Looked to be a fuel pump problem but likely a fuse panel problem

[nkemp]

Here is some info in case others have similar problems. Maybe I should call this "When 12Vdc isn't 12Vdc). Just because there is 12Vdc on the meter does not mean there is sufficient current to do the job.



The car periodically would "buck"...usually with a bit of a bang. This may or may not be related
It would start fine but on some days it would die. I could often get it started again. Over the years it has had 2 flatbed rides
Some times it would simply start later on and that makes it VERY difficult to diagnose
Seems that it was dying more lately
Multiple times I tighten a ground and it starts
Installed a ground bus and I'm very pleased that the volt gauge has some spark in it (pun intended) Thanks Bill R for the motivation and idea!
The other night the wife had to pull it home (that went well :pitty:)
I jumpered the RPM relay and no pump action
Measuring 12Vdc at the pump connection
At this point I'm thinking bad pump but it has less than 5K miles on it
Took the pump out (my least favorite job) and connected it to a battery and it runs. Hmmm
Connected to car harness, jumpered the RPM but no action even though 12Vdc at connector
Jumpered the ground to the engine block just for good measure but no pump action
Checked again and the pump runs when connected directly to the battery.

At this point the passenger car seat is out and I'm digging into the fuse block. #7 fuse block location looks like overheating (probably due to previous bad ground) Will update here as it proceeds.

[Bitsyncmaster]

Don't measure voltage at the pump to a ground. Measure from one terminal to the other. That way you know if your ground (inertia switch is grounded). If the inertia switch is open you will measure 12 volts on both pump terminals referenced to ground.

[nkemp]

I measured it a number of ways. When I measured across the pump terminal (no pump attached) I saw 12Vdc+/-. When I measured across the pump terminals when it was plugged into the harness I saw no voltage which confused me but it may have been a clue that I missed.

As I recall, the old style meters (with a needle) were less likely to display phantom voltage. The digital meters will display the phantom voltage. As I also recall, even the slightest load across the digital probes will disperse the phantom voltage which is what may have happened when I measured across the pump. BSMaster (maybe that is not a good way to abbreviate Bitsyncmaster ;)) do you have a more knowledgeable opinion?

Nick

[nkemp]

I got the fuse block to a more workable angle and the terminals are toast. There is a lot of discoloration on the extend a fuse spades. This looks to be a prime suspect.

The power routes from the fuse to the Lamda relay where it uses that point as a terminal before going to the RPM relay. It is not very long (12"?) but the wire from the fuse to the relay is a lighter gauge than the rest of the bundle. I was planning on extending the brown wires a bit to make it more workable and will replace the light gauge wire with something heavier. No use adding a bit more resistance albeit small.

I'd prefer to replace the whole fuse panel after seeing the panel sockets and the damage they can do but that isn't going to be happening for a while. I'll run the extend a fuse spades through the panel and make the connections behind the fuse block. That way it is not substantial different from original yet it will have better connections (bigger & soldered).

The questions are:


Why do those terminals get so hot?

It is a given that the female connections are light duty and poor

Is it due to a poor ground in the circuit?
Is it due to that ground going back to the bulkhead bolt being sized wrong for the load?

That circuit carries the fuel pump, Warm Up Regulator (resistance heater) AKA CPR, freq valve and other circuits.


Time to warm up the soldering iron and make some improvements.

Nick

[LEVY]

Bitsyncmaster gave you the best advice, and for a good reason.

If you measure the voltage at the pump, from + to - you are actually checking both + and - at the pump and you will never know where the problem is. If you check + at the pump and - some where else, you are now checking both, + & -.

Now, do this:


Disconecct your battery!

Disconnect your battery!!

Connect one wire from the negative side of the pump to negative side of the battery (straight to the battery).

Connect your battery.

Bypass the pump relay and check if your pump is now working.

If your pump is now working, now you know you have a bad ground! (Bitsyncmaster gave you a clue of what your problem is).

If your pump still not working:


Disconnect the battery.
Disconnect the battery!

Leave the ground wire straight to the battery but now add another wire from the battery straight to to the + side of the battery.

Your pump should be now working, now, just to double check, disconnect the negative wire from the pump just to make sure you only have problems on one side of the circuit. If the pump still running, then you know for sure you have problems on your + side.

Explanation:

You are checking for the "phantom" voltage you described on your post, if you have bad connection on one of the wires + or - , you might ready voltage at the pump when the pump is not connected, as soon as you connect the pump, the circuit will be interrumped due to the bad connection and the pump is trying to put a load on the circuit, you will read 0 v. You know you have a connection problem but you don't know on which side.

Now, this procedure should only be used if you have enough knowledge, wires at the pump should always be tight before restoring power, you don't want any spark close to the fuel tank!

