No-start gets more interesting

[JPaul]

Nevermind more infuriating...

So, several days of no-start diagnosis and I seem to be gaining on it but no breakthroughs.

Checked the DME relay according to Bentley and everything checked out fine. Even used a backup relay to check (thanks John Walker!!!), no difference. Could feel the relay making both connections, one during power on, one during cranking.

Got a loaner brain just to throw in to see if it'd fire up, nothing.

Took the next step and crawled under the car to test fuel pump during cranking, no vibration. Grabbed multimeter to check for fuel pump voltage while cranking, nada. Huh?

Got back under seat and pulled DME relay, jumped 30 to 87 (as they suggest in Bentleys) without the DME relay in place. Crawled back under car to look for voltage at fuel pump, NOTHING!!!

Okay, at this point I'm thinking the security system is faulty. As far as I know, it's a straight run from 86 to the fuel pump (haven't looked at schematic yet). Next phase is to chase the wiring for continuity.

Does this ring any bells for anyone?? I'm getting to the point of frustration, although on the bright side I wasn't getting rained on today...

Appreciate any support!

JPaul

[my1st911]

I would expect the fuel pump to be fused. I might be able to look at a wiring diagram if you give me a year

[JPaul]

Sorry, you're right. The fuse panel fuse (#6) is good, without corrosion. Happy to give you a year to check schematic. Right now it feels like it could take that long to fix...

JPaul

[my1st911]

i checked your garage. Its an 87 with alarm. I have to boot up other computer

[my1st911]

I believe that when fuel pump is energized, it goes to fuse number 3 in the fuse box number 1. Also, if you take out dme relay, and crank the car, pin 86 should energize. That is coming from the alarm unit. If pin 86 doesn't power up, your alarm unit is keeping the injectors from getting power. I could be wrong, but the alarm keeps the injectors from getting power. That"s what I see from my wiring anyway. ck fuse 3 first, and then check your injectors for noid, or at the very least 12 volts. Others will chime in also i am sure

[JPaul]

Sorry, caveat is that the alarm unit is aftermarket. I don't know what they might have wired into...

I'm going to check continuity with fuse #3 and the fuel pump. Can't imagine what may be going on but this seems to be a process of elimination. I've given up on spirited epiphanies.

JPaul

[my1st911]

I'm not sure what they wired into either, but their usually is a bypass button

[ischmitz]

Could be a bad fuel pump or the aftermarket alarm was wired into the fuel pump circuit. Porsche alarm units prevent +12V from getting to the DME relay but that doesn't seem to be your problem.

It's really simple. Jumper at the fuse box to see if the pump comes on. +12V to fuse #6. If not ohm out the cable from fuse #6 to the pump and make sure the other side of the pump is properly grounded.

Finally, run straight +12V to the pump from the starter when you are under the car to make sure the pump can turn. That should narrow it down.

Ingo

[JPaul]

Thanks Ingo,

I have the fuel pump in hand just in case but today I think I eliminated it.

Even when jumping the wire harness for the DME relay, I got not voltage at the pump. I assume if the pump were compromised (shorted) and 12V were being delivered, it'd blow the #3 fuse (or #6 depending on how you count).

I did test the ground on the pump and found it to be good. I have to assume something is preventing voltage from getting to the pump and the only thing I can think of (or than a bad wire/connection) is the alarm system.

It's done this to me on a handful of ocassions previously and I've never been able to determine the cause. Although 'til now, it's self corrected within an hour or two. I'm now thinking there's something in the alarm system that has been going slowly wrong and it finally gave up the ghost.

Half the reason for choosing this car was the simplicity of its systems. Not like working on a mid-sixties VW, to be sure.

Appreciate your thoughts.

JPaul

[ischmitz]

The fuel pump is nothing but an electric motor. It gets GND on one side and +12V on the other. No reason to believe you would blow a fuse. The fuse is in line with the +12V in order to protect the wiring harnes when there is a short in the pump.

If you take a simple wire (you could use one with an in-line fuse if it makes you feel better) and connect one side to the hot side of the starter and the other side to the fuel pump it needs to spin. If not it is bad.

Then you need to put +12V straight to the fuse panel fuse #3 to the RED/GRN wire (it is the wire coming from the DME relay). This send the +12V through fuse #3 into the red wire going to the fuel pump unless the alarm is spliced into that red wire. This is perfectly save.

If you don't feel comfortable hooking up +12V use a DVM instead and measure at fuse #3 when cranking. You need to see +12V. If yes the fault (alarm splice) is in the wire from the fuse to the pump or the pump is bad. If not the fault (splice) is in the wire from the DME relay Term. 87B to the fuse.

Ingo

[JPaul]

Thanks Ingo,

This is where I'll focus first thing tomorrow. No fear in running wire!!!

Really looking forward to hearing that engine purr again. I've been missing it. That'll be the signal that I can start reassembling everything I've ripped apart in the past several days.

These little adventures are great teachers but I have to admit I'm learning a bit more than I hoped for on this one...

JPaul

[JPaul]

Okay, ran 12V from battery to fuel pump. Nice whirrrr!!!

Traced wiring to security system and found that it interrupts starter motor only, which is working fine.

Connected long wire to coil wire going to distributor and stretched around car so that I could hit the ignition key while holding the wire next to the door strike (thanks again, John Walker!). No spark.

I'm starting to think there's something screwed up with the ignition system, maybe with the ignition switch itself?

Sound viable?

Thanks in advance!!!

JPaul

[JPaul]

No joy. Can't imagine what's happening.

Checked ignition connections, all fine. Checked spark from coil. Nothing. 12V to coil and resistance of coil checks out but coil isn't firing.

I'm starting to think the ECM is bad or is being given information that's leading it to not energize coil and fuel pump. Funny thing is that the same is happening with a loaner ECM that's known to be good.

I'm flummoxed.

[JPaul]

Will the speed sensor actually prevent the car from running? I thought it may make running above a certain RPM problematic but if it will prevent starting, I'm headed there next.

[mholbrook38]

Sounds a lot like crank sensor.

[E Sully]

DME needs signals from the speed and reference sensors. Initialy, the fuel pump runs while the key is in the start position. If it doesn't sense rotation it will not keep the fuel pump running or signal the spark.

[JPaul]

Agree with you both, Sully and Holbrook.

Unfortunately, cannot get to the sensor area very easily. It's a ***** on your back and going through the wheel well is blind with no leverage. Don't know how others have worked on this unless they've got access to a lift.

Sully, I was testing fuel pump with ignition cranking yesterday, no volts. Weird.

Today, tested fuel pump with wire jumped from battery. Spun as expected.

If the speed sensor is what's faulty, I have to assume the ECM will not tell the fuel pump to pump or the coil to fire unless it's seeing signal from the speed sensor.

That's my current theory...

Thanks!

JPaul

[E Sully]

You can do a quick check of the speed and reference sensors at the connectors under the blower duct.
My issue was unfortunately the DME, but the rebuilt unit is now working fine.
It is difficult to get to the sensors with the engine in, but there are a few threads on it. The gap from the flywheel sensor is very important.

[JPaul]

I've been living this chart in the Bentley manual for the past several days. I just saw the section on testing resistance between the sensors. Will head out to do that now. Just don't know what I'll do about it. Looks like a job for John Walker's Workshop. I can take abuse, but when engine sheet metal starts shaving my arms of skin, I get a little testy...

Thanks Ed!

[paulgtr]

I changed those sensors. you have to get them from the passenger side with the rear wheel removed. definitely a contortionist job. a little worse than CHT removal, but doable if you take your time