MAP Sensor - Cleaning vs Replacement

addo · Post by **addo** » 12 Apr 2009, 13:20

From comments read on this Forum, I get the impression accuracy is frequently lost on these items over time, and with use.

Furthermore, it seems advocated that replacement with OEM parts is imperative. Nobody talks of cleaning them.

Yet locally (to me) people will spray them with things like carb cleaner, rather than replace.

Can any of the electronically inclined, please explain what goes on in the "life" of an MAP sensor as it becomes less reliable/accurate?

Thanks, Adam.

Post by **CitroJim** » 12 Apr 2009, 16:18

Do you not mean a MAF (Mass Air Flow) sensor as apposed to a MAP (Manifold Absolute Pressure) sensor Adam?

No worries, I'll generally describe both..

The MAP sensor, as it's name implies, senses the pressure in the manifold and normally they work on the strain gauge principle where a resistive layer deposited on a ceramic substrate bends under pressure and thus the resistance of the device changes proportionally to the pressure applied. These are normally very relaible and often as not the failure is in the pipe going to it rather than sensor itself. gentle solvent will clean them usually. Carb cleaner though, may be a bit too vicious on them.

The MAF sensor typically measures airflow by measuring the cooling effect of the air rushing past a heated platinum wire. It relies on the fact that the resistance of the wire changes with temperature and a sensitive Wheatstone bridge is used to determine the cooling effect like so:

A voltage is applied across the wire to heat it until a reference current flow through it is established before the engine starts. Then, as the engine rus, the airflow will cool it. The voltage is adjusted to again make the same current flow, thus re-balancing the bridge. The voltage applied will be proportional to the mass of air flowing across the wire. On it's own the raw mass of air flowing is not sufficient for the ECU to determine the volume of air flowing in order to set the injection pulse width correctly to achieve stoichimetric (unity lambda) so it must use the raw air mass data in conjunction with air temperature and barometric pressure to determine the volume. This was always the traditional problem with the hot-wire MAF and in the early days the "Flapper" AFM (Air Flow Meter) was popular as less maths were involved; it measure volume straight off.

The problem with the hot-wire MAF is that the wire, which is very fine, gets covered in deposits and this upsets its sensitivity and calibration. In recognition of this problem, the ECU, when the engine is stopped after a run, provided certain conditions are met, will pass a high current through the wire for a few moments to make it glow red-hot and burn off any deposits that may have gathered on it. The ECU won't do this in certain circumstances in case the engine has stopped suddenly as a result of an accident where the glowing wire may ignite a fire.

This "after-burn" will not clear all deposits that may gather and after a time, harder deposits do build up and these cannot be shifted with the "after-burn" either. It is here that gentle cleaning with a solvent can help. The MAF is a very delicate instrument though (the wire is very fine) and often they're damaged through trying to clean them.

Also, after a time, the wire, through constant heating, looses some of it's own mass and this upsets its calibration and range. In effect it slowly burns out.

There are other types of MAF sensors that work on measuring the intensity of a vortex created in a tube which again is apparantly proportional to airflow. Precisely how these work I don't know but I believe they use a ceramic pressure sensor much like a MAP sensor and thus, are presumably more robust and long-lived.

OEM ones are better quality. I've heard that some pattern MAF sensors don't work well at all, especially those intended for some Fords.

All XU petrol engines with Bosch MP injection use a MAP sensor. I believe the HDi Diesels use a MAF sensor of the hot-wire variety.

Adam, you've drained my entire knowledge now. I must rest

addo · Post by **addo** » 12 Apr 2009, 21:45

Jim, I did specifically mean the MAP type. The 405 is old enough to have the flapper type MAF, though (I've checked it for binding and cleanliness).

Getting back to the pressure type, isn't the XU10 unit just screwed to the underside of the manifold plenum? Just wondering how much risk is involved in "hosing" it out with carby cleaner.

Regards, Adam.

Post by **CitroJim** » 12 Apr 2009, 22:00

Sorry Adam

Wrong end of stick there...

