuit.

Currently, I have a jumper wire in place connecting the 85 (ignition power in) on the fuel relay to the 87 out (injector/pump)

This is almost insane. If the nomenclature is consistent, 86 is control voltage. 85 is control ground. 30 is hot. 87 is destination.Your jumper according to convention would be a dead short.

Either way, you must understand that relays exist to REDUCE AMP LOADS through CONTROL CIRCUITS. 8"5" is absolutely not supposed to drive the damn pump. You jump 30 to 87 to run the pump. That's it. Or you make the relay control circuit do it for you by energizing 86 and making sure ground 85 is good.

Are we on the same page here? I have not pored over my Bentley but you might make me have to. 85 ignition power in, what page?
Colin

I can verify 12v on the 86 side of the fuel relay. This should indicate that the relay is open

No. You can have 12 volts poised at the threshhold of a relay, but it is worthless without a ground. It does not mean that the relay is "open" when that control circuit has a completed 12 volt circuit, in at 86 and a good ground out from 85. When that circuit is energized, it CLOSES the relay contacts 30 to 87, a closed circuit has power, an open circuit has no power.

If 86 has voltage that you have clearly read comes from the guts of the ECU, then you must see how the ground path works. I personally think that 86 comes from the battery via whatever like ignition switches. ECUs prefer to do their work using the ground side of a circuit. For example, the bay window injectors all get power right off the battery via the double relay/ignition coil. They nonetheless do nothing until the ECU grounds them as it sees fit.

Understand that the ground path is the "safe" path of an electrical circuit.
The horn is an example, a nice tidy short little positive path from the fuse box to the horn coil. Where is the ground path? Crawling around the steering column. The reason for this is the "consumer" (the relay's magenetic coil) has already eaten the electricity, the ground path, if it should ever short out due to a frayed wire, would merely energize the relay (blow the damn horn without your consent). If the positive side (before the consumer) shorted, fires and fuses and smoking wires. That is why they do not send the positive up through the steering column to the button and back down to the horn where it would simply ground at the mounting bracket. NO. Figure this reasoning out.

The ECU does not want electricity that is upstream of the consumer. Any short fries the ECU. Instead it controls the circuit downstream of the consumer where shorts are mostly meaningless annoyance.
Colin

[Lanval]

Well, the relevant pages in the Bentley are 97.61 and 97.62. I've been using the Digijet protraining manual, though.

Here's the relevant diagram from the Digijet (I like it better because it shows the whole setup on one page ~ the Bentley splits it into 2)



Unless I'm crazy (and I'm not) it shows power IN to both the 30 and the 85. 30 is power in from the battery/starter. 85 though, is powered through the ignition coil.

In other words, 30 is always powered. 85 is only powered when the ignition is on.

I'm no expert in wiring/electrical, but the diagram is clear on this point. I don't totally understand what's supposed to be happening in the ECU ~ the chain of events, given the wiring is a bit unclear to me, but based on what I see in the Digijet manual, it seems like this is what's going on:

Ignition power closes (sorry, I said "open" meaning "closed" because in my world, electricity requires an "open" passage... now I know that I must use "closed" to mean continuous/connected.) the relays simultaneously, allowing power from 30 in on the ECU and Fuel relay to flow.

On the ECU relay it appears that the ignition power also does something IN the ECU (86 out on the Fuel relay goes to the ECU connection 20). What, I couldn't say.

Once the relays are closed and the 30s are providing power to both 87s, the ECU relay sends some signal to the ECU (that's 87 out to me), while the Fuel relay sends power from the 30 in to the injectors and the fuel pump.

**************************
I'm *guessing* something like this occurs:

Ignition closes relays
power from 30 simultaneously:
- power the ECU
- powers the fuel pump
- sends "fire" signals to the inejectors

**************************

If I *had* to guess, the ECU is adjusting the fuel flow rate through the injectors (the Digijet manual suggests this someplace ~ I don't remember where, but if you want it, I'll email it to you, or post it somewhere here. I have a PDF copy that I'm working from).

Since the van runs with the 85 jumped to the 87, I NOW think the ECU is OK, and that 1 or more connections are bad. A missing/bad ground perhaps, or one of the connectors in the ECU is bad, maybe.

**************************

I found a used but complete 85 harness for $77 up in NorCal. The harness is on its way now. Classes are finished on Monday, so next week I will:

Clean every ground on the van, and replace any ground wires which seem dicey.

Check/replace/fix alternator as per tests suggest

Replace the wiring harness, and try to verify the 3(!) non-connected wires that I currently have (including that crazy brown one under the A/C mount. I still have no idea what it is. I'm hoping that the new harness will make that clear...)

Finally, I apologize for the confusion created by unfamiliarity with electrical terminology. I'll try to read up on it in the next week as well, so I'm less confusing.

