Welcome to our resource for MG Car Information.
MG MGF Technical - Fuel pump out of MEMS ?
|I have to replace the MEMS 1.9 in my F. I bought a second-hand replacement and sent it to the garage -distant from home- together with the immobilizer and the bleeper in order to be TestBooked and codified which they did. The car engine in which they placed it started fine however the fuel is cut soon after ignition. In order to not wait for another replacement the garage people suggest to short-circuit the pump and connect it directly. I would like to get advise about feasibility and if possible a sketch on how to get it done. Thanks.|
|If imobalised the fuel pump will not run at all (so the programming has worked), de-immobilised the fuel pump should run for about 3 secs with the ignition on but the engine not running, then it should restart and run continiously whilst the engine is running, if you stall it should stop. |
So if the engine is not running, this is correct behaviour, if the engine is stalling for another reason, then this is correct behaviour, but if overriding the fuelpump relay leads to continiouation of the engine running then there is another fault.
The MEMS uses the crank signal, but this is also used for timing and metering, and without this it wouldn't work at all.
Note the safty aspects of a fuelpump wich cuts off if the engine stops vs the fuel being pumped around a hot engine bay after a crash.
Also note I once had a fuel hose 'disconnect' itself whilst I was warming the car (I was nowhere near) your bypass system would have continued to pump 30 litres of petrol all over the engine.
If you stick with the bypass _MAKE CERTAIN_ that the shock sensor is kept in the fuel pump circuit!!
|Mine only seems to run for about a second Will..so this time varies depending on the fuel already in the fuel line I assume..|
Once its primed it stops and waits for the engine to start..
|No way to tell the amount of fuel in the line, but the manual says 'max three seconds' IIRC, I suppose they might have shortened it for the F as the fuel tank is real close to the engine.|
|There aren't any ECU pressure sensors in the fuel rail; fuel pressure is regulated by a mechanical valve device that returns fuel back to the petrol tank when pressure exceeds about 3bar.|
So something that I've wondered about: why does the fuel pump whirr for what sounds like different durations? Is there a current load sensor across the fuel pump? Or is it just subjective imaginings?
|The fuel pump doesn't whir if the ECU thinks the rail is already pressureised (try ign on, off, on, it doesn't run the second time). I suppose that it's possible that a decay period is built into te ECU, this sounds very over the top to do digitally (why bother!), but if the timing is set by chargeing a cap then it's possible.|
|>> I suppose that it's possible that a decay period is built into te ECU, this sounds very over the top to do digitally (why bother!) <<|
This seems likely - afterall, the MEMS already has a delayed power down for both the engine bay fan and the purge valve.
|@ Mike, I spot on the thread, as promised by email.|
Sounds difficult to giving final advise. :(
I think the question is, can you or the garage put down the failure to the ECU or anything different.
However, I get Luc Vael's problems with engine cut out problems in my mind.
At last it was the ECU (see Archives from 2001).
However, will direct power to the fuel pump hurt anything or not ?
(notice Wills advise on the security circuit with shock sensor plus relais to cut out the pump in emergency case)
|I truly appreciate your opinions Mates. Now, here is the actual situation with my F. The engine does start and revs fine but a few seconds after the fuel pump stops pumping. Apparently the MEMS -or something else-cuts the pump. I really donīt know what else can be done at this stage. If connecting the pump directly is not feasible -or may be dangerous- I will have to wait for weeks to get from the UK a third MEMS, immobilizer, codify them, and whatever more is required to the process. |
Any good idea will be most appreciated. Mike
|If the engine is running OK if you feed the pump directly, I would get a new MEMS - or an EMERALD (that way you could DIY the coding). If the fuel pump circuitry is duff, then God only knows what else is knackered in the MEMS, I wouldn't take the chance - if the fueling is off as well you could dammage the engine!|
Where did you source the MEMS from?
Final test before you quit this unit is to check the voltage on the wire that feeds the fuel pump relay, there is an outside chance that the relay is coked up and getting hot and so stopping the voltage after a few seconds - if this is the case an external relay is easy to wire in.
|I'm beginning to wonder whether the fuel pump itself is in good health or not? I guess 'hotwiring' the pump power will give a clearer indication of this: if the pump still powers down despite a known power supply, then you have your answer: it'll either be the relay or the pump itself.|
Thinking about it as I was typing, you should be able to tell if the MEMS is powering the pump down prematurely: put a lamp or some other 'reporter' in the pump relay circuit. When the pump switches off, is it because the relay has ceased its 'pump on' signal?
|Clearly way ahead of you Rob ;-)|
Black wire with purple stripe runs from the MEMS to the relay pack, connect a instrument light (very low wattage) between this and a +ve source.
|Thank you Will, Rob, Dieter. MEMS was sourced by a fine Centre in Wolverhampton. I let them know that it may be damaged and have offered to send a new one. No problem from that side. A very fine lady owner indeed.|
My real problem is time and distance and programming again as my poor F has been stalled for three month now -and I am desperate-. I quite donīt follow your indications -technically very ignorant-. Could I somehow wire the pump directly to see if that is the problem? If so, wire from were? Mike
|you could bridge from the thick brown with silver stripe wire to the white with purple stripe wire that go into the black wedge shaped thing which is the relay pack - This would give you the pump and the inertia switch, but the pump will continue to pump with the engine switched off (not a problem if you're only testing)|
|>>you could bridge from the thick brown with silver stripe wire to the white with purple stripe wire that go into the black wedge shaped thing which is the relay pack - This would give you the pump and the inertia switch, but the pump will continue to pump with the engine switched off (not a problem if you're only testing>> A bit of worries about, Will. Could be possible perhaps to wire the pump somewhere else that allows it to stop if the engine stops? |
|>Could be possible perhaps to wire the pump somewhere |
>else that allows it to stop if the engine stops?
you could wire it to the ignition - but thats not the same thing. This advice is for testing only!
Unless you want play with relays the choice is inertia switch and manual switch off or igntition switch. Ignition switch is convienient, inertia switch is safe(er) in an accident.
This thread was discussed between 28/04/2004 and 07/05/2004
MG MGF Technical index
This thread is from the archive. The Live MG MGF Technical BBS is active now.