D-jet + fuel pump troubleshooting, worked fine last month, won't start now! Options

[Chris Pincetich]

Well I was feeling GREAT after all the work I put in to my 1972 914 with 1.7 FI last month: first engine drop, swapped flywheels, sideshift conversion, installed new front shocks, shifter, some little bits. Now I've got it almost all back together and it won't start (IMG:style_emoticons/default/headbang.gif)

What I did that could have changed things:while dropping engine snagged FI plug and had to splice in new spade connector, snagged other FI and had to replace damaged seal, snagged plug cable at top of distributor with CV and replaced plug cables (carefully swapped all wires in same locations), replaced old black coil with Bosch Blue (carefully swapped all 3 wires in same locations), replaced MPS to air intake(plenum?) hose.

First sign of problem was no fuel pump sound when key is on. So, pump tests fine with 12V direct, follow Clay's advice after a search:

"Take out relay #75. Make sure that pins #30 and #85 have 12v.
If both have 12v, jumper pins 30 and 87, and see if the pump runs.
If that works, put the relay back in and ground the front left pin on the 4 pin connector on the relay board (numbered III on the diagram). The pump should run."

All this works!

Pulled my relay board and used McMark's advice and tested Ohms accross all circuits, couldn't find ANY bad spots, and the board is not cracked badly underneath nor does it have tar missing. Tested all relay board relays on headlights = work. Also from McMark, got new 8A and 25A fuses and a spray can of STARTER FLUID. The engine STARTS! Runs for 5 seconds, I can rev it, then dies, cause the starter juice ran out. I think my issue(s) are fuel related, not ignition.

Time for the dreaded D-jet trouble shoot! I am really hoping to avoid this, I mean EVERYTHING worked fine before the side-shift work.

I want to read these sites below, but the links don't exactly work for me (from DD)

If you have D-jet, there are two excellent troubleshooting guides for your fuel pump.
http://www.914fan.net/fuelpump.html -- A text troubleshooting guide
http://members.rennlist.com/pbanders -- Look for the "fuel pump troubleshooting" link; these will be graphical flowcharts you can follow.

Found this link and am reading it
http://mail.symuli.com/vw/Djet.html

Tonight I jumpered the fuel pump relay so the pump ran, then starter fluid and engine starts, but THEN DIES, and pump is still running (IMG:style_emoticons/default/headbang.gif)

Tomorrow I think I'll pull the injectors out, place them in 4 small cups, and make sure they all deliver fuel (advice from p banders rennlist).

I REALLY want to get back on the road ASAP, have missed 3 AX races already when my "quick" sideshift conversion turned into an engine bay rustoration, 1 more this weekend and 2 the next. Help (IMG:style_emoticons/default/beerchug.gif) (IMG:style_emoticons/default/beerchug.gif) (IMG:style_emoticons/default/beerchug.gif) (IMG:style_emoticons/default/beerchug.gif)

[McMark]

It sounds like your brain isn't connected or isn't working. Is this the one you pulled out?

[swl]

I agree with Mark - the ECU is the one that turns on the pump for that couple of seconds on startup. If you are not getting that then you have to suspect the ecu - perhaps power supply. You tried hardwiring the pump and it still didn't work but that would consistant because there would be no firing pulse for the injector.

key on, feather the throttle - any clicks of fuel injector firing?

[swl]

here is a list of multimeter checks I developed when trouble shooting. These were my 'as measured' and not necessarily the 'right' answers. Put this together with Brad Anders work and you should have something you can trust. This involves removing the plug from the ECU and using mainly resistance readings to see what is getting to the ECU. It is checking your harness as well as the devices.

The basics:
Ground: Pin 11 to battery ground strap. 0 ohms on the lowest scale
Supply voltage: pin 16 and 24 - 12V (switched through main power relay) measured to ground.
Start signal: pin 18 12V on start - did not confirm this one, just from reading the diagram. Comes from ignition switch.
Fuel pump on: pin 19 12V - this is the cct that the ECU grounds to run the fuel pump
Injectors:
about 3 ohms to ground on pins 3,4,5 and 6
MPS:
about 100ohms between 7 and 15.
about 30ohms between 10 and 8.
Note: if there is a wire broken to the mps (very high ohms) you will not get any fuel pulse at all. BTDT.
Temp Sensor 1:
about 200ohms between pins 1 and 13
Temp Sensor 2:
about 3.5Kohms pin 23 to ground.
Trigger points:
Alternately closes pins 21-12 and 21-22 as the distributor turns.
Throttle Switch:
closes pins 9 and 20 as throttle is operated.

[pbanders]

Diagnosing the fuel pump circuit can be tricky, that's why I developed the set of flow charts. If you follow through them, you should be able to identify the specific cause of the problem.

[Chris Pincetich]

Mark - same ECU. I took good care of it when the engine was out, and even secured the wires where the enter the box with a screw that was missing on the round clamp. The ECU is definitely plugged in now, but not mounted on the brackets.

Paul - I just noticed that the fuel pump flow charts work in Explorer but they didn't work at home when I used Netscape (the arrows at the bottom were not there).

I was told by Bill E, my local 914 mechanic, that since fuel wasn't dripping out of the lines when I unhooked them, there is air in there, and I need to run the fuel pump several cycles of the 1.5 sec (this was before I found out it won't do the 1.5 sec prime) until the valve BURPS the air out. I have not heard this even with the pump running several 10 second cycles when I jumpered it. Food for thought....

I'll be checking ohms all around (thanks for the check list swl!) and double checking my brain connection tonight (IMG:style_emoticons/default/beerchug.gif) I was a little hung over this weeeknd, maybe my brain was loose (IMG:style_emoticons/default/laugh.gif)

[Chris Pincetich]

CASE CLOSED!! (IMG:style_emoticons/default/thumb3d.gif) (IMG:style_emoticons/default/pray.gif) (IMG:style_emoticons/default/aktion035.gif) (IMG:style_emoticons/default/mueba.gif) (IMG:style_emoticons/default/mueba.gif) (IMG:style_emoticons/default/piratenanner.gif)
Paul's flow charts rock - I owe you!
McMark's ECU diagnosis was correct!

I pulled the cover off the ECU to reveal the white cables all lined up nice. How the hell do I test that??! So then I touch it, it moved! I touched the other side, more movement. I press, it goes in 1/4 inch. I pressed the entire pin set in 1/4 inch. Turn key, pump throttle, engine STARTS! (IMG:style_emoticons/default/beerchug.gif) (IMG:style_emoticons/default/beerchug.gif) (IMG:style_emoticons/default/beerchug.gif)

So, the problem was that the ECU connection UNDER THE COVER wiggled loose, not too surprising after all the twists and turns during the engine drop and driveway work sessions. Time to torque CVs, finish shift rod install, adjust new RenShift and wrap this project up. Then I'll start my "Sideshift Conversion" thread, which I wasn't going to do until it had a happy ending. (IMG:style_emoticons/default/biggrin.gif)

[Dave_Darling]

BTW, the text troubleshooting document can still be found on the "internet way-back machine" --
http://web.archive.org/web/20051108221355/...t/fuelpump.html

--DD