Fuel pump not priming - car slow to start - why?

[Baba Galant]

Looks like the bloody relay buggered something up.
The cable came this morning so plugged it all up and have a bunch of errors that don't clear.
No flashing lights but it did start blinking the TCL light the other day in boost coming home. Didn't come back after switching off and on.

Got a feeling these may share the same 5v line.
Below is a snapshot of Evoscan live readings at idle. Starts and runs fine.

Not screaming help, but if anyone has seen these errors as a group it may save me some time.

 

[BCX]

Evoscans DTC implementation is crap in my experience.

Are you getting the same errors from CEL earthing out pin 1?

They all share 5v supply generated by the ecu - pin 81

also Ecuid of e7e1 is of interest to me, unrelated though.

 

[Baba Galant]

No, no hard error codes. The ecu is a metal box loaner from TME while my plastic ECU gets a remote tune

 

[BCX]

Wiring diagram of the relay if youre interested.

 

[Tony_T]

Thanks for the diagram, @BCX.
My concern was that for the relay coil to have burnt out an awful lot of current had to have flowed through pin 35 of the ECU, possibly damaging the transistor. In that instance transistors usually fail to a short circuit, but what isn't shown in that diagram is that output pins like that often have a high value resistor feeding in to a multiplexer which gives input to the microprocessor such that a 'logic high' is seen when the relay is de-energised as a check of the continuity of the relay coil so if the relay or equivalent load is missing or the transistor is shorted a fault will be detected. This may be the source of the error.

Alternatively the TCL is associated with the two solenoid valves and the vacuum storage canister which are all very close by that resistor and relay for the pump so could have been disturbed in the course of the other work. Any small leak in those hoses or the canister will give TCL errors.

And just to add to the confusion I've had bad contact in the stop lamp switch on the brake pedal cause a TCL error too!
Aren't these cars wonderful!
Edit: I'm too scotch to spring for Evoscan!

 

[BCX]

Ive got a few bench ECUs, considering they dont have any sensors, actuators, relays, solenoids, etc connected, they dont complain about much considering.

Analogue sensors out of spec (ie 0v) is usually what throws a cel on the bench ECUs, not much else.

ECU doesnt have much in terms of checking feedback, and even if the hardware did, the software doesnt look to do much.

My FPR solenoid is in the bin, plug floating about, no complaints.

 

[Baba Galant]

These cars are still intriguing to me, seperate ECUs, ayc etc and AWD.
Other than the timing belt and shocks I haven't had to fix anything.
Almost no one knows what they are still, suits me and best off all it shits my brother in law like you wouldn't believe!
He's been warned by Vic roads to get a LMCT licence or stop buying and selling cars.
Looking at new compressor wheels and housings ATM for a little more air at freeway speeds.
Working from home now almost permanently and need to entertain myself.

 

[Baba Galant]

Been at it for a while with the TCL error appearing again now and then, never though without boost.
Re-tested my vacuum control solenoid for the TCL side of the throttle body and initially was getting a 1K ohm's reading. Could barely get the solenoid the click until a bit of WD-40. From there the coil read about 38 ohm's. I believe the could may be dodgy do I'll need to source a new one without going broke or a good used one of any around.

After lots of reading and checking APS and TCL sensor's, I believe our write ups are incorrect.
The APS is a dual operation potentiometer, pins 3 and 4 set the base idle feedback to the ECU while pins 1 and 2 provide feedback to the TCL ECU for pedal position.
In the write up I've read, it's always stated to use a multimeter to set base APS when switch just goes off. This is in contradiction to APS operation tables as shown in the manuals.
Thinking about it if set to just open circuit you don't actually use the base idle portion as pins 3 and 4 remain open at all times.
The images bow show the operation at idle as continuos not open.
If anyone thinks this is wrong let me know as I'd hate to post misinformation.

 

[BCX]

I dont have tcl, so cant confirm... But the impression from my magna mate is the APS and TPS are the same sensor (so same pinout) just on different parts of the tb.


Pin 1 = gnd
Pin 2 = idle switch (tied to ground when on)
Pin 3 = signal of internal potentiometer
Pin 4 = 5v

 

[Baba Galant]

I'm quite sure they are too. They're listed as different part numbers and I haven't seen a cross reference that is conclusive.

