Fuel pump priming continuously

[jigarai]

When I switch the ignition on the fuel pump normally primes for a couple of seconds then stops. Today I noticed it continued priming indefinitely.

It is gti swapped but on a digifant ecu (with an Elvis loom) so starting point has many similarities to stock efi setup.

My first query is looking for an explanation of, or signpost to the mechanism for the fuel pump priming to switch off - does it aim to reach a certain pressure? Is it on a timer chip or similar? Relay issue?

[cobblers]

Engine ECU controls the fuel pump relay, initially when the key is turned to ignition it will prime for a few seconds - just a timer, there's no feedback from pressure etc.

Whenever the ECU detects the engine is rotating from the crank/dizzy sensor, it will enable the pump again.

[silverbullet]

I have a - presumably faulty - 191...numbered TCI that makes the fuel pump priming cycle every few seconds, rather than the single 1-2 second burst that is usual.
It also gets hot very quickly, which makes me think that it has an internal short that is sending a false signal to the ecu.

[cobblers]

I have a - presumably faulty - 191...numbered TCI that makes the fuel pump priming cycle every few seconds, rather than the single 1-2 second burst that is usual.
It also gets hot very quickly, which makes me think that it has an internal short that is sending a false signal to the ecu.

Rather than waiting for an accurate "lock" on of the engine RPM which needs quite a few pulses to calculate, the ECU is very keen to trigger the fuel pump and will do so as soon as it sees any change of state on any timing signal.
If a TCI is iffy backfeeding a bit of noise down the signal wire, this could happen.

[jigarai]

thanks for the replies folks - I've been on a bit of a journey with this as it went down hill quite quickly and ended up with me 500 miles from home and rigging up a chicken switch to turn the fuel pump on and off. 

I've had a good look at the wiring diagrams in the Bentley and for the golf the ABF came from and worked out where it all crosses over.

Van is actually with Evil Ben's in Chacewater to have a panel replaced now but when it is back my intention is to check the power into the trigger side (86) of the relay which comes from the switched live via the coil and then if that is good, I'll replace the wire to the ECU (85 on relay to T68/6 on the ABF set up Digifant. I was interested to work out that the ECU makes a circuit to trigger the relay by earthing it. 

I'll triple check with another spare relay before I start pulling wiring looms apart though!

I'll report in again when I've had time to do a bit more work on it. 

[jigarai]

I have a - presumably faulty - 191...numbered TCI that makes the fuel pump priming cycle every few seconds, rather than the single 1-2 second burst that is usual.
It also gets hot very quickly, which makes me think that it has an internal short that is sending a false signal to the ecu.

From what I understand the ABF doesn't use a TCI module that is external to the ECU, it's a digifant system from around 1996. (Was the 7 pin TCI on the Digijet?) - I guess the same process of isolating inputs and checking they are all as expected and checking Earth pathway is good too, before then spending out on a new module applies here also...