Detailed fuel pump problem on my 93 - humming n worse

[Curtis_H]

I know you said you cleaned the tank before you put it back in, but sometimes heavier grit can get trapped in there and not come out so easily. It is possible that you have grit/slugs in the tank that are being sucked up into the screen when the pump is running and causing it to cavitate, hence the humming noise. they can either cause the pump to overheat and shut down or when enough have been sucked up it can clog the screen and kill the pump. Then, when the pump is off and no vacuum is present at the screen they will float back down and settle away from the pump until the entire cycle starts back up. Check the tank thoroughly with a flashlight, run a magnet on a stick and some reverse wrapped (sticky side out) duct tape on a stick all through the bottom of the tank before you put it back in the vehicle, just to help eliminate this problem.

 

[Nick_XJ]

I know you said you cleaned the tank before you put it back in, but sometimes heavier grit can get trapped in there and not come out so easily. It is possible that you have grit/slugs in the tank that are being sucked up into the screen when the pump is running and causing it to cavitate, hence the humming noise. they can either cause the pump to overheat and shut down or when enough have been sucked up it can clog the screen and kill the pump. Then, when the pump is off and no vacuum is present at the screen they will float back down and settle away from the pump until the entire cycle starts back up. Check the tank thoroughly with a flashlight, run a magnet on a stick and some reverse wrapped (sticky side out) duct tape on a stick all through the bottom of the tank before you put it back in the vehicle, just to help eliminate this problem.

oh this reminds me of something i forgot to mention.

I put 6 magnets on the bottom of my tank, 3 on each side (drivers and passenger sides of the bottom of the tank) before i put the plastic cover back on when installing my tank.

the magnets are small but fairly strong. I did this so that if there was anything left in the tank hopefully the magnets would draw them to the sides of the tank and not let them touch the fuel pump.

would these magnets have anything to do with the way my fuel pump is working?

also when I cleaned out my tank i looked in with a flashlight and i could not see anything, not even dirt. nothing rattled around when i moved the tank around either. I'm pretty sure its very clean. However I will double check it when I replace my pump again.



I checked for the bosch pump from Advance in Adrian, MI. I probably wasn't clear on that though, they do have the bosch pump in stock, but no sending unit/pump combo.

 

[Nick_XJ]

Alright here we go again with more info. :laugh3:

I was able to talk Autozone into exchanging my fuel pump for another AirTex (even though they already returned it). I installed it and yet again the exact same humming.

With that being said, I got my hands on a Multimeter and started testing my electrical. Here were my findings:
Where the ground from the battery attatches to the inside of the fender, it was rusty and wasn't grounding correctly. I cleaned that up and replaced the ground. Nothing changed.

I then unplugged my fuel pump at the connecter and tested the voltage there. There are 3 wires, Black, Purple, and Gray. When I connect Black + Gray via the multimeter, it shows 6.25V for about 2 seconds then goes to 0V. When I connect the Black + Purple I get 6.25V constant. Isnt' it supposed to be 12V?

Also, I am trying to figure out if its a ground problem or voltage problem, so i want to connect the positive end to the Purple Wire and the negative to a chassis ground. Should i connect the negative to the frame or to the bottom of the unibody?

 

[ChevelleSSLS6]

Is it airtek or airtex? I do believe airtex is an oem supplier, kinda like visteon (makes radiators and stuff), and ngk, bosch, etc... I wonder if the stuff they sell to RicerZone is the same stuff they sell to GM or Ford or mopar.

 

[Jeepcreep]

I fitted a new Airtex pump + frame and sender unit, about 6 months or so ago, I can hear it when it 1st. primes the system, but have never heard it after that. Your voltage numbers do not look correct to me, I would have expected around 12v, could be low voltage is your problem. Let us know how you get on.

 

[xjnco]

Have you replaced the fuel filter itself.
I recall having these symptoms several years ago, Replaced the filter , and WALLAH....
Sorry, I didn't see this thread earlier.

