Nominal Afm Measurement

[mhyphenl 10 1 Cars]

Can anyone with a 1.9 DKZ engine and AFM 0280202202 do a measurement of voltage for me!? I'm trying to correct a tampered with AFM and I'd like to know what voltage across terminals 2 and 4 is at tickover with engine running. revs at 950? I know it's a pain but it's the only way I can see of getting the correct setting. Obviously on a car that is running well with no faults. I'm hoping this method bypasses any issues with air leaks etc!

[Telf 222 2 Cars]

I'm not sure you can fix it in this manner- the spring is calibrated - if its been messed with its almost impossible to correct (at least that's my experience/understanding).

 

Also the variables involved make a measurement between different cars highly unlikely to mean much.

 

I would look for another AFM- at least for diagnostic purposes.

[mhyphenl 10 1 Cars]

Surely the AFM is only calibrated to its own part number, ie all 0280 202 202 AFMs will be the same!? They can't calibrate the afm to the actual car, if so getting another wouldn't work either. I'm pretty sure that, amongst other things the ECU is looking at the voltage across pins 2 and 4 and calculating the the injector pulse width.

If I send this AFM away to be refurbished they set the spring tension to give a certain reading given a certain pressure on the air flap. Thing is this AFM is new and the garage that did my MOT fiddled with it to get it through. I don't really want to pay for the whole thing to be refurbished when it only needs the spring tension setting.

I agree that getting readings from someone elses AFM is taking for granted that there's is in good working order and not molested in any way! But thats the same as buying a used one! The good thing about mine is that it's only done 300 miles and so the resistance tracks are pretty much new.

 

I still think getting a reading from a good one will get me close, at the moment it's miles off and was very rich. I've spun it round a bit but don't want to go too lean!

[Telf 222 2 Cars]

I have a host of AFMs in the shed. When you measure the resistance output from the wiper at min or maximum deflection- indeed with the wiper held in a 'identical' position the readings are slightly different on them all.

 

Its this spring calibration that as you say is most important.

 

If your garage has messed around with it I would approach them to get it fixed. I cannot think of a single reason they would need to be messing around with the AFM (other than the CO2 screw).

 

I believe there are a few links on the forum detailing spring setup for the AFMs.

[Telf 222 2 Cars]

From one of the articles:

 

 

http://www.205gtidrivers.com/articles/e3-injection.html

 

 

AFM testing (notice that there are different AFMs for Motronic and Jetronic systems):

- For Jetronic: Disconnect the 5-pin connector from the AFM, and measure the resistance on the unit between pins 5 and 7. It should be between 60 ohms (flap closed) and about 1 KOhm (fully open), with a smooth transition. Check the resistance between 8 and 9, it should be between 160 and 300 Ohms. Check the resistance between 5 and 8, it should be between 350 and 450 Ohms.

- For Motronic: Disconnect the 5-pin connector from the AFM, and measure the resistance between pins 2 and 4. Like the Jetronic type, it should vary with the movement of the flap in AFM. Resistances between 2 & 3, and 3 & 4 should be around 300 Ohms both. Check the resistance between 4 and 5, it should be around 2-3 KOhms at 20 Celsius.

- Pinout of the AFM connectors (pictures show the unit side):

 

Jetronic Motronic 5 Earth 1 not connected 7 Flap signal 2 flap signal E Not connected 3 +5V supply 8 ? 4 sensor return/ground 9 Power supply 5 air temp. signal

 

 

Hope that helps

Edited June 27, 2016 by Telf

[mhyphenl 10 1 Cars]

Thanks for the link telf, I've done all the tests and I'm confident mine's in good working order! definately AFMs with different part numbers will be different by definition, the only difference between them all is the spring tension and voltage given and a set volume of air. Oh and whether they have a CO screw or not! If you've got 2 with the same part number they'll be very similar although position of the wiper is not the key, it's the measurement of the voltage with a given volume of air flow in cubic feet per minute, (or any other unit). That's why measurung the voltage on a good car at given revs would be a fairly accurate way of calibrating one.

