electrical / fuel system help needed!
 

Guys,

I have a 94 single turbo w/ PFC, 1600cc secondaries, RP fuel pump, aeromotive FPR and supporting mods. The car was built and tuned by Steve Kan.

I ran into a knock problem a few weeks ago at low boost (6psi)/part throttle (4000rpm), during which I think I heard knock. Lasted maybe 2 seconds. Knock reading 140s (usually 10-20). Car drove fine afterwards, so I took it home w/o boost. After this, car started idleing very poorly and I thought the worst. Compression test was ok. Ended up being a fried MSD box and pulled ignition harness connection causing the car to run on one rotor. Ignition harness was returned to stock and the car idled/drove well.

I carefully fed in some boost (2-3psi) at about 3800rpm, and the knock level started going up (60s). I of course backed off immediately. I thought it must be a lean condition, likely something to do with the secondaries. I went to check the fuel system today, and I'm running into trouble immediately.

I jumped the F/P to ground and checked voltage at the pump (W/R wire)...6.6V! WTF? :mad:. I pulled the connector off the fuel pump, and read 12.5V (batter voltage). Weird. I plugged it back in and rechecked with the same findings. I checked the Fuel Pump Resistor, and its within spec. I bypassed it anyway and still got only 8V at the pump w/ the harness plugged in. The Fuel Pump (speed) relay works normally when voltage is applied across it.

Low voltage at the pump could certainly explain my lean condition. I just can't figure out why the voltage is low. I've looked at the wiring diagram, but I'm stumped. Searching got me this far, but now I need some rotary guru love.

Help!

Thanks,
Stew

Ok, I rechecked everything again, and I put a jumper across the fuel pump resistor and got 9V at the pump. It also definitely sounded like it was flowing more fuel. Grrrrrrr. I need to figure this out or this car will drive me postal.

Stew

check the voltage across the alternator. Make sure that you're getting at least 13V according to the PFC. The alternator could have been slowly going bad. I have seen this problem before.

OK, I'll check this evening. I seem to recall getting decent voltage at the PFC.

BTW- I checked the voltage at the W/R wire in my other (for sale) RX-7 and got similar voltages. Both cars' pumps DO switch over to battery voltage at the W/R wire if you get the RPMs up to 5k or so neutral.

Thanks for chiming in, Steve. I'll get her running soon. PLX-250 WB on the way!

Stew

I have seen a bad pump that still worked but offered too much resistance and caused a voltage drop in the system when plugged in, but when checking the system "unloaded" with the pump unplugged you would have full power.

Tap your voltmeter probe into the wire before the fuelpump plug...jumper the switch, and observe your 12vdc. Now plug in the pump, you'll hear it run immediately and then see the voltage drop. A bit of a drop, a volt or so, I would consider normal. More than that, and you have a pump slowly going bad that is putting too much load on the system.

I've been too busy with work to play with the 7, but your test is essentially what I did. I tested the voltage at the plug with the pump unplugged and F/P jumped to ground, which yielded ~12.8V (same as battery voltage). Plugged the pump in, dropped down to ~7 Volts. Battery voltage remained constant w/ pump running.
So you think this may be a pump on the way out?
Thanks,

Stew

IF it drops down that low, yes. How else would you explain a drain of almost 50%? You should be able to probe that wire with the pump running and get 9-12vdc (depending on whether the resistor is on or off, for low speed or high speed driving).

I thought something might be wrong with the resistor or the fuel pump speed relay, but these tested within spec. With the resistor connection jumped, I got about 9V at the pump with F/P jumpered to ground. When I measured voltage with the car running, I could get about 12.5 V at the pump if I took it up to about 5,000rpm quickly (in neutral). Hmmm...I wonder how old this pump is. I'll ask Steve Kan if he knows.


Stew

Jumping F/P to ground should give battery voltage at the pump, regardless of the resistor, right?

Stew

I don't think we ever replace the fuel pump here. I think it was done previously by someone else so I would guess that the pump is at least 2 yrs old.

Is there a way to measure resistance across the pump to see if its bad? If not, which pump do you guys recommend (damn money pit) :).

I'm suprised no one has mentioned this yet, or maybe I have no clue.

Either way this may be a helpful read.
https://www.rx7club.com/3rd-generation-specific-1993-2002-16/fuel-pump-upgraded-wiring-476124/

I've never tried, and with something dynamic like an electric motor I'm not sure it would work. These are just my observations from dealing with 7's for a few years, especially when it comes to making cars run that havent run in a while, you run across siezed or weak pumps a lot.

Unless you're going over 400, I'd just run a walbro 255 for 100 bucks. Steve might beg to differ with me on that, and I would bow to his opinion. ;) Otherwise some of the standard more expensive stuff like cosmo/supra pumps or dual pump setups might be best.

Thanks for all the help, guys. I'll probably replace the pump AND rewire to be safe. I still need to check the injectors to make sure nothings sticking before I install the wideband and take her out for a test run. With my work schedule, I anticipate completion of this project in 3-5 years ;).

Stew

it sounds like youre doing this with the car off since you say you get 12.8 (battery voltage) at the pump connector. it could just be your battery dying and its voltage drops under load. try it with the car running and put the meter inline and check the current to see if its abnormally high. If the current draw is normal I would say the pump is fine.

I've tested with the car running as well. I get about 7V to the pump at idle and full voltage (14+) at the battery. If I rev the car to about 5k rpm, I see 14 V or so at the pump. I am concerned that at lower revs with load (boost), the pump may not be working efficiently. I know that voltage to the pump is determined by temp, fuel temp, TPS, boost, etc. I haven't had a chance to hook a voltmeter up to the fuel pump plug and drive the car, watching voltage real-time. I don't want to risk running lean during this type of test.

BTW, any idea how much current the pump should pull?
-Stew