911SC only starts with alternator disconnected

[Walt Fricke]

Color me dubious about turning the key from lock to run and having the red relay momentarily direct current to the fuel pump.

First off, the fuel pump pressurizes the system almost instantaneously, so something like prepressurization isn't needed (except when the engine is hot, and the pressure retention system deals with that). When you put a fuel pressure gauge in place (say for CIS diagnostics), the pressure is up to full as soon as you start the pump.

Secondly, my recollection is that the only thing different with the red relay as compared with the other black relays is that it has a diode inside. This supposedly is to prevent voltage spikes. An energized coil, when the ground is lifted, produces a high voltage spike. That is how traditional spark coils worked - points open, there is a spark voltage. This is not so important that you can't use a back solenoid if the red one fails when you are far from help.

While the use of a relay and the safety switch back on the intake and the ignition switch design acts as an ingenious logic circuit, nothing in this system could act to run the fuel pump for any meaningful amount of time when 12 battery volts is not switched to the pump.

Figuring out how this circuity does its stuff is not especially easy. Opening a circuit causes the fuel pump to run? How's that. Easier to understand is that when you switch to start, the switch sends current which causes the relay to send 12V to the fuel pump, as well as sending 12V to the starter solenoid. And when the engine starts, and you release the key so it moves back to run, not only does the starter stop (that's easy to understand), but the relay sends current to the fuel pump, which continues until you turn off the ignition key, or the engine stalls.

I haven't hauled out my circuit diagam to help solve the problems here - I don't have that completely memorized. But it sounds likel you are at least closing in on solving the problem, which is good. By now you know about being a pretzel to deal with that.

[anthony]

Color me dubious about turning the key from lock to run and having the red relay momentarily direct current to the fuel pump.

Would you like to see the video? In my test, I have the AFM plate raised so the fuel pump should run when I turn the key to ON. It runs for a brief moment as the key goes from OFF to ACC.

http://youtu.be/7jNOwEY4SDc

[mytoy]

Originally Posted by Walt Fricke

Color me dubious about turning the key from lock to run and having the red relay momentarily direct current to the fuel pump.

First off, the fuel pump pressurizes the system almost instantaneously, so something like prepressurization isn't needed (except when the engine is hot, and the pressure retention system deals with that). When you put a fuel pressure gauge in place (say for CIS diagnostics), the pressure is up to full as soon as you start the pump.

Secondly, my recollection is that the only thing different with the red relay as compared with the other black relays is that it has a diode inside. This supposedly is to prevent voltage spikes. An energized coil, when the ground is lifted, produces a high voltage spike. That is how traditional spark coils worked - points open, there is a spark voltage. This is not so important that you can't use a back solenoid if the red one fails when you are far from help.

While the use of a relay and the safety switch back on the intake and the ignition switch design acts as an ingenious logic circuit, nothing in this system could act to run the fuel pump for any meaningful amount of time when 12 battery volts is not switched to the pump.

Figuring out how this circuity does its stuff is not especially easy. Opening a circuit causes the fuel pump to run? How's that. Easier to understand is that when you switch to start, the switch sends current which causes the relay to send 12V to the fuel pump, as well as sending 12V to the starter solenoid. And when the engine starts, and you release the key so it moves back to run, not only does the starter stop (that's easy to understand), but the relay sends current to the fuel pump, which continues until you turn off the ignition key, or the engine stalls.

I haven't hauled out my circuit diagam to help solve the problems here - I don't have that completely memorized. But it sounds likel you are at least closing in on solving the problem, which is good. By now you know about being a pretzel to deal with that.

Walt;

You are completely correct. I mistakenly mixed up the Motronic system and the CIS systems operations as I am in the midst of an EFI conversion. Looking at the diagrams for the CIS you are exactly correct and my statement pertains only to the Motronic system and not related to the CIS system as stated in my post. Please disregard my previous post and I apologise for bringing in any confusion to the situation.

[rick-l]

I missed the fact that in your previous test step you had the air flow plate propped up to hold the switch open when you turned the key from off to run.

In this condition the fuel pump should indeed run unless the fuel pump relay is defective or there is another ground path for relay terminal 85 other than through the flow switch. Have you thought of putting your test light on term 85 and verifying it stays on for this test?