 

[winterbeater]

You should be getting 12 volts to the pump. (Assuming that you either don't have or have already bypassed the ballast resister.) But with the key in run position, the voltage only comes through for 3 seconds unless the PCM sees a signal from the CPK (crank position sensor). Best way to test for voltage is from the power wire to a clean spot on the body. If you get 12V that way, but then 6.5V when you test from the power wire to the black (ground) wire, you have found a problem. The ground has too much resistance. You can also put the multimeter on Ohms. Touch your meter leads together and you will get "0". This means no resistance, which is good. Then if you probe the black wire and the other probe to your clean spot on the body, you should also get "0" or at least very close to zero. If not, your ground is no good.

 

[Shorty]

I would suspect a poor ground for the pump in this case. The best thing you can do (for little to no $$) is run an additional accessory ground to the chassis. It will never hurt to have a better ground and it can eleviate that as a cause. Also, verify a good connection from the pump body to the sending unit assemble as well and you can scratch a poor ground off the list for a couple bucks in wire and a little of your time.

You should be getting 12 volts to the pump. (Assuming that you either don't have or have already bypassed the ballast resister.) But with the key in run position, the voltage only comes through for 3 seconds unless the PCM sees a signal from the CPK (crank position sensor). Best way to test for voltage is from the power wire to a clean spot on the body. If you get 12V that way, but then 6.5V when you test from the power wire to the black (ground) wire, you have found a problem. The ground has too much resistance. You can also put the multimeter on Ohms. Touch your meter leads together and you will get "0". This means no resistance, which is good. Then if you probe the black wire and the other probe to your clean spot on the body, you should also get "0" or at least very close to zero. If not, your ground is no good.

X3^

 

[Nick_XJ]

Thank you very much for pointing out the voltage. I called Airtek support and confirmed this information:

An Airtek fuel pump has 3 wires.
Gray - Positive - requires 12 volts
Purple - Sending Unit - requires 5+ volts (in the middle)
Black - Ground

The stock connector on my 93 Cherokee has 3 wires.
Green / Red stripe - 12 volts for 1 second
Blue - 6 volts constant (in the middle)
Black - Ground (beeps when u do a continuity check to ground)


The strange thing is the connectors dont line up correctly. Heres how they connect:
[fuel pump to the factory connector]
Black Wire connects to the Green/Red Stripe
Purple Wire connects to the Blue Wire
Gray Wire connects to the Black Wire

I asked Airtek's support what would happen if the pump was wired backwards and they said it would not pump any fuel and my vehicle would not run. However the wires tell a different story.

Tomarrow I plan on running a wire from before the Ballast Resistor to the Gray wire on the fuel pump, and connecting the Black wire to the chasis ground. I will post the results. If it still runs hooked up this way, I will drive it and see if it solves my problem. Hopefully it does! I will keep anyone watching this thread updated on my trials.

 

[winterbeater]

That will work. Or you could just cut the pump wires and splice the black to the gray and the gray to the black. This would be easy to solder and heat shrink on a bench.

 

[Nick_XJ]

Today I checked a lot more of the electrical connections to my fuel pump and here is my findings:

I checked the voltage on the ballast underneath the hood with the vehicle running. On one side of the ballast is a Green/Black wire. That runs the correct 14ish volts. However when I check the other side of the ballast, the Green/Red wire, it only has 6 volts.

It is a brand new ballast but assuming the worst I completely bypassed the ballast by connecting the green/black with the green/red directly. I took it for a test drive and the humming still persists without any change.

I then unhooked the fuel pump wire connection near the gas tank and with some gas still in the lines started it back up. Now suddenly I get the same voltage on both the Green/Black wire that connects to the ballast, and the Green/Red wire that connects to the other side of the ballast while the Jeep is running.

I reconnected the fuel pump connector and started the jeep back up, it went back to 14v @ green/black and 6v @ green/red.

Assuming maybe it is a bad ground I ran a new ground to the chasis and connected it directly to the ground wire on the fuel pump connector near the fuel tank. No change.

2 posts ago I described that the gray wire on the pump connected to the black wire, and the black wire on the pump connected to the green/red wire...I ran wires directly from the front and connected the positive wire from before the ballast directly to the gray wire (positive) of the pump. I then hooked the black wire of the pump directly to my new chassis ground. The pump turns on when you turn the key on but when you try to start the jeep it smells like gas really bad and will not run. It is pumping but I believe it's pumping in reverse.

