hotwire on a 3.5

Educated guess from your description of the symptoms and my experiences

- Throttle Position Sensor.

On or around Mon, 20 Sep 2004 21:29:17 +1200, EMB enlightened us thusly:

...

broken? maladjusted? Hmmm. I think I have a spare one, p'raps I'll try fitting it.

On or around Mon, 20 Sep 2004 21:29:17 +1200, EMB enlightened us thusly:

TPS passes ohm and volt checks, as does the AFM. doesn't mean that they're not faulty of course, could be an intermittent.

spare TPS isn't buggrit, different part number, different plug. Bet it's the same unit, too, buggrem. 's prolly off a range rover, or a different year.

Sounds vaguely familiar - my Disco used to do this. Fixing the split on the vacuum pipe made lots of difference - replacing it entirely completed the job (Warren found it split at the dizzy end).

So perhaps the advance isn't advancing for whatever reason.

On or around Mon, 20 Sep 2004 19:48:19 +0100, Tim Hobbs enlightened us thusly:

I think I looked at all the vacuum pipes before. Might stand a second look.

forgot to say that this is a petrol-specific fault, it's running fine on gas.

I gathered it was petrol specific from your original description, but oddly enough the gas was much less affected than the petrol when I had the problem. Am I right in thinking gas wants less advance at low rpm and more at high rpm?

A hairline crack in the carbon track that drops signal for a minute bit of it's operation will do this and it's bloody hard to see unless you use an old undamped analogue meter to check it - digital ones tend to respond too slowly to show the fault. Did you check the throttle closed contacts too?

On or around Tue, 21 Sep 2004 08:19:45 +1200, EMB enlightened us thusly:

ain't any so far as I know. not according to the book, anyway. The ECU is supposed to learn the throttle positions all by itself, or somesuch.

throttle pot test is in 2 halves - resistance check end-to-end (which would show up a broken track, I guess) and voltage check from the output wire -

mind, I was checking in the engine bay. I guess there could be iffy connections twixt there and the ECU, but it's a lot more hassle checking at that end :-)

On or around Mon, 20 Sep 2004 20:11:26 +0100, Tim Hobbs enlightened us thusly:

other way about. more initial advance, and not so much less as no extra dynamic at high speeds, IIRC. so rather than say 6 - 30 BTDC say, it wants more like 12-30. those are just example figures, not known to be accurate on any real engine.