Fuel pump shuts off after a couple minutes

[underscore]

Currently when I start the car up it will run and drive fine for several minutes, then shut off. It'll crank no problem but won't fire. If I jump B+ and FP in the diagnostic box it runs fine. I previously had an issue with a replacement EFI relay that was bad so I figured either that or the Circuit Opening Relay was bad. I swapped them both this morning, and it did the same thing. I thought maybe those were bad too so I immediately swapped the old ones back in just to move the car. Normally I'd just jumper it and swap parts after. However this time even with the cold relays swapped in it wouldn't start until I jumped the fuel pump.

This makes me think the COR is fine, and the EFI relay has to be fine or it wouldn't run even with the jumper (I realized just today that I had been reading the diagram incorrectly, whoops). However looking at the diagram for a 185 the COR is controlled by the ECU. What would make the ECU shut the fuel pump off after a couple minutes and refuse to turn it back on again until the car has sat a while? At this point I'm thinking I should swap in my spare ECU and see what it does?

 

[alltracman78]

You need the wiring manual. It's more complete than that.

The ECU uses the RPM signal from the distributor to confirm the engine is rotating to close the circuit opening relay.

I assume your fuel pump relay is already jumped? I know either Position will allow the pump to run but electronics can do funny things sometimes.

Lots of people have been having trouble with their ecu's lately. Wouldn't hurt to make sure it's good in there.

 

[underscore]

I think I have the wiring manual but I didn't know it went into ECU functions. I saw the FC pin on the ECU pinout and assumed that was all I could learn about that path. If it goes off the distributor then hopefully this doesn't mean that's on the way out.

Yes the fuel pump relay has been replaced with a factory jumper. I meant to mention that but forgot.

The ECU had the caps replaced a couple years ago now but the car has been parked for the last couple years so it's probably worth taking another look in there.

 

[underscore]

I plugged in my spare ECU and the result was the same. The wiring manual doesn't say anything about how the FC pin is controlled but I followed the checks and the distributor tested okay while cold, which I expected. I guess next is to try and test it while still warmed up.

 

[underscore]

Alright so I ran it til it shut off, which was quicker this time, and the distributor tested slightly differently but still okay according to the resistance checks listed in the wiring manual. I swapped the COR again for kicks and it ran much longer than before, the fan actually cycled a couple times, but I only made it about a block before it shut off and I had to use the jumper to get home. No codes when I check using TE1 and E1 either.

 

[93celicaconv]

On an ST185, you also have a Fuel Pump Control Relay and a Fuel Pump Resistor. Have you checked both of these when you have an engine stall and/or no-start condition? I'm thinking your issue resides there, not in the COR or the ECU (especially given you swapped both of these and had no real change in symptoms.

The COR is first energized when the starter relay energizes (when your engine starts cranking (or turning over)). As soon as the engine starts and you release the key (which makes the ignition go from the START position to the ON position, the starter relay de-energizes, which cuts power to the COR...except the ECU, now getting input from the distributor, knows the engine is running, and energizes the COR. Since your engine seems like it starts (or wants to) and often runs a bit, it would seem the ECU is properly controlling the COR after the engine has started.

The ECU also controls the Fuel Pump Control Relay. This particular relay is always closed in either one position or another (it simply switches which position it wants to close regarding how to feed power to the fuel pump). One position is for full 12V power to the fuel pump. The other position feeds power first through a Fuel Pump Resistor (which reduces output voltage) and then to the fuel pump. To me, it seems like the side of the Fuel Pump Control Relay that feeds power to the Fuel Pump Resistor is not closing properly or the contact of this side of the relay went bad, or the Fuel Pump Resistor itself has gone bad. I don't know all the conditions that need to be met for the ECU to make a switch, but it seems to me when the ECU wants to make this switch, something on the side that involves the resistor isn't working well anymore. Just an assumption based on what information was provided.

When you jump B+ & FP, you are bypassing not only the COR, but also the Fuel Pump Control Relay and the Fuel Pump Resistor (as shown on your wiring diagram).

 

[underscore]

The fuel pump relay has been removed and replaced with a jumper so it and the resistor pack are out of the equation, but I could try putting it back in to see what happens.

 

[93celicaconv]

The fuel pump relay has been removed and replaced with a jumper so it and the resistor pack are out of the equation, but I could try putting it back in to see what happens.

