MG-Cars.info

Welcome to our Site for MG, Triumph and Austin-Healey Car Information.

Recommendations

Parts

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 - strange HGF

on way home from work tonight glanced at temp gauage and it was in the red, pulled over, the coolant in the bottle was had risen to almost the top. no mayo on dip stick, and no oil in the coolant. so could it be gas escaping into the coolant and causing the rise?
anyone seen this before
richb

>anyone seen this before

Yep... *some*
No reason to panic... currently.
Actions in sequence:
- check for blown radiator fan fuse
- check the radiator temp sensor (brown coloured on the top left of the engine)
- check whether the radiator fan itself is damaged
If you can't do yourself, then
- tow the car to a dealer and let them verify the above
... and get a new coolant expansion tank cap.

Good luck
Dieter
Dieter K.

thanks dieter, there is hope then !!!! i will give it a go, but how can i "- check the radiator temp sensor (brown coloured on the top left of the engine)"
richb

If the radiator fan is coming on with the temp gauge in its normal place, then this should be OK. If the fan stays on and the temperature continues to rise then you may well have a head gasket failure.
(Just start the car and let it idle for this test - you can then watch the coolant bottle, and temp gauge; and listen for the fans)

Does the heater work - do you get HOT air out of the heater or just lukewarm ?

Steve
Steve

>If the radiator fan is coming on with the temp gauge in its normal place, then this should be OK.
>If the fan stays on and the temperature continues to rise then you may well

Have a faulty theromstat (more lightly)
Have a faulty water pump

>Does the heater work - do you get HOT air out of the heater or just lukewarm ?
If normal then errm well forget that
If ROASTINGLY HOT then thermostat/fan (more lightly if going slowly, fan is only really needed when not moving) / HGF
If cool, like a sea breeze then you have a water pump problem

Will
Will Munns

richb, I had exactly the same problem - and Dieter has already covered it - the MEMS water temperature sensor was causing problems, meaning that the overheated - but only in slow moving traffic. The rest of the time the water temp was absolutely fine.

The other tell-tale sign was the idle speed - slightly on the high side - caused by that temp sensor fooling the engine management into 'thinking' that the engine was still cold, and therefore keeping the choke open (fuel economy had also deteriorated - as I noticed the improvement after replacement).

Best way to check this part is to replace it with a new one (costs 10 quid, and is pretty easy to do as found on http://www.mgf.ultimatemg.com/HGF_diagnosis.htm or http://www.mgf.4mg.com/HGF_diagnosis.htm if the UltimateMG server is down - I appear to be getting connection failures at the moment... :o( )
Rob Bell

so to sum up my plan of action
check to see if rad fan is working ok.
replace MEMS water temperature sensor
check water pump operation
get a new coolant expansion tank cap

then cross fingers !!!!!!
richb

Hi,

>replace MEMS water temperature sensor

I think another first indication on a faulty sensor is the resistance of the part.
Could easy be measured with any multifuncion instrument at the connection.

Should be about 2.8 KOhm at about 17° C
Less temperature less resistance.
http://www.mgfcar.de/sensor/index.htm
Rob ?

HTH
Dieter
Dieter K.

rich,

I'd check/replace the thermostat too. Happened to me in an old car. AA 'geezer' (and normally I am very appreciative of these guys) took one look at it and decided it was the radiator fan that was the problem, and hooked up a 'bypass' so it was running all the time.

Luckily I was 3 miles from home, so knew there was a problem when the temp went beyond 1/2 way even with the bypass running. When I felt the (cold) hose from radiator to thermostat, I knew what the problem was straight away.

Waited for the car to cool, drove 1/2 a mile to a car parts store, drove back again and fitted the thing myself.

Leigh
L Reid

run the car upto operating temp and beyond, no sign of fan cutting in so changed the fuse, with no luck , so i will go to MGR today and buy a mems temp sensor and a thermostat
richb

>>I think another first indication on a faulty sensor is the resistance of the part.<<

I tried this with a faulty sensor from my car, and another from Per's Elise. In both cases, the resistance values measured with my multimeter were within the anticipated range for cold and boiling water. I have no clear explanation for this, unless the original problem was, in fact, due to a poor connection between the sensor and the loom connector.

