Greetings everyone,

I went to the grocery store today, and when I came out my fuel pump decided to die. I tried having a kind fellow crank it while I smashed the gas tank with a rubber mallet... didn't work this time. This is the third time this has happened within the last year and I am trying to figure out the cause. Here is some information about the vehicle.

D15B7 swap into 1991 hatch.
Rywire conversion (cut and solder, not full harness).
Last two replacement pumps were Advanced Auto pumps.
Main relay ~6 months old. Clicks on and off.
No fuses blown under dash.
No fuses blown under hood.
Fuel pump does not prime when jumping pins 5 & 7. Also tried jumping pins 1, 5 & 7, still nothing.
Ground at thermostat is clean and tight.
Pulled a CEL code 3 MAP sensor earlier in the week.

That is all I have now. I am stuck at the grocery store and I left my multi-meter sitting on my motorcycle so I can't test for voltage to the pump. Last two times voltage was at the pump. Replacing the fuel pump let the vehicle fire up no problem both of the previous times. I had one failure occur within 2 months of the first. I assumed a poorly made fuel pump was at fault. Starting to think otherwise. The fuel pump was literally priming fine every day since I have replaced it last time. Every time I start the vehicle I allow the pump to fully prime before starting the vehicle.

If anyone has any ideas what could be causing such frequent fuel pump failure please feel free to expand. I really don't want to continue wasting money on pumps. Thank you.

Also, I have not tested another ECU after failure.
The vehicle idle lopes at low temperature. Have yet to clean spring in IACV.

 

Generally speaking original Honda fuel pumps will last almost indefinitely. I'd tend to doubt the quality of the aftermarket pumps more than anything else at this point, unless you have a lot of rust or debris floating around in the tank.

 

When you jump main relay connector pins 1-5 or 5-7, check whether the fuel pump connector gets battery voltage to body ground. If so, check the fuel pump ground wire. I think the pumps you replaced may have been fine and that you instead have an intermittent fuel pump ground wire problem.

 

I got the chance to get the multimeter on it. Here is what I found.

First I disconnect the fuel pump connector located under the rear seat. Then I grounded the multimeter to chassis.

With the relay in I got ~18v for the duration of normal fuel pump priming. Continuous voltage to the fuel level sensor.

With the relay removed and pins 1 and 5 jumped I got 0v to the fuel pump. Continuous voltage to fuel level sensor.

With the relay removed and pins 5 and 7 jumped I got 0v to the fuel pump. Cotonuous voltage to the fuel level sensor.

Any ideas? Thanks in advance.

 

Something is very wrong here^ because battery voltage is only about 12.5V. Do you mean 18 milli-Volts? Battery voltage should only be detected for 2 seconds after turning the key from off to ON(II).

What voltage do you measure across the two battery posts?

Sorry, the wrong main relay plug pins are being jumped here^. Redo your voltage test after jumping pin 1 to pin 7 in the unplugged main relay connector. Do the test with the key in the off position.


Redo this test as I described for the 1-7 test, except have the key in the ON(II) position. You should measure continuous battery voltage.



Main relay connector for 91 Civic:

.

 

Alright, I got the chance to run those diagnostics.

First I disconnected the fuel pump connection under the rear seat. Then I plugged the positive multimeter lead into the yellow and black (fuel pump power) wire. Then I grounded the multimeter to the chassis.

With main relay pins 1 & 7 jumped and the key OUT of the ignition I recorded ~18.2v

With main relay pins 1 & 7 jumped and the key in the OFF position I recorded ~18.2v

The main relay pins 1 & 7 jumped and the key in the ON(II) position I recorded ~18.2v

When testing the battery I recorded ~18.2v.

Looks like we have found a possible underlying problem here! The voltage regulator has failed and the battery is overcharging. My assumption is that, due to an overcharged battery the fuel pump was killed by a surge of voltage while attempting to prime.

So, I have a few more questions now.

1. What other electrical parts are likely to be damaged due to the electrical surge? (injectors, sensors, etc)

2. Does the alternator contain the voltage regulator within itself?

3. If the voltage regulator is contained within the alternator, would putting an aftermarket inline voltage regulator be a practical?

4. Can I do multimeter tests (resistance, continuity, etc) on a junk yard alternator to check whether or not it is in working condition?

5. Should I just fix my motorcycle?

Thank you for the help.

 

Yes

No.

If it's a mitsubishi alternator (which I kind of suspect since this is a typical failure for them) you're probably best off to replace the alternator, especially if you happen to live in area that uses road salt since the chances of it ever coming apart in that scenario are slim. If it's a denso it's probably worth buying a regulator for.