Here are my new questions:
When I have my fuel pump hooked up to the factory connector, why does 1 side of the ballast read 6v and the other side read 14v?
What would cause this voltage to drop so much if its supposed to be 14v all the way back to the pump?
Any other ideas?

 

[winterbeater]

The ballast resistor's purpose is to make the fuel pump quieter after the vehicle is started up. You don't need it, and it is the first thing that you should eliminate if you have any problems. Most people got rid of that a long time ago. Bypass that and take it out of the equation. That way you will always have 12 V+ going to the pump. The resistor is supposed to drop the pump voltage from 12 to 6V after everything starts up. It's all working like it is supposed to. When the key is in start, the pump should get 12 volts, once it's in run, it should get 6v. Unless you bypass the resistor like everyone else has done to get 12V all the time. The big question is whether it runs good all the time now?

 

[89xj]

Here are my new questions:
When I have my fuel pump hooked up to the factory connector, why does 1 side of the ballast read 6v and the other side read 14v?
What would cause this voltage to drop so much if its supposed to be 14v all the way back to the pump?
Any other ideas?

the airtek pump i installed on my 89 had the + and - terminals on opposite sides than the oem pump had them on.

why does one side read 6 and the other side 14V. because a resistor decreases voltage. thats what resistors do, as resistance goes up, voltage goes down.

to voltage isnt supposed to be 14v all the way back. winterbeater %100 correct on how the system works and yours is working proporly.

 

[Pelican]

Interesting. On my 93 I have a green/black wire into the resistor with a reading of ~14.4v and a green/white out of the resistor with ~11.3v, both with the engine running. I have to question whether the pump will run on 6v (assuming I am following you correctly). In any event, unless all of your xjs are a lot different than mine, something is wrong if nick_xj has a 6v reading at the ballast resistor.

 

[Nick_XJ]

I would hate to jump the gun, but I believe that I fixed my problem! I will find out for sure tomarrow when I take it on a longer test drive. It turns out my Crank Position Sensor was fulty. It was not throwing any codes, but was not functioning properly. It only cost $61 at autozone and was pretty easy to replace. Yes a tight spot but only took about 20 minutes after I got the right tools. It's on the drivers side of the belt housing held on by 2x 11mm bolts.

After installing the new sensor I took a 20 minute drive and there was no humming, the pump did not give out, and my problem seems to be solved. However I will find out for sure tomarrow. Thank you all very much for your help!

Interesting. On my 93 I have a green/black wire into the resistor with a reading of ~14.4v and a green/white out of the resistor with ~11.3v, both with the engine running. I have to question whether the pump will run on 6v (assuming I am following you correctly). In any event, unless all of your xjs are a lot different than mine, something is wrong if nick_xj has a 6v reading at the ballast resistor.

winterbeater was correct about this, when you start your vehicle the fuel pump should get battery voltage (usually around 14v). After the vehicle is running it drops to 6v and that voltage goes up or down as ur rpms go up or down to send more or less fuel. Apparently the Crank Position Sensor controls this.

 

[Nick_XJ]

I'm sorry I have not been on to update this.

This is confirmed, my Jeep is fixed.

The Crank Position Sensor was the problem.

Again thank you all for your help, you can move this into a fixed or closed state.

 

[orange]

Excellent News! Congrads!

--Good Luck with "New" Jeep !--



(Thanks for update)

 

[joe_peters]

CPS = noisy fuel pump.

Gotta put this one in the note book.

 

[xjbubba]

CPS = noisy fuel pump.

Gotta put this one in the note book.

I'm sorry, but I don't buy it.
The pump's operation, other than not running if the ECM is not getting an input from the ignition module, is not affected by the CPS.
If so, I'd appreciate the "science".

 

[joe_peters]

I'm sorry, but I don't buy it.
The pump's operation, other than not running if the ECM is not getting an input from the ignition module, is not affected by the CPS.
If so, I'd appreciate the "science".

I think the "science" is there for the major part of his trouble--did you read his original post on 2-10-2009?

I don't know why his pump is less noisy now, but at least his xj is running, which I think was the major part of his goal.