I could confirm this by replacing my old temp sensor in place of the new one... but I just can't bring myself to do that at present!!! ;o)
Rob Bell

I would let it be as well ;)

NTC resistors are funny behavior sometimes when broken. So the static resistance is for sure only a first indication.
Can't explain all about NTS's with my few english words, but I know how they got made.

Can someone please explain electric 'rauschen' ? is it *noise distortion* ?
C a r l ????? B r u c e ???

Not the so called 'white noise' from acoustics, but similar happens if the NTC internal bonds are bad. Together with a sensitive input at the ECU this could make false signals.

Dieter
Dieter K.

what seems strange to me is why has the coolant in the header tank risen almost to the top?
richb


Hi all,

just a follow-up on what Dieter wrote , The diff. between quality and "cheap" NTC /PTC sensors are the bonding of the wires to the ceramic sensor. When this bonding deterioate with age / humidity / vibration and all the evils car-related electronics are prone to only the very best sensors survive... A sensor can show just fine values when measured at 20 deg.C but totally flip out when re-connected in the car and under vibration from a running engine. Here the "Test-Book" readings are at their best , they measure the choosen sensor "at work" ! Unfortunatly we all know there is a price to be paid for a "Test-Book" run and even then not many of the operators know the full potential of the system...
Those of You who want to digg further into this : Unscrew the ambient engine bay temp. sensor (the one at the far upper right corner of the engine bay ). Look inside the open brass tubing and note how the sensor is mounted. Imagine the vibration at resonance of that tiny sensor that has no end support. A dab of any good glue will fixate it to the inner wall of the tubing and prolonge itīs life.

BR, Carl.
Carl Blom

Points taken guys - thanks. Maybe the resistance test should be performed with the sensor in situ and disconnected from the loom plug so as to expose the terminals? Then run the engine @ idle and monitor the resistance changes as the coolant reaches operating temperature?
Rob Bell

Rich, that is probably to air into the cooling system. Performe a bleed on all 3 points and also have a check on the "one-way valve" in the head / inlet mainfold. Be prepaired to fill up during the bleeding process. Plain clean water will do for the moment until there is a total flushing and re-newal of the cooling fluid.
BR, Carl.
Carl Blom

Are HGF's commom to all "F" or are they on only older cars? Has this been fixed on say a new TF or can I expect problems around the 45K mark.

Still thinking of buying a new 115

Bill
W Sutton

>what seems strange to me is why has the coolant in the header tank risen almost to the top?

The expansion tank cap is simply broken. Replace as recommended.

Rob,
see what Carl wrote regarding bonding of connections to the ceramic body.
In some cases you will not be able to measure only with a resistance meter. It is to slow. The signal should be spotted by use of an oszilloskop.
The 'noise' from a bad bond connection will probabbly be only high frequency.
IMO put any resistor 100kOhm ? in line with the sensor. Then a load from a simple 1.5 Volt battery to the in sequence resistors.
Connect the Oscilloscop to the sensor and see the signal in high resolution to see the high frequency distortions / *noise*.
Anyway, that only makes sense if the MEMS input is sensitive to such HF signals.
If someone could point me on the related input connection to MEMS, then I can look for the input circuit in Lucs broken MEMS ECU. (R-C combination ... hopefully)

Dieter
Dieter K.

Dieter wrote
<<Can someone please explain electric 'rauschen' ? is it *noise distortion* ? >>

As far as I can tell it means electrical noise. This is difficult to see with a normal resistance meter because the changes are too quick. Usually the result of a dry joint (solder) or an oxidized connection.

HTH Dieter

Bruce
Bruce Caldwell

This thread was discussed between 30/09/2002 and 05/10/2002

MG MGF Technical index

This thread is from the archives. Join the live MG MGF Technical BBS now