 

Absolutely, messing with it was a bad idea, in fairness to them they asked my permission because I needed it driveable. There is no CO2 screw on the 0280 202 202 AFM, it relies solely on the Lambda to adjust the mixture, (actually the screw is there but not connected). That's why it's causing me so much bother.

While I agree that there will be slight variations between cars they will be very close at a given revs, (revs translating directly to air volume measured, particularly on these engines as there is no unmeasured bypass.)

 

There may well be other unresolved issues which is why I don't simply want to stick a gas meter on it and re-adjust, I need to try and set it to what a well running DKZ engine has it then progress from there.

[Telf 222 2 Cars]

What is actually wrong with the car?

 

If your having issues with the idle I would suggest looking at every air/oil pipe. I've had problems with mine, initial examinations all appears well but when I actually had a good look at the pipes I found perished areas, vlucanised/porous rubber (especially on bends) and tears (in the vicinity of jubilee clips)

 

Any mating surface that might let in air needs looking at- the Haynes manual has a guide to likely areas.

[mhyphenl 10 1 Cars]

Everything!

 

The idle is quite lumpy and its running rich, it's putting the error light on because it can't bring the emissions under control with the lambda!

[mhyphenl 10 1 Cars]

Ok, done a bit this morning.

 

Idles like sh1t on cold start, very lumpy. push the flap open on the afm slightly and it smooths out. That means lean right? Likely air leak?

 

Then after a minute or so (lambda kicks in) the revs rise and it smooths out (on it's own, not toughing the afm) Which suggests the Lamda is signalling lean and richening the mixture. at this point the engine light comes on, possibly because it can't make it rich enough.

 

Does this sequence of events and my thinking make any sense to anyone???

[Telf 222 2 Cars]

I cant really comment on the lambda side of things- I've no experience other than I think I may have a sensor in my shed!

 

when you say lumpy do you mean hunting up and down the revs?

 

Hunting is often a mixture issue. Air leak it usually just runs at too high revs (or too low if you have set it up whilst the SAD is in operation.)

 

Have you checked the actual lambda sensor wiring to ensure its serviceable?

[ALEX 98 1 Cars]

Nothing wrong with adjusting the tension of the AFM spring to get the car running correctly. sometimes its a requirement to get it through the MOT.

Though its always beneficial to mark anything adjustable first, so at least you can put it back.

I've had one of these apart as Intended to have the casing vapour blasted and noticed the bearings had seized. this for sure would affect the spring calibration on the flap.

In the end I dropped on one on ebay brand new, so they are still available.

If your not doing a full restoration but buy a brand new AFM then keep it sealed for when/if you do a full restoration. It'll be better not try and calibrate the AFM spring inside to compensate for other worn components, use an old 2nd hand AFM to mess with.

[2-Pugs 57]

Martin, come to PugFest next weekend and I'll hook up my code reader to it. It might help?

Rob

[mhyphenl 10 1 Cars]

Martin, come to PugFest next weekend and I'll hook up my code reader to it. It might help?

Rob

 

Hi mate,

 

that's really kind of you, would be nice to catch up! I'll try and wangle Saturday off work. In terms of the code reader I built myself one with a proper plug and everything, just a switched led in a posh box really.

[mhyphenl 10 1 Cars]

Wow, ok we might be getting there. I received a known good proper Bosch AFM from the states today! Fitted it and drove to work! What a difference! Smooth acceleration, quieter engine!! Light still comes on though I've replaced the temp sensor with a new one which i tested before i put it in!

It's going in to the garage to have a new Lambda fitted and the exhaust checked for leaks up to the manifold gaskets! Hopefully that will be the end of it!

[toolie72 67 1 Cars]

Wish you all the best-might be worth checking the values etc to see what the difference really is

[mhyphenl 10 1 Cars]

Yes my thoughts exactly. Would be interesting to set the other one exactly the same and see if it works well. I have my suspicions that the Pattern ones are just not good enough. All sorts of things like the mass of the flap, the level of air damping provided by the chamber, quality of the resistive track are all going to effect values across the whole rev range. There are only about five or six cog teeth that it can be adjusted to and I've tried them all, none of them performed very well si I basically think the cheap ones are pants!

