Found that my fuel pump wasnt getting power. Turned out to be a bad relay. However, now the pump runs but continues to run when the engine isn’t on. It used to shut off after 1-2 seconds when the key was turned to the on position with the engine off.

Tried another relay but no change. Are the round relays all the same? Could they be stuck ‘open’?

Also I have 6 relays that are apparently bad. Are they repairable?

Fuel pump relay also runs the the blower fan in the engine compartment. Make sure the switch lever on the floor with the red handle is not shorting out and the fan connections are good. Quick check would be unplug fan and see if fuel pump still runs after shutdown.

Quick check for relays is to plug them into head lights and test if they go up and down. Check for corrosion on the pins.

I disagree with the above post (unless I misunderstand)...in the 2L there are two separate and distinct relays for the fuel pump and blower fan. Those relays use a common power supply, but each circuit is controlled separately: the blower motor by the switch in the console and the fuel pump by the ECU.

All the round black relays are the same, they're interchangeable. It is unusual for one to fail closed, they usually fail open. I am aware (and I have) one silver relay which is a jumper wire that keeps the controlled circuit on all the time, but it is silver...

Electrical diagrams are here. Use the '74, I understand it's the same: https://www.pelicanparts.com/914/914_electrical_diagrams.htm

I suggest using the blower circuit to test your relays. Here's a link to the board; you'll notice the blower is the bottom center one. Pull the fuel pump relay (just above the blower relay, just so it's not running all the time), key on, stick the relays into the blower circuit one at a time. Reach into the car and flip the blower lever up and down and make sure they each work. If one does not work, take a razor blade and gently SLIGHTLY spread out the quarter/slotted male tips to ensure they get connectivity and try again. Mark the ones that test good. Use dielectric grease on all the terminals.

https://www.pelicanparts.com/914/parts/Elec...lectric_73E.jpg

If the fuel pump is still powered when the key is off, and you're using a known-good relay, then the problem is in the control circuit. With the key on and the engine not running, pins 85 and 30 should have 12V, pin 87 goes to the fuel pump, and pin 86 goes to the ECU. Disconnect the wiring harness from the ECU; if you have a dead ground at pin 86 then you have a shorted wire you're going to have to trace down.

Troubleshoot and report back!

Edit: just realized the relay board image is unavailable. I've attached it here.

Edit 2: Quick tip: I find it a real PITA to quickly know how the round relay fits onto the board (why make them round??) They'll only fit one way but it's still a pain to quickly find that, especially if light is low. Note on the board there is one pin location that does not have a terminal in it? That's pin 87a on the relay. So I take a silver Sharpie and mark a vertical line on the relay alongside 87a, and then a dot on the board for that empty pin and it makes an easy reference on how they go in.

Click to view attachment

Sorry about my error about fan relay and fuel pump relay I had them confused.

Thanks for the detailed reply. Will test the relays as suggested.


The fuel pump doesn’t run when the key is in the off position, it just doesn’t shut off in the on position like it’s supposed to after 2 seconds if the engine isn’t running. This is what I’m trying to solve. It was working correctly before the relay died.

I will try to explain how to test a relay.
relays are switches that open and close.
In a classic four prong relay, you have two large wires and two small ones.

So, the two small ones are the trigger circuit.
One side is ground the other is power.
When you connect those two properly, the other two prongs now form a circuit, or the switch closes.

So, to bench test a relay, you need a 12vdc power source, and an ohm meter.
You will test the two large prongs to see if they have continuity no power connected to anything.
If yes, bad relay, toss it.
If no, proceed.

Now, hook up your 12vdc power supply ground to one of the smaller prongs.
while checking continuity of the large prongs, touch the 12vdc positive lead to the other small prong.
The result should be the two large prongs now have continuity.
the switch closed.
Remove the positive lead, the switch should be open again, with no continuity.

Large prong, small prong does not apply to our relays,
some one else may be able to post the pinout function of our relays so you can test them.

Alternately, if your lights go up and down when you turn them on, those two relays in the front trunk are working.
I use that as a simpler test, because its easy.

Rich

Thank you, Rich!

Roger. Then start with testing with a known-good relay. If the fuel pump still does not turn off with key on after 2 secs but before starting the engine, then you're looking at either a short in the wire between pin 86 and the ECU, or an ECU fault (the ECU grounds that circuit to energize the relay.)

Someone needs to post up the fuel pump diagnostic Flow Chart for him I’m sitting in the airport or I’d do it.

Wish I knew about it, or I would...!

Actually one other easy-to-do test: does the fuel pump stay on when you remove the relay entirely, then go back on when you re-insert a known-good one?

Update:
After verifying which relays were good via the headlight relay swap test, I verified continuity on the relay board traces and went through this excellent troubleshooting guide:
Bowlsby’s most excellent fuel pump troubleshooting guide

Passed all the steps (didn’t bother testing the ECU) so I cleaned all of the board and relay contacts. Reinstalled the board and the pump is working again as it should. Should’ve just cleaned it all up before going through all of the other tests, but doing so gave me a good understanding of how it all works.


Thanks to everyone that chimed in!

It was listed in "Brad Anders '70 914 2.0L" guide posted at Rennlist however that file is now empty! IIRC, I printed it and will post it when I locate it. Unless another member beats me to it.

Did you check the link in my post above? It links to Bowlsby’s troubleshooting guide and is the most comprehensive list of steps I’ve seen ever posted on the subject.

I've got a hard copy of the flow chart to which I think you refer, but it's not in great shape. I agree, it used to be readily available and I don't know where it went. I've recreated it, hopefully faithfully, in PowerPoint (yes, I know the EE's see that and are like , but sometimes PowerPoint is the easiest to use for simple stuff). I can also PDF it if it will be useful for rjames or anyone else.

If you have the url for where it used to be, you might try plugging it into the wayback machine to see if it pops up. It can be handy for recovering lost web content but isn't 100%.

Couldn't find it on the net but here it is after I copied it out of my garage notebook. Sorry for the non-professional appearance. In most cases you'll figure it out in 2 pages but here are all six. Kent

Click to view attachment

Click to view attachment

Click to view attachment

Click to view attachment

Click to view attachment

Click to view attachment