Does this help?

Michael L

[Lanval]

PS ~ the bottom line on pic posting is that this forum doesn't like pic posts from Picasa. I set up an account at Photobucket, and it works OK, so I'll use that from now on, for this forum anywhoo.

ML

[Amskeptic]

As per our phone call this morning and for our readership:

Yes, 85 is the positive side of the control circuit to each relay and comes directly from the coil #15 terminal.

Therefore 86 is the ground side of the control (master) circuit. The fuel pump relay's control ground is handled by the ECU (so it can shut the pump off when the ignition pulses stop due to a stall).

All 30s are hot supply to the destination. All 87s are the relay-blessed permission to operate.

Colin80%Remaining

[Lanval]

OK, Colin yelled at me for awhile, but I'd heard most of it before

His main point was that the system itself is pretty simple, and I should long ago have ferreted out the problem. Sheesh, just learnin' Colin...


So here's what I did before I ran out of light this evening:

Verified the Power Supply Relay (hereafter PSR) is receiving 12 volts at pin 85 from the ignition, and closing. Click is heard, all is good.

Verified the Fuel Pump Relay (hereafter FPR) is receiving 12 volts at pin 85 from the ignition, and not closing. No click, problem.

Verified continuity between:

FPR 86 to the Fuel Control Unit (hereafter FCU) at pin 20. Continuity is good.

PSR 86 to head. Continuity is good.

PSR 87 to FCU pin 13 continuity is good.

FCU pin 7 to ground on head, continuity is good.

FCU pin 25 to ground on head, continuity is good.

FCU to Temp II sensor. Continuity is good.

Temp II sensor ground to head. Continuity is there, but spotty (it was getting dark. I'm going to check again in the morning, but the reading was switching between "I" and a resistance value).

******************

Those last two seem immaterial to the starting issue ~ I just wanted to thorough and check everything I can. There are several points to be made:

1. The fuel pump wasn't running to start with
2. In attempting to troubleshoot the system, I shorted pin 30 on the relay to pin 86 on the relay
3. After shorting the relay, I fixed several wires including grounds.
4. The van runs with FPR 85 jumped to FPR 87

So either:
1. I had a wiring problem which I fixed, but shorted/killed the FCU
2. I had a wiring problem which remains, but shorted/killed the FCU
3. I had a wiring problem which remains, and the FCU is OK

My money is on 1, but there's one more thing I want to check in the morning.

Previously, I verified 12v in at FPR 85 and out at FPR 86. Since I know that wire has continuity, but the relay is NOT opening, I want to make sure that in fact we have 12v going to the pin out on the FCU (pin 20 on the FCU).

If in fact we're getting 12v all the way to the FCU, then there are two options, I would think:

1. The FCU pin and the little connector aren't connecting
2. The FCU is bad

Any problems with my logic here? Any checks/verifications I missed?

*************************

Now for the bad news:

Even with the relay jumped for power, the engine was hiccuping/dying (on occasion). Driving along, the RPM indicator drops straight to zero (no bouncing around ~ straight fall) sits for a second, and picks up again. Sometimes the engine dies, sometimes it doesn't.

If the tach is going out, I'm wondering if the coil is dicey (since the tach draws directly off the coil) or if I need to be looking at several things. In any event, this suggests to me that there is another ghost in the machine, and simply replacing the FCU without attention to the electrical system broadly is likely to result in another dead FCU ~ or some other problem... fire, death, that sort of thing.

Best,

L.

[satchmo]

I'm not gonna be much help, but remind me again if you cleaned all the ground connections. I mean ALL the ground connections.

I'm not sure what you have in the 85, but in my 87 there are a couple ground connections on the left side of the engine bay (one of which is a big strap from the 3/4 head), one at the transmission nose to frame, and a couple 'rosettes' on the body just above the fuse box under the dash where all the ground wires from the dashboard attach. Make them all shiny steel and put a little dielectric grease on each connection. Then there is also a ground connection from the Hall sensor on the distributor to the engine case (does the 85 have that?).

My van had all kinds of electrical weirdness until I got the grounds sorted. The disappearing signal to the tachometer sounds vaguely familiar....

Good, logic based sleuthing on your part so far though.

Tim

[Lanval]

Satchmo,

No, and I'm with you. Tomorrow morning I get to clean and check grounds all over the place. Other wiring too. I've identified at least 3 things that ought to be connected.

But the bottom line is that the damn thing should fire. It's not, and the only thing that I can see is a problem is either the wire to the FCU (which if it's broken, is broken after the first couple of inches out of the FPR) or the FCU itself.

If everything works with the FCU out of the loop then:

1. FCU wiring is bad
2. FCU is bad

I await the word of the master.

L.