So how do you have the jumper set? Jumper'd to the 12V full voltage side (so you have the highest flow rate possible at all times)? Why did you remove the Fuel Pump Control Relay & Resistor? Is it possible your engine stalls after it warms up is due to flooding with fuel?

 

[underscore]

Yes, it's got the factory jumper from some of the 2WD models to run 12V all the time. I have an aftermarket fuel pump and since I was upping the boost I wanted to eliminate a possible point of failure that could cause it to run lean. I'm pretty confident it isn't flooding because as soon as the jumper is in place it fires back up and runs fine. I'd expect it to still fail to start or still shut down every so often if it was flooding because it would still flood with the pump jumpered.

 

[alltracman78]

That jumper in place of the fuel pump relay can't cause the engine to flood. All it does is supply battery voltage to the pump. That's not enough to come even close to flooding the engine.
Honestly, it's almost impossible to flood an EFI engine to where it won't start. You would pretty much have to have all your injectors stuck open for that to happen.

Have you checked to make sure the ECU is still grounding the COR when it won't run?
And that the COR is still getting power?

Set everything up ahead of time so all you have to do is look at the meter.

It's possible you have a damaged wire or connection somewhere in the circuit.

When it won't run wiggle the wiring around the COR. See if that does anything?

If your COR is getting the power and ground it's supposed to I would check for voltage along the rest of the circuit when it won't run.
I laid out all the connection points in that dude from Colorados thread. With the black 92 GT4.

 

[underscore]

Throwing a meter on the FC pin to confirm my suspicions was the next thing I was thinking of too. With the jumper in place I'm thinking I can just unplug the COR and put a nice spade terminal in the plug and see if the ground connection drops out whole driving around.

 

[alltracman78]

*I copied this from the other thread, in case someone is looking in the future, so all the info is in one place*


The AFM isn't the problem. They're probably confusing with the 165. IIRC The fuel pump circuit runs through the AFM on those. That's the safety, instead of from the ECU like the 185 and up.

AFM wiring is the same on all 185s. 5V supply voltage, airflow signal, air temp signal, and ground. Homologation got an extra ground for some reason. But like I said it splices into the other ground about a foot back. I don't have the pinout in front of me but probably the same order too.

There may not even be that many (7) differences between the homologation and the 92/93 (facelift) pinouts. And the differences are things like EGR , EGR temp. Plus of course the IC level sensor. Nothing to do with the fuel pump. I have the pinouts for both, but not in front of me. The ECUs actually have the pinouts printed on the board FYI. You can open the top cover and read them. I forget if the second board (on the facelift ECUs) is in the way or not. I don't think so.

Fuel pump wiring is going to be the same across the board. Wires possibly in a different position, but works the same way. You can use the regular manuals for this. Just double check you have the right wire.

I doubt it will do anything for this issue, but don't forget you can put the ECU into a more sensitive diagnostic mode by jumping TE2 and E1 and drive with the jumper in place. Benefit of having a facelift ECU.

Did you check and see if the ECU was still grounding FC when the pump stopped working?

I don't have any other info about the FC circuit specifically, but I can't imagine it's any more complicated than the ECU sees the engine rotating (from the dist) so it grounds the circuit.

The wiring manual has info on checks you can do for specific circuits. Tells you what voltage or resistance to look for for specific wires. It's on the page after the diagram. Or second page maybe. Probably won't give you anything but worth a look.

You may have 2 parts that are bad or going bad.

You can make jumpers to eliminate certain parts. For example removing the FC pin from the COR connector and adding in your own pin with a wire. When you ground your wire with the key on it will have the same effect, causing the COR to close and make the pump run. It will do it even with the engine not running, just like jumping FP and B+.
This will eliminate the ECU and the dist. See if it still cuts out.
Thinking about this, I doubt it's the dist. If the ECU isn't getting an RPM signal it shouldn't send an ignition signal to the igniter either, which it obviously is if the engine runs with FP and B+ jumped.

It's also possible one or more of your engine harness grounds is bad or corroded. Anything that grounds through the ECU (like your injectors or the COR) ultimately grounds through the E1 and E2 wires. If there's a bad, damaged, or corroded splice or connection that will effect how much voltage things see. Too little voltage and things won't work, like the COR dropping out. Wouldn't hurt to do a voltage drop test on the E1 and E2 wires. I wouldn't suggest checking resistance, doing that can sometimes miss only partially damaged wires. A voltage drop test will let you know if there's an extra load (in this case damaged/corroded wire).

