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 - Idle high, intermittant sticky purge valve

Thread was: *Final answer to running problems*

Neil started asking about high idle problems and some more like me myself found and that they see similar syptoms at their cars.

Probably most common causes for idle high:
- sticky plastic throttle body
- *carpet crumpled under accelerator pedal*
- sticky throttle bowden cable
- broken wires to TPS (throttle position sensor)
.. could not be at mine, cause found good.

Echoing Neil from the other thread ...
or Coil, HT leads, engine management system ECU, sparks,
People have written in over the years saying they are trying to fix it with a new: -
Engine Bay Wiring Loom,
wiring to rear-crank sensor play,
MAP signal,
Coolant Temp. sensor,
throttle potential meter,
inlet manifld pressure sensor,
solenoids, and
inlet manifold vacumn pipe.

Anyway, I think Stu might suffer with the same failure like me.

Intermittant idle high after long rides.
When I release accelerator and push the clutch, the tachometer *glues* at current revs and decreases very, very slow down to app 1500revs. Standing there, even if I top and restart the engine.

The solution to temporary overcome the problem was to reset the stepper motor (ignition key on to position II, whithin 10 seconds kick the accelerator 5 times and then turn ignition key back, restart engine.)

The problem could put down to any intermittant leakage in the inlet manifold vaccum system (all pipes and related parts can be involved. Like manifold and connections to the brake servo, MEMS ECU sensor and last but not least the purge valve.

The official desription of the purge valve:
When the engine is above 75C, the purge valve solenoid will become operational (modulated ON and OFF) whenever the engine speed is above approximately 1800 rev/min.

... and...
The amount of fuel vapour which enters the cylinders can affect the overall AFR, therefore the ECM must only open the cannister purge valve when it is able to compensate by reducing fuel injector duration.

>(modulated ON and OFF)
... was found at my car with a frequency of app 3 to 5 pulse per second. Sounds like a quick steam train or a slow membrane pump when you pull of the valve block from the cannister.

The purge valve *sometimes* sticks open or closed at my car, IMO depending on how long it worked *modulated* or/and related to ambient temperatures. The warmer it is, the more often it gets sticky.
I'll spot on the failure until it happens again and I'll then measure the power to the valve to get last approval about the theory. (spare purge valve block in the bonnet of course.

It's easy to testing the valve function.
A slight *knocking* with 3 to 5 Hz frequency can clearly be felt if the valve works.
But the engine must be warmed up over 75C and revs need to be increased to 1800 by pulling the throttle cable.

Measurement reading with a digital multimeter is not the best. An analogue meter will be better. Though the power at the purge valve wires (between brown-pink(+) and yellow-orange (pulled to ground modulated by the ECU) should change from app 6 to 9 Volt (integrated) when the valve works.
I think an LED with applicable resistor will show the pulse much better.

A open sticky purge valve acts like a leak to the manifold providing *wrong air* in idle. The stepper motor (IACV) isn't designed to correct this in idle, IMO.

Not sure about Eriks *saw tooth* idle. The frequency of the tachometer going up and down at his car (to be seen at Eriks website) looks different to the switch modulation of the purge valve. Though the signal might be faked with all the connected hoses and volumes. (mechanical resistor-coil behavior leading to modulation overlay)

Dieter K.

I also mentioned this in the other thread 'Final Answer etc.'

<<Could the purge valve in the EVAP Canister be incorrectly operating and causing this strange idle? The MEMS may be faulty and cause the purge valve to operate at idle revs instead of above 1800 rpm. The increased flow of air to the mainifold would then make the revs rise. It would be easy to check by connecting a wire to the purge valve connector and run it through to the cockpit to a small light so you could see when it comes on.>>

I have had the purge valve causing funny sounds after a long run but never actually sticking to my knowledge.

Bruce Caldwell


thank you for taking attention to the matter again.

Verifying about MEMS failure OR Purge valve, yes, that's why I didn't just change the part. I'm after it and wait for the next occurance.
May be tomorrow morning when I go for the next ride to the workplace.

>to check by connecting a wire to the purge valve connector and run it through to the cockpit to a small light so you could see when it comes on.>>

Feeding wires is not my favorite ;). The arm chair in the garden is more fun currently and I need not sweat ... ;)
On the other hand this would not help me, cause I do the autobahn usually over 1800 revs ;)

Things I've checked and found until now.
- The valve works with solenoid. A small 9 Volt battery bloc is able to let it open.
- Mine was found closed when I looked the first time, even above 1800 revs (75 ). Unfortunately I didn't measure the yellow-orange wire due to lack of knowledge.
- Mine worked with ECU modulated power feed *normal* on Friday.