Unknowns for me. Does the RPM limiter provide a ground to the relay and where does the current trace for Alarm System go?

[anthony]

I've tested the ground at 85 with the test lamp and it tests good.

I read some other threads about the RPM limiter and didn't see anything that would indicate it as the culprit. I also disconnected it since it was easy to do and the car didn't start.

[rick-l]

I mean test it operationally as you do the test with the plate lifted and the key being turned to on.

Your voltmeter should read 12 volts there with the plate lifted

[anthony]

You mean test pin 85 on the FP relay socket or the rpm rev limiter? I've tested 85 with the lamp. Is it instructive to test it with the voltmeter?

[mysocal911]

Originally Posted by rick-l

Unknowns for me. Does the RPM limiter provide a ground to the relay and where does the current trace for Alarm System go?

Per the wiring diagram, either the rev limiter or the sensor plate switch can ground pin 85,
thus activating the relay and shutting the FP off.

[rick-l]

Yes (Pin 85 of the relay) if you test it operationally. Turn the key on there should be 0 Volts there. Start the car it should go to 12.

20-20 hindsight says that should have been step 2

[anthony]

When I test 85 with the relay connected I do get a small voltage. .1 volts right now. I actually did that test and reported it on page 2 of this thread. I was told though to use a test lamp for a more accurate test. Does .1 volt on 85 mean a short circuit?

From page 2

Test #3
Relay installed
key turned to on position
86 --- 12v
87 --- .03v (should be zero)
87a --- 12v
85 --- .14v
30 --- .02v

I put the test lamp on 85 yesterday and the car just started and the lamp lit up. :-) But it didn't restart after that. :-(

So, backing up. Do I have this test right? I put the multimeter set at volts from the battery to pin 85 with the relay plugged in. 85 is a ground so I need to test from 12v, right?

When I do that I get a reading of .11 volts. (Should be zero I guess.) When I turn the key to start the voltage jumps to about .3 and the car doesn't start.

2nd thing - ignition switch

I'm still confused by the ignition switch. When I do the fuel pump test (lift AFM plate and turn key to on). The fuel pump should run but it doesn't. But I can repeat every time turning the ignition switch a little bit to Accessory and having the fuel pump run for a brief moment. Isn't that wrong for it to do that at the ACC position?

I'm wondering if my ignition switch is just a little flakey though maybe it might not be the ultimate fix for the problem. Some how though it's completing the circuit and making the car do what it's supposed to do.

3rd thing - ignition switch plug

I put my multimeter on the pins of the ignition switch plug. The only reading I get is 12 volts on two of the three holes marked 30. Looking at the wiring diagram, 30 should be direct to the battery and putting out 12v. Have I found something there? Or is there a 30 on the socket that is special or blank?

[anthony]

pic of the plug

[ant7]

HI Anthony,
First of all, your multi-meter, make sure you have it set on DC volts, not AC,
Secondly, as for the ignition switch, if you look back at my input on the first page, i did sugest that it may be an ignition switch issue, i realy hope that it turns out to be just that, as i think and fair play to you, you have had a good go at checking much more than you originaly thought you could!
A...

[rick-l]

Let's break this down a little bit

• Relay installed - key turned to on position
  • 86 --- 12v Top of the relay coil 12 Volts in Run or start.
  • 87 --- .03v Yellow wire from the starter control
  • 87a --- 12v 12 Volts in Run or start = will provide power to the fuel pump when the relay is de-energized (normally closed contact)
  • 85 --- .14v Bottom of the relay coil (switched ground). The higher voltage above your meter offset probably means there is current flowing in the circuit so the relay is likely active and the fuel pump will be powered by the start position of the switch - 87 and 30 are connected together
  • 30 --- .02v Output power to the fuel pump

That is entirely the correct state for the conditions of the test.

Originally Posted by anthony

I put the test lamp on 85 yesterday and the car just started and the lamp lit up. :-)

This is also the correct operation. The flow switch opens and there is now 12 Volts on the bottom of the relay coil, current flows through the test light and it lights. Note: This might be a better application for a multimeter, the test light might have a low enough impedance to affect the relay operation.

