The MGF Register Forums

I recently had the head refurbished on my first F, a Jan '96 MEMS 1.9 MPi, and all went back together nicely. There was so much compression in a 215k old engine that I thought the battery might be at fault. Performance was great for a couple of days then after a 20 mile motorway run the revs just stayed high which is very embarrassing in stationary traffic. I drove home with difficulty and the steady speed was uncomfortable, acceleration was lumpy and tickover a horrible 2-3k rpm; something was very wrong.
I work away a lot - typically just after it happened - and was desperately searching this and other F sites for a clue to 'over revving' while I couldn't be at home to fix it, but there was no mention!
Conventional wisdom suggests an air leak is responsible for over revving; most would suspect the workmanship on the recently replaced head, but I was there and know the greatest care was taken by a very thorough, knowledgeable F mechanic. As a first resort I used a pscan and found no fault codes, then examined the plugs, distributor, throttle butterfly and IACV, all appeared clean. When I sprayed every bit of the inlet tract with lighter gas there was no change in engine note which suggested it wasn't a leak.
It turned out to be the TPS - fitted a known good one and the tickover subsided to a disciplined 850 and acceleration is keen again. Ahhh.
So now if someone searches 'over revving' in this archive they have one clue with a happy ending!

Good result. I wonder why pscan didn't reveal the fault?

Reckless Rat wrote:I wonder why pscan didn't reveal the fault?

A sticking TPS doesn't have a fault code in MEMS 1.9 or MES 2.0J.

If a sensor puts out a reading that is in range/valid and nothing else contradicts it then wouldn't be considered faulty by the system - in this case it would consider you had the throttle operated

OK - I lied. It wasn't the TPS alone, as the problem gradually returned over the next couple of days and then I was miserably sent away just before Christmas with work. Not long back, so I started tracing the wiring today and came upon this damage where the TPS wiring splits off the main engine bay sensor loom (at an acute angle, who would have thought that was a bad idea, just like the infamous boot loom). I cut the sub loom out to straighten them: If the photo isn't clear the two wires on the left were twisted so tightly together that several strands parted and the remainder broke through their insulation and occasionally merged - with evidence of some heat generated. The TPS is just a big variable resistor, so the merging wires fed the ECU with a voltage suggesting that the throttle was being pressed, which isn't a fault.
In the same way as the boot loom can be repaired, I chopped and resoldered them with strain relieving heat shrink and stripping back the loom gathering outer and tape. So far so good - but I am reasonably confident that is the problem finally sorted! Back to beautiful car.

A quite typical fault, Charles, and probably to be seen more frequently in the future, as our cars – and their electrical intestines – get older. I had that same fault on my first MGF, you may remember my tales of woe. In fact I could rev up the engine to 2.5k by just willing the wires on the TPS. As other sensors gave erratic errors as well, I had to swap the whole engine harness.

HNY Michael! Good to see you here again - was that Lizbeth? So maybe I got lucky after so many years/miles to only have the one sensor (touch wood) misleading the ECU!
MOT for it tomorrow so we will see if the adaptive ECU has had a chance to give me the usual stress free emissions readings. I spent most of the daylight derusting the discs and servicing the brakes, which is all I ever do for MOTs (then a thorough 'Italian tune up' on the indirect route to the test station). All the best and catch up sometime,
Charles