Ladies & Gentlemen,

I need a little assistance figuring out a problem with my 1975, 2.0, 914, Little Red. He is refusing to start and it appears that it is a fuel pump problem. I have tested the relay board and all the circuits are good and appear to function properly and the wiring from the relay board to the fuel pump is good. The problem appears to be a lack of ground on pin 86 of the fuel pump relay. The circuit diagram that I have indicates that this should go to ground through terminal III at the lower left hand corner of the circuit by grounding through pin 19 on the computer/brain.

I have checked continuity between terminal III and pin 19 and it is good. I can make the fuel pump relay kick in by using a test light to ground with the key on. I think that I may have fried the computer/brain when I was trying to solve another electrical problem, but don’t know how to test this component. Is there away to test the computer/brain, and if not where can I get the unit repaired?

Thanks for you help.

Tom Cooper

Remember, D-jet will only run the pump for ~1.5 seconds when the key goes from off to on, or the whole time the starter is cranking, or when the engine is actually turning over. It runs the pump by grounding that pin.

Have you gone through the troubleshooting flowchart on Brad Anders' D-jet web page? http://members.rennlist.com/pbanders

--DD

Dave,

I have power at the fuel pump for only a fraction of a second before the needle on the volt meter drops back to zero when the ignition is turned on. It does not even get to 12 volts before it dies. In the past, I could hear the pump rum for a second or two, that is not happening now.

I have been using an article by Kjell Nelin that is/was on Rennlist. I'll check out Brad's this morning. Thanks for the link.

Tom Cooper

If you're volt meter is dropping to zero when you turn the key on, then it seems to me there would be an issue with getting power from the battery to the board as opposed to from the board to the pump. Try checking continuity of the wires from the battery to the board- you may have a bad ground in there.

As for your computer, I don't know if it applies to older models, but on new model cars, the computer can be "reset" so to speak by unhooking the battery for 12 hours. (Actually I don't remember if it's 8 hrs or 12 hrs, so any time I need to do it I just do it for 12 hrs just to be sure.) So before you go replacing the computer, you might try this and see if it does any good.

Good luck and keep us updated.

Edit: And one more thing, your fuel pump will only run long enough after the key is turned on to prime the fuel lines for ignition. Once they're primed and no air is in them, the pump will shut off until you start cranking. So consider only hearing it run a second or two normal.

Just for clarity - when you ground pin 3 and the relay kicks in - does the pump run too?

Where you in the area of the ECU when you were troubleshooting? One of the guys a while back tracked a similar problem to the harness not being all the way home on the ECU.

Maybe I wasn't clear enough on that. When I turn the ignition key on, the volt meter jumps toward 12 volts for just a fraction of a second before falling to zero, it's kind of a blip. Previously, I could hear the pump run for a couple of seconds before it would stop. That's not happening now.

I have considered that I might have a problem with power to the computer/brain. However, when I check the power to pin 24 on the computer/brain. I get 12 volts. I also have continuity on the wire from terminal III on the relay board to the computer/brain.

I think that I may have fried the computer trying to fix another electrical problem, but want to make sure that it is getting power. Are there other pins on the computer/brain that should be getting power?

I did take the computer/brain out of the car overnight, so that should have reset the computer/brain if that works.

Thanks for the help.

Tom

That's a good question. I need to get some one up front to listen if the pump is running when I ground terminal III on the relay board. This old guy doesn't hear that well anymore. FYI, the doctor said my hearing was normal in my good ear for a person my age.

I didn't move the ECU when I was trouble shooting, but I may have given it an electrical shock. I have now unplugged it and checked the the voltage and continuity to ground and reinstalled it so that should be okay.

Thanks for the help.

Tom

Disconnect the ECU and ground Pin III on the back connector by hand and see if that makes the pump run ...

Andy

Like this, except, don't make it permanent like in the picture (that is a reference for carb conversions) ...

This should get the pump going as long as you keep pin 3 grounded.
If so, something is wrong with the brain. If not, the problem is in the board, relay, wiring or pump ...

Andy

"Reset"? This thing is as close to crude as you can get. Removing the power from it does absolutely nothing. While an arc might hurt it, that's unlikely, too. Compared to rocket science, the technology and the component sensitivity is like throwing rocks. The Cap'n

Okay, maybe I need some clarity now.

When in reference to cars, somebody says "computer" "brain" or "control unit," my thoughts go directly toward a box with a circuit board in it that distributes electrical information to sensors or any type of electrical hook up that needs controlling.

