Welcome to our resource for MG Car Information.



MG parts spares and accessories are available for MG T Series (TA, MG TB, MG TC, MG TD, MG TF), Magnette, MGA, Twin cam, MGB, MGBGT, MGC, MGC GT, MG Midget, Sprite and other MG models from British car spares company LBCarCo.

MG MGF Technical - MEMS ECU Connector / idle high and sensor troubles

Apology, just saw Steves comment to the other thread and hijacked the subject.

From: steve ireland on 03 September 2004 at 11:35:10 (UK time)
Latest news is that it is one week since the guage moved and has not budged from where it normally resides. No water usage above normal ie 100ml-200ml / 6000 miles and no waterin the oil.

The only plausible reason I have for it apart from imminent HGF is that the large black plug ( very technical ! ) for the mems was attaacked by myself one evening at the side of the road. The reason for this was that other irritating fault of over idle. Now when I say I attacked it I mean exactly that. Guess what, for the first time in 8 months I have had perfect idle for I guess 3-4 weeks. So I may of upset something else along the way but I doubt it as I would imagine that the guage works straight from the sensor and not via the mems. Can anyone confirm ?
From: John McFeely Gloucester U.K. on 03 September 2004 at 14:09:57 (UK time)
Love it! He who 'attacks' his multiplug thus rectifying his idle problems is no mug! ...LOLOLOL

From: steve ireland on 03 September 2004 at 17:25:37 (UK time)
>>>Love it! He who 'attacks' his multiplug thus rectifying his idle problems is no mug! ...LOLOLOL

I lived with this problem on and off mostly on as I said for 8 months.

No amount of tender care shown to the named plug would make a permanent fix. By that I mean crimping the connections tighter together and using most of a tin of kindly donated contact cleaner. Thankyou work !

Anyway one or two tempremental bends of the cable has done the trick. I recommend it to anyone. The joy of not surging on after taking the foot of the throttle at 70 mph is worth the risk of wrecking it. Oh and the obligatory 2000rpm revs at start up on a cold winters morning as well. Petrol as well, thought I was driving a Volvo for a good while.

Incidently about a year ago whilst being unfortuanate enough to be at a Rover dealer I noticed a tech with a new liner in his hand. Upon asking what it was for the reply was it was for a Rover 200 which had suffered bore wash. Why, because the bloody auto choke wouldn't switch off. Wonder why.

Oh and I will ask again is ther a way of forcing the rad fan on ?

From: Dieter K. Hersfeld Hesse on 03 September 2004 at 20:19:03 (UK time)

I can confirm !!
The same is with my MEMS ECU connector.

However, let's talk about in a new thread.

<MEMS ECU Connector / idle high and sensor troubles>

Yep, this one.

Steve asked for. Lots of are in the archives MGF ongoing intermittent idle problems.
Ended up once in not starting engine:
I found the harness loose below the connector (cable tie not fitted to ECU bracket)
Realigned the harness, problem solved, .... almost.
Sometimes the idle high appears again and my usual action is pulling to the road side, shut down ignition, open boot and rattle at the wires.
Then reset TPS and it's cured. Idle normal.

Some more facts.

The connection diagram.
Notice the wire locations of the thin sensor wires at the ECU connector
Sensor wires are thiner then the other wires. Bad crimps possible ?

ECU diagrams

And the simple of sensor circuit if you simply put them all together from both above on one sheet

The idea is that a bad connection from bent receptable contacts or poor crimp could cause high idle.
See wire i.e. the earth wire #30.

Tony has some details on receptables probably used from AMP TYCO here.

comments :)
Dieter K.

Well there is not much else that I can say to all that has been said.

The only thing noticeable was the fact that I gave the last 6 or so inches of the cable a different posture than before. Whether this gave a more secure connection at the actual plug is debatable, or it may be a brake in the cable itself. I would doubt the later is true as on previous attempts just moving the plug itself had an impact on the idle speed.

All in all the plug and socket arrangement is very poor. It should have securing screws especially as it is subject to a fair bit of vibration.

Just drove in again and it's still okay. Believe me I will not touch that thing again unless it is absolutley necessary like the last time, a new engine !

BTW Dieter. If you do decide to buy a new harness 40 pounds from mgf center ! I was quoted 100 from others.

It would be worth it to get rid of the extremley annoying high idle not to mention the risk of bore wash with a rich mixture.


>I gave the last 6 or so inches of the cable a different posture than before.
Did the same only, and found the mentionend cable tie not fitted to the hole in the ECU bracket.
Refitted, and this worked for me over 6 or so month.

Don't think i's a break in cable.
The crimp reliability for the thin sensor wires is another question.
If done to strong, inner thin copper wires break, if done to less, then bad connection.
There are international standards for harness wires to be used and I daubt they selected the wires from highest standard ;)

> If you do decide to buy a new harness 40 pounds from mgf center ! I

Good point with changing the harness, I've a spare myself from a 97 wreck _anywhere_.
I think Tom Randall (K&N bracket master) got this done and his MGF problems got cured. Anyway, swapping in the spare is no fun for DIY.
Dieter K.

>>>> swapping in the spare is no fun for DIY

Too right, unless engine is out.

>>>>>There are international standards for harness wires to be used and I daubt they selected the wires from highest standard

Right again. I am used to working with what they call super flex cables at work and there is absolutley no doubt that the quality of wire in the harness is something that needs to be improved.

Hi All
Amazing coincidence that this topic should come up right now. As I've posted before, I am having troubles with my car revving too high (2000 rpm) following a change over from a standard TB to a GM TB.
What I forgot to mention to you Dieter, is that while digging around, I decided to spray some connector cleaner on the large mems plug. Is it possible that I may have replaced it incorrectly / bent something?
I have checked the TPS, which is working correctly, and cannot find any air leaks. One additional characteristic of my problem is that tick over is at 2000 rpm initially, however this drops to 1500 after a little rev. to 3000 or so. Does anyone else recognise this?
Dieter, I shall be poking around the electronics tonight with a multimeter, looking for poor connections etc., is there a chance I could call you for a bit of guidance if I get stuck? If this is ok, you can get me on: jim dot farrington at proseis dot com to let me know the most convenient time / tel. no to get you on.

All the best

J. A. Farrington

This thread was discussed between 03/09/2004 and 06/09/2004

MG MGF Technical index

This thread is from the archive. The Live MG MGF Technical BBS is active now.