[toolie72 67 1 Cars]

Hmmm you might be right-I still find it amazing that pattern parts can be so bad, why bother tooling up to make something then make it out of tolerance?

[mhyphenl 10 1 Cars]

Totally agree, although you have to think, is a real Bosch AFM £500 when a pattern one is about £130 just because Bosch are a bunch of robbers or because thesres a significant difference in the quality of the components. Bit of both probably. I'm lucky I found a really good one from the US when I did!

[welshpug 1,621]

a remanufactured unit is about 150 from atp electronics.

[mhyphenl 10 1 Cars]

£150 for a bosch unit reman would be a good price, anyway we're getting even deeper now.

 

So had a hole fixed in the exhaust and a new lambda fitted at the weekend and guess what..... no difference!! So that's new coolant temp sensor, known good (and tested) AFM, new lambda and still the same fault.

 

Now I had the garage do a quick check of the gasses and concluded rich running. I decided to check the output of the lambda to see why it wasn't adjusting the mixture to lean it out, turns out the lambda is reading about 0.2 volts. Now I know that your 0.1V is a lean reading and 0.9V is rich and it should cycle between each end of the spectrum, I'm also aware that on a normal digital voltmeter it may not refresh quickly enough to show the swings but should average out to 0.5V ish. Question is, why is a new lambda (the other had only done 2000 miles) reading lean when it's actually rich and pushing it richer?? I've checked continuity on the lambda from the ecu and there's about 0.3 ohms along both wires. I've also read the running voltage accross the pins both at the lambda and the ecu and it's the same.

[DamirGTI 323]

Bad lambda sensor wiring/no power to the sensor heater element maybe .. check the heater wiring and check if the sensor heats up .

 

D

[mhyphenl 10 1 Cars]

Bad lambda sensor wiring/no power to the sensor heater element maybe .. check the heater wiring and check if the sensor heats up .

 

D

 

Hi yes my thoughts too. Had a couple of hours this morning to do more tests and it gets stranger:

 

Completely renewed both loom connection for the lamda as the pins sometimes pushed in. New connectors and pins, crimped well with the proper tool! There is Battery voltage to the heater connector and 0.45 volts across the sensor pin which is what it should be with no lambda present. I haven't connected battery voltage to the lambda element directly to see if it heats, it is a new lambda though??

 

Ok so when engine starts there is 0.15v across the lambda which suggests lean and I am on the hunt for a small air leak. When she warms up and the ECU kicks in it tries to richen the mixture, the revs increase and smooth out but the voltage only goes up to 0.26v. That made me think that maybe the leak is bad enough that it can't richen the mixture enough to flip the lambda. It's suggested on a Lamda web site to release butane / propane into the air inlet to richen the mixture and test the lambdas response, it made no difference. The other suggestion is to move the flap on the afm more open with a screw driver to richen the mixture, again no response on the lambda, just sits at 0.26v. If I rev the engine in drops back to 0.15v then climbs back up to 0.26. When pushing the flap on the AFM I could smell it was rich so it must be an issue with the lambda response.

 

I've tried 3 differend ECU's and this is the second lambda. Continuity on the lamda signal is good, only 0.3 ohms across the length of the wire. As far as I know the cable for the lambda signal goes directly from the lambda to the ecu. the grounded shield on the lambda signal is earthed at the ecu end and not at the lambda end. Is it possible to have battery voltage at the heater but it not be able to draw enough current to sufficiently heat the element??

 

I'm glad I've found this issue because it's almost certainly the major reason I'm getting this issue, if I can get the correct lambda signal to the ecu I'm sure it will be able to govern itself.

[welshpug 1,621]

have you checked the alternator output?

[mhyphenl 10 1 Cars]

have you checked the alternator output?

 

In a word no! I'm not having an issue with battery charge, could this effect the lambda in any way? Interestingly that little battery light on the dash often comes on until I drive away!?