Remember, always connect at the pump first, then at the battery to avoid any sparks close to the fuel tank!

Remember, always connect at the pump first, then at the battery to avoid any sparks close to the fuel tank!

Remember, always connect at the pump first, then at the battery to avoid any sparks close to the fuel tank!

Note:

If you have a discharged or a bad battery, it will release hydrogen and it may explode if you have a spark close enough, so be very careful!




Don't measure voltage at the pump to a ground. Measure from one terminal to the other. That way you know if your ground (inertia switch is grounded). If the inertia switch is open you will measure 12 volts on both pump terminals referenced to ground.


I measured it a number of ways. When I measured across the pump terminal (no pump attached) I saw 12Vdc+/-. When I measured across the pump terminals when it was plugged into the harness I saw no voltage which confused me but it may have been a clue that I missed.

As I recall, the old style meters (with a needle) were less likely to display phantom voltage. The digital meters will display the phantom voltage. As I also recall, even the slightest load across the digital probes will disperse the phantom voltage which is what may have happened when I measured across the pump. BSMaster (maybe that is not a good way to abbreviate Bitsyncmaster ;)) do you have a more knowledgeable opinion?

Nick

[Bitsyncmaster]

When I measured across the pump terminals when it was plugged into the harness I saw no voltage which confused me but it may have been a clue that I missed.
Nick

That tells you the ground is open since you saw 12 volts reading on one terminal.

Your not drawing any current so the voltage would be the same on both terminals.

[Bitsyncmaster]

Measuring the voltage over the pump terminals also will point out a poor connection if the voltage reads to low. On my car I found two bad connections even thought the car runs fine. It's just a good preventive measure to check that voltage.

Then I did measurements of each point using a long ground wire from my meter back to the battery ground post. That way you can record your readings and determine where the voltage drops occur.

1) One terminal of the two pin connector up near the pump was dropping 0.50 volts. I fixed that by removing that unneeded connector.

2) One terminal of the 20 amp fuse for the pump circuit was also dropping 0.50 volts. I fixed that by installing a new DMCH fuse block.

3) The ground wire from the inertia switch was dropping 0.40 volts. I fixed that by installing a new ground wire going to the frame.

[David T]

There is a connector up inside the washer bottle area that gets corroded. A good place to start is right at the terminals on the pump but if you got nothing it doesn't tell you if you are missing power or ground. Next step would be to see if there is power by going to a known, good ground. Also make sure the inertia switch is on the side of the footwell (not on the Lambda switch) and it has a splotch of white paint on it. Just doing a visual can get you close. Some of the more common failure points are:
#7 fuse and connectors in the fuseblock
Bad inertia switch
Wiring mods by PO's
Bad RPM relay
Corroded connections by the pump or washer bottle
If all else fails you could have a bad pump but rule out all other possibilities first.
David Teitelbaum

[nkemp]

The point I am trying to make in this thread is that even though your digital meter reads 12Vdc ( or something similar) does not mean that you have sufficient current flow to operate the device. I have seen this on the D and on other electronic devices I was involved in diagnosing.

When I was diagnosing my perceived fuel pump problem, the digital meter read 12Vdc at the pump but the pump did not work with the ground as wired or with it jumperd to known good ground. At firsty glance one would assume that the pump was bad. But when I took the pump out and wired directly to a battery it worked. I ruled out the ground as an issue when it did not work with the pump grounded directly.

I started diagnosing the positive side of the circuit and it turned out that fuse 7 was badly damaged but not sufficient to read zero volts at the pump. I'm not sure if the term is correct but I refer to this as "phantom voltage" since it is not real in terms of an active circuit or in terms of sufficient current flow to operate the circuit.

I have since replaced the fuse 7 terminals and the pump now operates (I just used it to pump down the tank).

I now plan on locating the connection in the washer bottle area to validate that connection. I like the idea of working back and determining the voltage drop at the various points/junctions.

Over the years, these connections have corroded and cause problems. For instance ... I lost the whole electrical system (not a single light or item worked) at a corroded bolt at the ground on the trailing arm bracket. And the bolt was TIGHT.

Nick

[Bitsyncmaster]

Your pump probably stopped turning at a point where the bushes were not making contact. Hence you read the full voltage because no current is being drawn. When you removed the pump it must have moved the brushes a little and then would run.

You would have seen the same results with an old VOM as you saw on the DVM.

A bad connection can read full voltage with no load causing no voltage drop over the bad connection.

[Ron]

When I measured across the pump terminals when it was plugged into the harness I saw no voltage which confused me but it may have been a clue that I missed.