I'll get the second door key for the boot with me tomorrow and the meter to see what's happening finally at the yellow-orange. Will be better to let the engine run after I joined the car park.

btw, I had the option speaking to a german MGR technician here yesterday and asked him about the matter.
He said: Sticky purge valve is no common failure at the MGF but he knew from some similar failures at the R200. The valve gets blocked non operating cause of a kinda *charcoal derivates* inside.
Hmm, I saw nothing on Friday inside the entry pipe. There's a clean small 1.5mm shaft and coil spring but no derivates.

Let's see what it gets tomorrow. If it fails, I'll measure and then just change the purge valve solenoid with pipe to the manifold to the spare I have with me.

I've the idea checking the *broken* under heat (90C ?) then with help of the 9Volt battery. The kitchen oven will provide the heat ;).
Later dismanteling might be possible on the right hand side. There's kind of plastic bolt with triangular lock.

Dieter K.

Dieter (and others),

The problem did happen again on Friday evening, unfortunatly I hadn't had the time to use the multimeter.

Anyway, in my situation the purge valve shouldn't work, as the revs are well below 1800rpm. Disconnecting it had no effect either.

Anyway, I fiddled with the cables, but no effect at all (broken cables).

And no sticky revs when the problem is there.


>Anyway, in my situation the purge valve shouldn't work, as the revs are well below 1800rpm. Disconnecting it had no effect either.

Of course. It should not get powered with the modulated Voltage, but who knows. Finally you need to measure at the yellow-orange (6V is OK under 18000. 9V would be wrong).

However. I think there are lots of doubts that it is so easy and the valve is responsible in your case.

Different at mine (IMO).
Did the ride today with 300km and again NO problem.
I wonder when it gets back. Me is prepared for measures and replacing on the fly :)

BTW where is Stu ??
Would be nice if anyone could report to him when you meet at Silverstone.

Dieter K.

Dieter, sorry, I haven't been following this thread all that carefully. I guess you've checked the TPS on your Opel-sourced TB? Might it's adjustment be a little too-far out?

I know that re-setting the TPS should compensate, but might the physical position of the TPS on your adapter plate be such that the values from the sensor are too far out of range at the idle position to be correctly compensated for?

Probably just a long shot - but a thought nonetheless. Have you tried replacing your original TB?
Rob Bell

Hi Rob,

It's not the TPS at mine and approved not the TPS at Stu's or Neil's car. Broken wire was on case at the other thread. I had this problem already before I went for the GM body.

Though you are right due to checking the signals of the TPS. We DIY GM builders had only resistance values from old to new installation (comparison 1:1).
But now in any later electrical manual is written the *idle* voltage values on which the reset should reffer to. Looked up, but can't recall correct. I'll amend the site if I find the values.

The next home ride is wednesday evening and I hope it will good warmer wheather and the engine will get above 120C .....
Looking forward to meeting the *valve gremlin* then ... LOL

Dieter K.

Hi Dieter

Well when you want it to go wrong it doesnt do it !! had a long run down to stanstead airport using airport transfers tunbridge wells last week and back today in this heat and its not played up at all...hmmm

the plot thickens...

I'm sure the subject will come up at the 'heep big Pow wow' at Davids tent on saturday...

but so far i'm lost as to why its so intermitent on mine..


Ah, okay Dieter - I figured that this must have been something that you had already considered! LOL

So what, in terms of simple principles, could the cause of this problem be?

* Too much air entering the manifold at idle
* Inappropriate MEMS 'soft landing' (the holding of rpm following the shuttting of the throttle butterfly - uses to cut emissions under normal operation)
* sensor failure causing (forcing?) inappropriate fast idle for the conditions
Rob Bell

Hi all,

I give in ;). Let the *failure* be where it is !!
Again it didn't get up today. How boring..
Got the oiltemp on todays 300km home ride up to *vertial* indication (130C) over a longer distance.
The car ran fine and sometimes with 6300 revs in fifth ... ;)

I checked the idle as usual about each 50 km ... but nothing to worry.

HEF-MG96, MPI 1,8i now nearly 75k km
Dieter K.

This thread was discussed between 14/06/2003 and 18/06/2003

MG MGF Technical index

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