• Welcome to the new NAXJA Forum! If your password does not work, please use "Forgot your password?" link on the log-in page. Please feel free to reach out to [email protected] if we can provide any assistance.

Power to fuel pump questions


NAXJA Forum User
Waynesboro, VA
I know that when the ignition is in the start position the fuel pump gets the total voltage from the battery. Give or take 12 VDC.

When the key is in the on position, it gets about 8 VDC. Measured at the 3-way fuel pump plug.

Well what I am getting is more like .8 VDC in the on position an 1.2 VDC in the start position. I checked my meter and made sure that I was using it right and it read about 12 VDC at the battery.

So what in the fuel pump curcuit could devide the power by 10?

Keep in mind that the balast measured 1.2 ohms.

Tim, does the fuel pump run? If these voltages are correct, the pump shouldn't run. The only cause that I can think of is a faulty pump relay.
Not sure if the problems I had are exactly the same as yours, but they sound similar.

Might be your ignition switch under the steering column. Mine was burned out -- preventing juice from getting to my pump.

One way you can check this is to get a couple of long wires and hook your pump directly up to your battery. If you get a start, then you know something in between is keeping your juice from getting there. If you've already swapped a good relay into your fuel pump relay (should be second one from the firewall), and checked your resistor, the ignition switch is where i'd go.

It's a bit of a pain to get to, but at least it's cheap.
unplug the pump and see what the voltages are (or was that measurement made with the pump unplugged) -- a shorted pump would show values like that and the Balast usually will keep the fuse from blowing (with the balast in the circuit there's only like 11 amps being pulled by a shorted pump).

You didn't mention any voltage measurements at the balast resistor - take those under bothe sets of conditions -- the pump side of the balast should "roughly" correlate with the info you collected at the pump plug and the other side of the balast should show 12V with key-on...

The thinking is that if the plug shows data which does not correlate with the plug - then you've developed resistance or damage in the harness between those two-points, if the pump side of the balast shows what you're seeing at the pump and the other side of the balast is not exhibiting battery voltage, then the issue is supply side from there (relay/socket/fuse etc...)

When you find that the supply at the ballast is inadequate, pull the fuel-Pump relay (the 2nd on from the front - the front one is the o2 heater relay) and probe it's contacts -- I believe that pin 5 is where the fuseable link connects (verify this the ballast wasn't in the original '87 design, so some of the installations were retrofitted to include the necessary, if pin 5 isn't hot key-off relay out, then pin 3 should be - you'll get the idea -- if 5 is hot then 3 represents the output of that relay to the balast.. or numbers are swapped) - Verify teh connections for the fuseable link and it's integrity - you can load the circuit (jumper from 5 to 3 - and measure there -- jumper will pass the pump current - could get a little warm be careful.

For reference (something not often posted) I've added the pin-out and socket diagram for the relays used on the MJ/XJs

Euro Std     Jeep Socket        -- function
 (Bosch)        number      
________       _______           ______________ 
  -30-            1              Common
  -85-            2              Coil (usually the ground)
  -86-            5              Coil (usually the V+)
  -87-            4              Normally Open
  -87A-           3              Normally Closed

Looking at the SOCKET (relay removed) kinda is like this
                      |         |
                     2|  ____   |5
Uhmmm. Thanks satan. :D I will run through all that tonight. <Print selection> ;)

The measurements from above were with the fuel pump unplugged.

I will check the ignition switch. I have a spare. :)

I am also going to do the wire straight from the battery for a second or two.

If that data was with the pump unplugged then I'd suspect the harness or the source voltages -- a quick look at the ballast numbers and the open-soket relay numbers should identify "how far up" the path to look...

Pray for low at the relay socket so you've only got a foot or two of harness and a fusible link...

Now something that I shoudl really ask -- and PLEASE excuse me if you're already here, but sometimes, y'gotta check...

you posted that you verified the meter...

... but your "key on" voltage measurements are made with the key on AND when the pump is supposed to be running (the first 1 to 3 seconds after the key is turned "on"), right?
(good thing it's hard to mess-up the cranking position or I'd be asking another stupid question)
I checked last night.

To answer your questions: Yes, I did check it imediately after the key was turned on (when the pump would still be priming to the correct pressure). No stupid questions can be asked. Only not asked. :)

Well I used a jumper and the new pump works. I also started the wonderful process of cleaning dirty wires to find the color and probing to find voltages. What I found is that the primary side of the fuel relay is fine but the secondary side (pump side) is not passing voltage. I probed this from the bottom of the socket so the connection is still in question but it looks like it is right at the fuel pump relay. Funny thing was, I swapped the O2 heater relay in (same part number) and it appears to be fried too. That would explain the fuel comsumption and realitively long open loop running. :)

I am going to get some blade connector and 10ga wire and wake jumpers for the secondary side of the relay slot. But I am pretty sure it won't start because the Primary side of that relay is the ground for the ignition coil, I think. But the pump will run like it normally would.

I am also going to get a hand full of pick-and-pull relays to use and have for spares. Probably a Balast resistor too. And a couple fusible links. And....Well you get the idea.

Thanks a bunch for the help.
Actually (if I remember right) the relay feeds the injector-drivers (current path is back to the ECU where it's switched-out to fire injectors, should work if you supply voltage to socket #4 of the plug, though..)

Somewhere someone should round-up all the OEM relays in the yards and make a killing selling em back to us!

Good luck

I got a handful of relays from the Junkyard. I hard wired a new power from the orange/black wire that goes to the Ballist basically taking it out of the equation. Then I hard wired a new ground from the tank/pump to a nice chassic ground.

It now starts and dies instead of not even starting.

There is 9.xx VDC at the ballist connection when cranking but as soon as the key is put if the RUN position the engine just dies. The voltage fall to about 50 mVDC.

The fuel pump is a 92 model and it is on an '89 fuel system.

Any suggestions?

Tim (lacking the time to properly diagnois)
Did you ever check that ignition switch?

I know you mentioned that you had an extra.

May be what's keeping the juice from continuing to the pump when it's switched into "run".

When I had fuel pump problems it was the last thing I checked. It was literally working one second and then not the next. Checked it out and part of it was burned out/melted. But not the entire thing, mind you, just one section. Might explain why yours is obviously sending power in one position and not the other.

Other than that, I'm stumped.



I will be switching it to a toggle ignition as soon as I find a good switch. It worked great for my race car and I bet it will work for this too.

I am actually pretty sure that the RUN position is working because most of my power readings have been in the first 3 seconds of putting the key to the on position. But the toggle will fix that entirely if there is a problem. Thanks a bunch for you help too.
Thanks for the help. I have been doing a bunch of probing and finally just gave up a rewired a brand new 10 guage wire from the balast to the pump and then a ground back to the engine bay. It works now perfectly.

I am pretty sure that is was the bulkhead connector was the culprit. There was 1.6 ohms of resistance between the engine bay and the pump. I didn't feel like messing with it so I just rewired. :D