 

[BCX]

They are slightly different, seem to be in reverse order:

APS pinout:
Pin 1 = 5v
Pin 2 = signal of internal potentiometer
Pin 3 = idle switch (tied to ground when on)
Pin 4 = gnd

TPS Pinout:
Pin 1 = gnd
Pin 2 = idle switch (tied to ground when on)
Pin 3 = signal of internal potentiometer
Pin 4 = 5v

 

[Baba Galant]

Hai caramba! Fortunately I went with the part numbers and have already installed my APS side.
The TPS was an equivalent for about $20.
It's here but not checked or installed.

 

[Baba Galant]

Back on to the original thread, the slow to start dilemma has been giving me the shits and solved my pain in the butt issue.
Forgetting about the dual speed relay set up and fuel tank seals, I set up a relay to power up the fuel pump at battery voltage at all times.
However, my car doesn't prime every time I want to start it.
Only at cold starting will my fuel pump prime. After that it's no prime and I suspect this is related to our original FPR's and fuel pumps where pressure holds after stitching off.
I run an Aeromotive stealth 320 and a Tommei FPR, once the car shuts the fuel pressure is gone.
I believe theoretically that EFI vehicles run with a cold prime and from there it's oil pressure based to switch on the fuel pump at cranking.
We may have engine temperature come into the equation but I can't verify this.

Today I wired in a momentary on button that bypasses the relay and energises the fuel pump for about 1 second.
Instant start every time hot or cold. For now the button sits near the fuel cap release lever and I'll change that in a few days after I am certain a 20 amp mom on switch is enough to cope.

 

[Baba Galant]

The crank but won't fire gremlin is back.
Regardless now if pump is primed, 1 in 3 starts takes 4 -5 secs of cranking.
I have an led running off the fuel pump feed lines so when it cranking you can see it light up, fuel pressure normal.
Seen a few glitchy starts where it actually activates the fuel pump but only for a split second, hot or cold dose not seem to be a factor.
All I can think of now is to monitor the 12 volt feed to the coils and look over all connections that feed these lines.

 

[Kaldek]

The crank but won't fire gremlin is back.
Regardless now if pump is primed, 1 in 3 starts takes 4 -5 secs of cranking.
I have an led running off the fuel pump feed lines so when it cranking you can see it light up, fuel pressure normal.
Seen a few glitchy starts where it actually activates the fuel pump but only for a split second, hot or cold dose not seem to be a factor.
All I can think of now is to monitor the 12 volt feed to the coils and look over all connections that feed these lines.

Dude you really, really need a fuel pressure gauge. Buy the kit from 3SX and progress from there - unless you have an existing 1/8 NPT tap you can use, in which case get one of the cheap electronic pressure sensor and gauge kits on eBay. Pretty much all the sensor kits use the same sensor design.

 

[Baba Galant]

My FPR has one attached, turbo smart. It only pressurises when cranking and drops to zero as soon as you switch off.
The only thing I haven't checked is the ISC, idle is steady but it's been a while since I've cleaned the throttle body and checked for build up.

 

[Kaldek]

My FPR has one attached, turbo smart. It only pressurises when cranking and drops to zero as soon as you switch off.

That's abnormal. Your pressure should hold for minutes, sometimes hours. Anything else is indicative of things like pump failure or regulator failure.

 

[Kaldek]

For example here is my car even after sitting a while. This was taken years ago. Note the pressure holding at 40psi:

 

[Baba Galant]

That's abnormal. Your pressure should hold for minutes, sometimes hours. Anything else is indicative of things like pump failure or regulator failure.

I am sure that only a factory FPR holds pressure after switch off. Aftermarket like Turbo Smart, Tommei etc drop to zero on switch off. I'd love to be wrong, I could just add in a one way valve on the return side.

 

[6A13TT TYPE S]

Alot of aftermarket regs dont hold pressure, you're right. And that combined with an ecu that doesnt prime till cranking leads to a bit of time on the starter.

You could wire in a short timer relay to prime with Key "on" if it really botheres you.

Ive also found that the higher flowing aeromotive regs. Are unable to hold a steady pressure with low fuel pump flow. i.e: when the intank pump is running though the low speed resistor. So when putting in a aftermarket reg I always strip the resistor and speed relay out of the loom. No only does this eliminate the fuel pressure issue at low pump speeds but it tidys up the firewall and gives you a spot to mount your reg where the stepdown relay and resistor used to be.