Originally Posted by anthony

But it didn't restart after that. :-(

Voltage on 85? Was the light lit? When did it go out?

[anthony]

Voltage on 85? Was the light lite? When did it go out?

Yes, same exact test. First it worked and car started. And then it didn't. Intermittant problems....grrrrr!

HI Anthony,
First of all, your multi-meter, make sure you have it set on DC volts, not AC,
Secondly, as for the ignition switch, if you look back at my input on the first page, i did sugest that it may be an ignition switch issue, i realy hope that it turns out to be just that, as i think and fair play to you, you have had a good go at checking much more than you originaly thought you could!
A...

I have learned a ton here. I hope it's the ignition switch because it feels like I'm starting over though I've ruled out a lot of stuff.

Anyone have thoughts on the voltages at the ignition switch plug? One of the 30 holes has no power. Maybe that is the 30 that goes to the light switch on the electrical diagram? All three 30s are tied together so shouldn't they all have power? Maybe the one that goes to the light switch is not grounded? Maybe I should try and turn on the light switch while testing for voltage.

Pelican also hasn't sent the ignition switch ordered on Saturday. How does an "in-stock" item take a week to ship???

[rick-l]

Originally Posted by anthony

Yes, same exact test. First it worked and car started. And then it didn't. Intermittant problems....grrrrr!

At some point in time the light had to go out. When was that?

The sequence should be

1. Crank the car
2. light goes on
3. car starts - release the key
4. light stays on (or in your case goes out ????)

[anthony]

I can't remember the exact sequence when the car started. And I can't reproduce that. It was all in an instant like it should be.

Now,
crank the car
light does not come on
car doesn't start

It seems like I've done enough testing to figure out this problem with logic. The bottom line is that the fuel pump is not running. I've manually tested the AFM circuit to death. It works. Key on, lift AFM plate, test light goes on as it should. If the the rev limiter or the alarm which are on that circuit were interrupting that process wouldn't it affect the test?

I've tested the FP relay, the fuel pump, the FP relay socket.

The fact that during the "lift AFM plate" test, the fuel pump runs when I first turn the key to the ACC position still intrigues me. I can repeat that every time. That proves the AFM switch circuit does work when the ignition switch is in a certain position (but the wrong position). Can anyone say why that would happen in ACC? Is this a clue leading me astray?

I'm thinking the new ignition switch is going to fix it. Pelican, please ship it!!! Anyone is the SF easy bay with a spare they want to loan me?

[T77911S]

the FP could/can run for a split second when you turn the key to ON. it takes time to energze the relay and the FP can run before it does.
i have not tested this on the 911 but the 930 does run but there are more relays involved in NOT running the FP.

stop screwing the the AFM sw. form what you have done the only way that part of the system can not run the pump is if the wiring is shroted to ground.

one thing youcan do is remove the cover on the FP relay and see if it is energized ot not when the car will not start.

[anthony]

Is it a valid test to put an ohm meter between the pins/contacts on the FP relay to see if they are opening closing?

the FP could/can run for a split second when you turn the key to ON.

This is the key at the Accessory position. The car shouldn't run there right? That is at least the way I read the wiring diagram.

[rick-l]

Originally Posted by anthony

Now,
crank the car
light does not come on
car doesn't start

I understood the symptoms to be (per the video) that the car started when you turned the key but immediately quit when you released the key.

If "car doesn't start" follows the definition above (per the video) and the light does not come on (de-energized relay) when the engine runs with the starter your problem is in the air flow switch.

[rick-l]

When I watch this I see the alternator light, oil light, guage bounce and hear I clunk I think you describe as the fuel pump all at the same time. Is that with the key between acc and run?

Originally Posted by anthony

I made a video. At 4 seconds and 11 seconds in, just as I turn the key the fuel pump runs for a brief moment.
https://www.youtube.com/watch?v=0EnVJgug074

On the other hand it seems like the car should run longer with a pressurized fuel system and maybe run/start power from the ignition goes away.

Of course you could answer this easily by putting your test light on either terminal 86 or 87a of the relay. EDIT:Under the conditions where the car doesn't start.