*One other thing to add, if you do end up doing anything with the wiring, don't use aftermarket spade connectors or paper clips or anything like that. It's one thing to use it in something like the DIAG box that doesn't have anything plugged into it. Unless it's the same size as the original pin you'll "stretch" the female pin and the original pin may not correctly make contact afterwards. Which causes obvious problems. I only use Toyota pins for things like this. I have a bunch of spare wiring that I cut up as needed. Old O2 sensors are perfect for this, a lot of the pins in the engine harness are the same size.*

 

[alltracman78]

So I took a quick look at my almost 20 year old hand drawn diagrams comparing the 2 ( RC and 92). Looks like there are more than 7. But nothing to do with the fuel pump. Things on the RC like exhaust temp for the cat, and the EVAP system (controlled by a thermal valve on the US cars, instead of a VSV), no O2 heater, and interestingly the RC apparently has an ABS output. The US cars don't.

By the way, did you have a homemade jumper in the fuel pump relay location at any point? If you did, you may have spread the pins too far and your factory jumper may not be fully contacting? This could possibly be causing a problem as well. And jumping FP and B+ will bypass this....

 

[underscore]

The jumper is a factory one from a FWD Celica, it was snug before but I'll double check it still is.

The AFM isn't the problem. They're probably confusing with the 165. IIRC The fuel pump circuit runs through the AFM on those. That's the safety, instead of from the ECU like the 185 and up.

The only way I could see it mattering is if the ECU is using the signal from the AFM to decide what to do with the FC pin to control the COR.

There may not even be that many (7) differences between the homologation and the 92/93 (facelift) pinouts. And the differences are things like EGR , EGR temp. Plus of course the IC level sensor. Nothing to do with the fuel pump. I have the pinouts for both, but not in front of me. The ECUs actually have the pinouts printed on the board FYI. You can open the top cover and read them. I forget if the second board (on the facelift ECUs) is in the way or not. I don't think so.

Some of them I can find descriptors for, but others I have no clue. The RC apparently has:
EVP - ???
ISCV - ???
RLY - ???
ECO - ???
ACT - ???
ATS - ???
CCO - ???
WIN - Intercooler ECU
that the USDM doesn't have.

The USDM has:
HT - O2 heater
STJ - Cold Start Injector
EGR - EGR VSV
that the RC doesn't have.

I doubt it will do anything for this issue, but don't forget you can put the ECU into a more sensitive diagnostic mode by jumping TE2 and E1 and drive with the jumper in place. Benefit of having a facelift ECU.

I might have to try that next to see if it can find anything.

Did you check and see if the ECU was still grounding FC when the pump stopped working?

I tried, I can report back that with the probe stuck into the back of the plug with the COR installed it reads a bit of resistance and voltage to ground when running. That was annoying because I couldn't just leave the meter on short test and listen for the beep to stop as I drove around. If you switch the engine off and the key to ON it shows 12V as expected. The car did shut off once, but when I coasted to a safe spot to look at the meter and crank the engine it just fired up and ran normally the rest of the evening. :bangshead:

I don't have any other info about the FC circuit specifically, but I can't imagine it's any more complicated than the ECU sees the engine rotating (from the dist) so it grounds the circuit.

The wiring manual has info on checks you can do for specific circuits. Tells you what voltage or resistance to look for for specific wires. It's on the page after the diagram. Or second page maybe. Probably won't give you anything but worth a look.

That's the issue I'm having, there's no info on the FC circuit. Correct operation, testing procedure, functionality, nothing. The only place it's mentioned is in the terminal list: FC - Circuit Opening Relay.

Thinking about this, I doubt it's the dist. If the ECU isn't getting an RPM signal it shouldn't send an ignition signal to the igniter either, which it obviously is if the engine runs with FP and B+ jumped./quote]

I would expect it to throw a code for that too (same reason I was doubting the AFM), since there's 2 possible codes for dizzy faults.