Yep, generally they're bolted to the inlet manifold on the XU engines and yes, I see no reason why a dose of solvent of your chose should not be worth a go. Guaranteed it'll be a bit mucky.

I must whip mine out of the V6 plenum and see what that looks like before the engine goes back in...

What specifically are your problems Adam, in regard to MAP sensors, that is

The Flapper on the 405 can be a source of all sorts of nasties. The big spring (under the black plastic cover) weakens over time and upsets the mixture. Also, the carbon cermet tracks can wear, again upsettingt hjings and causing the output signal to be jerky. They can be retracked with care so the wipers work on a fresh arc of track.

Going back to the weak spring, the usual symptom is a rich mixture on idle and using a CO meter, it's possible to click the cogged wheel around a tooth or three to make it right although, strictly, if you do this, you really need to check the mixture all the way up, uner load, with a wide-band lambda probe. Surprising how well you can do with a gentle tweak and experience though.

Anothe thing that really upsets the mixture on the Mi16 is wear in the butterfly spindle bosses on the throttle body, again upsetting idle mixture and speed normally.

Takes me right back to my 205GTi owning days this does

addo · Post by **addo** » 12 Apr 2009, 22:26

Takes me right back to my 205GTi owning days...

Don't tell me you fitted brightly coloured coolant hoses and a body kit?

Your description of the 405 flapper internals is handy info for the future (it does run rich at idle and just turned in 23MPG

).

I'm just wanting to eliminate variables in a basic troubleshoot of the XU10/AL4 system. There's a bunch of other (related) missions such as dialling the TPS in small steps to check resistive conformity. Overall, the motor is a little sluggish and thirsty, and I don't just want to poison a new oxygen sensor if the problem lies upstream.

Also I will be dialling the cam to check it's not mistimed from when the belt was done, but that's a mission I probably won't attempt today. Don't even know if the valve cover gaskets can be re-used.

The MAP curiosity was seeded by people here extolling how their car ran so much better with a new MAP sensor from a genuine source. At least cleaning it won't bust the thing after all! I'll try thinners and a gentle blow out.

Cheers, Adam.

addo · Post by **addo** » 13 Apr 2009, 00:17

It was barely dirty:

What a fiddle to get in and out, though! I managed to "lose" a 10mm Blackhawk socket to the engine bay, so the car is a few grams heavier now.

VertVega · Post by **VertVega** » 05 May 2009, 16:11

Hi, It was interesting to read the thread. Some years ago this MAP sensor caused all kinds of nice surprises on the road

I still keep the old one (as a ref.) and I wonder if it is possible to fix it by cleaning. Did you try? Does it work?

XantiaMan · Post by **XantiaMan** » 02 Nov 2009, 11:27

Rather than start a new one i thought i would tack on to this thread...

What are the symptoms of a MAF sensor fault? Would it cause a flat spot?

Briefly, my symptoms are a flat spot at all stages of acceleration, be if light load through the rev range, medium or full throttle. Once it passes this flat spot, it goes like a rocket! I have no fault codes stored apparently. Seems to have got worse as its got colder too.

HDI · Post by **HDI** » 02 Nov 2009, 11:57

The MAF sensor failed on my HDi , when I examined it some of the connections to the sensor element itself had corroded through. I reckon this was probably due to salt spray taken into the air filter during winter.

Faulty MAF's can cause flat spots and non linear response due to the ECU filling in the gaps caused by error signals.

XantiaMan · Post by **XantiaMan** » 02 Nov 2009, 12:10

Just spoke to the expert, mine doesnt have a MAF (assume thats diesels) only MAP. FWIW, there is a tiny bit of oil inside the MAP sensor line trap, but nothing drastic. My problems could be throttle pot for example, or even Lambda. Popping to Jim's shortly to resolve

MikeT · Post by **MikeT** » 02 Nov 2009, 12:21

Typical sensor causes of too rich are the HEGO and ECT's.