As I recall, the old style meters (with a needle) were less likely to display phantom voltage. The digital meters will display the phantom voltage. As I also recall, even the slightest load across the digital probes will disperse the phantom voltage which is what may have happened when I measured across the pump.

Hey Nick,
I believe I can help clear this up for ya.
First, you're not accounting for impedance (a unit of measurement for the opposition to AC...Basically the same as resistance with DC, but accounts for capacitance and inductance effects on AC which DC does not have -- With DC, Impedance = Resistance.)
Second, phantom voltage is with AC only (except for jargon when intensionally mixing the two, but we don't need to go there). It is caused by capacitive and/or inductive coupling from nearby AC circuitry.
When it comes to if a given meter will read it, it's not whether the meter is digital v/s analog, it's high v/s low impedance of the meter. Obviously, a low impedance meter will be a heavier load than one with high impedance. A high impedance meter's load is small enough that is can read the phantom voltage without "dispersing it".

Now, what appeared to be a phantom voltage on the pump circuit-
The pump circuit is DC, so the meters "impedance" (think opposition to AC here) means nothing*, so the voltage present was simply read -- But the bad connection would not handle the relatively huge load of the pump.
(This is why I like an ice pick light to (solely) test for hot or ground. A bulb is usually a reasonable load....and serves as a fair fuse if I screw up ;-)

* This is also why a different meter would read the same (as Bitsy said).

Hope this helps...


Your pump probably stopped turning at a point where the bushes were not making contact. Hence you read the full voltage because no current is being drawn. When you removed the pump it must have moved the brushes a little and then would run. It seems unclear (to me) as to whether he he read across the pump with it installed and got 12v. If so, I see what you are saying...and he will need a pump soon because the brushes should always be contacting the commutator. (Maybe that was your point? -- But it seems that the problem wasn't the pump (alone anyways). Time will tell....)

[Bitsyncmaster]

It seems unclear (to me) as to whether he he read across the pump with it installed and got 12v. If so, I see what you are saying...and he will need a pump soon because the brushes should always be contacting the commutator. (Maybe that was your point? -- But it seems that the problem wasn't the pump (alone anyways). Time will tell....)

I was thinking the same....I would not trust that pump.

Back in the old days a VOM (volt ohm meter) was just a micro-amp meter with resistors to change the scale. Usually the meter was rated at 50 Kohms/volt. Which just means on a 1 volt scale the resistance load was 50 Kohms. On a 10 volt scale the load was 500 Kohms....etc. Todays DVM are mostly 10 Mohms or higher but even with 500 Kohms load you will most likely not see any voltage drops due to bad connections.

[Ron]

I'm with ya.

"Old days...VOM"!!!
I still have a nice, working, VTVM....somewhere.

[Bitsyncmaster]

I'm with ya.

"Old days...VOM"!!!
I still have a nice, working, VTVM....somewhere.

How old are you? Even I forgot about VTVM....the vacuum tube volt meter. That one did have a high impedance. I did have one as a kid.

[Ron]

lol...probably older than you.

[nkemp]

The car has a new fuse 7 circuit and is back together. Runs and runs much better. I think the fuel pump is quieter (Since day one it has sounded like a meat grinder). Now I'll have to try harder to hear if it is running. Also, I think the WUR works better.

I never say that the D is fixed... but I think this problem may be on the mend.

Nick

[dn010]

I just went through all this crap too - it was the same symptoms and everything. After cleaning, replacing, grounding (bus), etc; the only thing my situation can be is a bad pump. I took the pump out and ran it on a battery and it ran, put it in and didn't run (caused the wires to get extremely hot as well as the fuse) leading me to believe it was vehicle electrics. I fixed/upgraded the above mentioned things and the pump sprang to life and was almost silent. I was extremely happy to fix it and it ran like nicely.... for about 15-20 minutes, and then it gave up again. Jumped pump with fresh battery directly and it doesn't run, just burns up the wires. I haven't gotten a new pump, but I have gently tapped on the broken one with a rubber mallet which has freed it up and allowed it to run - there is either something internally wrong or somehow (even with a nice clean tank and clean pickup filter) something got sucked into it lodging it. New pump and I believe my troubles are over.

I hope yours doesn't seize again and it is fixed.

[nkemp]

I've been running mine the last days and it has been running great. While debugging mine:


I tried running in the car, no luck
Put it on a battery and it ran fine
Put it back in the car, no luck
Back on battery and ran fine

At that point I narrowed it to the car electrics. When I dug into the #7 fuse I knew it needed replacing. Wasn't sure at that point if it would solve the problem. When done with the electrics it fired right up.

As before, I will not declare it fixed and jinx the repair. But I think mine is solved.

Good luck with yours.
Nick