So are we talking about something of that sort, or are we talking about the relay board, as SirAndy pictured in his post? Because last I was aware, they weren't the same thing...

Yes, "computer" "brain" or "control unit" and not relay board.

However, this was Bosch's first electronic fuel injection and the "ECU" does not look like anything somebody younger than 40 would describe as a "computer".

There's really nothing to "reset" in the ECU. It's a bunch of diodes and capacitors and resistors arranged in a somewhat clever fashion.
Andy

Thanks for all the comments and help. I'll try Sir Andy's approach tomorrow and see if that solves the problem so that I can determine if the ECU, computer/brain is really the problem.

Tom

And a few transistors thrown in to turn a few things on and off too....

Here are the circuit schematics for those interested:

ECU page 1
ECU page 2
ECU page 3

--- bill

If the voltmeter on the fuel pump power wire jumps when the key goes from off to on, the fuel pump control circuitry is working. The most likely reason it doesn't get to 12V at that time is because the meter is lagging.

How about when you crank the starter? Does the pump control wire go to ~10V during that?

How's the ground for the pump? Does the pump run when you jumper the pump relay? Have you tried swapping the relays on the relay board with those for the headlights? Or any other spares you have on hand?

--DD

Meh... I guess that's just one of the curses of being born after 1900...

That doesn't sound good Tom. The ECU controls the duration of that key on running of the pump. There's no intelligence in that cct - just a simple timer. If the pump is turning on for a blip it suggests the ecu is there and trying to do it's job. If the blip is really short then it has to be the ecu. Maybe it could be a bad contact reducing the power available to that cct. It has been a lifetime since I've looked at flip flop ccts but my guess is that at a lower voltage the discharge would happen more quickly.
You have 12volts on pin 24 of the harness. Do you also have 12v on pin 16? Do you have a solid ground on pin 11 of the ecu harness?

Ladies and Gentlemen,

Okay, I tried Sir Andy's trick of running a wire directly from ground to Terminal III on the relay board, the fuel pump relay kicked in, but the fuel pump didn't run. So, I disconnected the hot lead and the ground from the fuel pump and connected them to my volt meter and tried again. That showed twelve volts at the fuel pump. Bad fuel pump right?

Just to make sure, I reconnected the leads to the fuel pump and tried again. This time I got a strange noise that I hadn't heard before, and checking the fuel pump it was running and dripping a small amount of fuel. Then I reconnected the plug to Terminals I through IV on the relay board and tried to start Little Red. He started right up and run like nothing had ever happened.

My only conclusion is that I had fuel dripping on the leads to the fuel pump and that caused a discontinuity in the circuit and after unplugging and replugging the leads several times to test I finally made contact. The only other possibility is that I somehow disturbed the connection when I remounted the fuel pump, but having pulled and reconnected them several times I don't think that was/is the problem.

Now my question is now do you get at the fuel pump in the front trunk? The opening isn't very big, and I don't see a way other than pulling the fuel tank to get in there and fix the leak. I am sure that some of you have been there done that and know a few tricks that I don't.

Thanks for all your help, and if anyone needs the diagram and explanation of how the ECU works I now have a hard copy in my library.

Tom

Are you talking about replacing the pump, or just replacing the line that's leaking? Because that's the difference between pulling the tank and not pulling the tank. And you say this...



So that tells me that you have enough access to replace the pump if you need to. But if you're need to replace the lines going to and from the pump, you'll need to pull the tank. Unless something's escaping me, there's just no other way.

Well, I guess that's not entirely true. If you get underneath the front end, you should see an access plate you can pull off that's removed for things like replacing the master cylinder. When you remove that, there should be an access hole you can see that will allow you to disconnect and plug fuel lines coming off the tank. I suppose if you were motivated enough, you could feed new lines using that access hole and the access hole for the feul pump. But that sounds like a much bigger hassle than just pulling the tank and having the ease of room to work with. See below (this is a pic of the front end of my '75 2.0)-

Click to view attachment

If you have a helper pulling the tank is not difficult. I've been able to do it with the hood still on by removing the expansion tank first. Trick is to get the fuel out first.

If it has been a while since the lines have been replace getting them off the nipples can be challenging. Your best bet is to commit to replacing the hoses and just cut them through the port in DB's photo above. Then the tank will come out easily and you can work on getting off the old hoses with a bit of working space.

Something that I did when I first pulled the tank on my teener (because of the 3/4 tank of 10 year old gas in it), was just took a pair of vice grips and clamped both lines through the access hole, then just cut right below the grips. It's not 100% leakproof, but it keeps the gas from dumping so fast it just makes a smelly mess of things.