It's also possible one or more of your engine harness grounds is bad or corroded. Anything that grounds through the ECU (like your injectors or the COR) ultimately grounds through the E1 and E2 wires. If there's a bad, damaged, or corroded splice or connection that will effect how much voltage things see. Too little voltage and things won't work, like the COR dropping out. Wouldn't hurt to do a voltage drop test on the E1 and E2 wires. I wouldn't suggest checking resistance, doing that can sometimes miss only partially damaged wires. A voltage drop test will let you know if there's an extra load (in this case damaged/corroded wire).

That's another good thing for me to check, the inconsistency is killing me though. I much prefer when something breaks and stays broken so I can just find it and fix it lol.

 

[underscore]

I checked on the fuel pump relay jumper. It's still snug and the terminals are still clean with the dielectric grease on them. For kicks I opened up my spare ECU to try and trace the FC pin backwards but after a couple transistors it went into an IC on the secondary board with a lot of pins so I gave up.

 

[alltracman78]

Some of them I can find descriptors for, but others I have no clue. The RC apparently has:
EVP - ??? *Evap electric VSV. USDM uses a bimetal VSV.*
ISCV - ??? *I don't have this on my RC pinouts.*
RLY - ??? *To the level sensor in the IC. Grounds while "full". This is the pin you ground if you're using a facelift engine harness with the RC ECU*
ECO - ??? *Economy button?*
ACT - ??? *To AC control. US has this also. Possibly engine RPM?*
ATS - ??? *To AC control. US has this also.*
CCO - ??? *IIRC This goes to the EGT probe after the cat*
WIN - Intercooler ECU
that the USDM doesn't have.

The USDM has:
HT - O2 heater
STJ - Cold Start Injector *JDM (including RC) has this also.*
EGR - EGR VSV
that the RC doesn't have.

Answers are in bold.

I tried, I can report back that with the probe stuck into the back of the plug with the COR installed it reads a bit of resistance and voltage to ground when running. That was annoying because I couldn't just leave the meter on short test and listen for the beep to stop as I drove around. If you switch the engine off and the key to ON it shows 12V as expected. The car did shut off once, but when I coasted to a safe spot to look at the meter and crank the engine it just fired up and ran normally the rest of the evening. :bangshead:

From what I understand, originally when it died it would stay dead for a while? If that's the case it seems like disturbing the FC wire makes a difference? You may have an issue there? Take a very close look at the wire.
I've had a code for the AFM signal because when they crimped the wire at the factory a TINY bit of insulation was under the crimp. Drove for years without a problem and then when I disturbed the harness and connector (was actually when I was playing around with combinations of the RC and US ECU and AFM) the fault popped up.

That's another good thing for me to check, the inconsistency is killing me though. I much prefer when something breaks and stays broken so I can just find it and fix it lol.

This is the fun part! It's what separates the men from the boys, as they say.


If this makes you feel any better, here:

http://wilbo666.pbworks.com/w/page/4360 ... ng%20Relay

https://www.toyotaguru.us/engine-contro ... relay.html

 

[underscore]

Yes originally it would die and stay dead until I put the jumper in. The wires got a fair bit of jiggling and being moved to different orientations because when I swapped the ECU and COR I didn't fully install/uninstall them. It's a bit hard to get good eyes on it when it's all jammed under the dash but at least if I need to replace the FC wire it's not a long job lol.

Those diagrams are very helpful! It looks like Toyota used the same setup for multiple cars so I'll consider it safe to assume the 185 is the same. Though I'm not too sure what's going on with those ignition status signals coming into the ECU in the second one. Annoyingly the NE signal dropping out only triggers a code if the engine is above 1000 rpm, so if it's dropping out at idle and killing the engine I won't get a code for it.

More testing I suppose.

 

[Gert]

Can't help you here. Most things are already spoken.
Other ECU = same problem.
So, nothing left then check if relays are receiving power and ground and broken or corroded wiring.
If all that is good and the "kill signall" is coming out off the ECU, I really have no idea how or when the Toyota ECU is programmed to do this.

Good luck.

 

[underscore]

I put over 1000km on the car this weekend and it never shut off so I'm not sure what to do at this point. My best guess is something had corrosion on it after sitting for a couple years but now it's been cleaned off.

The alternator wasn't working on startup a couple of times, so I might have to dig into that now instead lol. It always came back if I shut the car off for a minute and then tried again though.