Hey everyone,




The E30 Air Flow Meter is shown above... The copper arm rotates on a shaft connected to a flap, as the vehicle pulls in more air the arm rotates and moves along the voltage divider network.  All of the wires from this sensor go to the car's ECU.  What's interesting about this is the resistance across pins 7 and 26 aren't linear as you rotate but the voltage output is.... Still don't quite get this, obviuosly it works but i wonder why it isn't a normal voltage divider?


I was also interested in the current draw on the +5V source and output voltage wires.  So i utilized a current measuring tool and took measurements on two cars:




What I thought was interesting here is I expected the output signal (pin 7) to have very little current on it as I assumed it was going straight to the ECU's ADC.  The other odd thing is the large measurement difference here, used the same gauge, cars have the same ECU and AFM part #'s.  

I decided to dig in and see what i could find on the ECU, picture below is the car's wiring diagram (left side) vs the Motronic 1.3 schematic.  Note that I believe on the schematic they have the two names flipped... Reason I believe this to be correct are two fold.  1.  The pin #'s are correct otherwise and 2.  I assume since the temp sensor (pin 44) is a NTC thermistor it would have a voltage divider as you see on the pin 44.



I'm not seeing anything on here that sticks out on why i'm seeing so much current on this voltage output (pin 7).  Any ideas?

Hi vettett,

this is a very interesting project which may well be useful to brick pilots in the future as the wipers on  these superb instruments wear out beyond repair.

Firstly read my write up in again in detail . I think I covered reason for the non linearity: air flow is not linear either. The potentiometer is roughly logarithmic  to compensate for the exponential relationship of 'barn door' flap position with air flow volume. If your mag device is giving a linear reading w.r.t. angle you may need to compensate in software.

Take care to look at the extra resistances which are added on either end of the printed resistance chain. One is in parallel with the air temp thermistor.

Two cars different: without knowing the difference in the engines, let's just guess that the airflow, idle, wear or setting of  zero point are not the same. 

Bear in mind that the original Bosch unit is an analogue computer, ( well hybrid analogue/digital at least ). Those inputs are part of an op amp circuit , not just an ADC.  This is probably a key understanding that you are missing in looking at this circuit.

IIRC, I linked to a circuit of the electronics , so I suggest you look at that and try to see the implications.

I think what are doing is perfectly attainable but is probably a bit more subtle than you initially thought.

Check out the points I have raised and post back if you need clarification.

Beamer, thanks for the reply.  I'll have to go back but i believe at some point Bosch switched it up such that the AFM put out a linear relationship between flap angle and output voltage and then they let the ECU do the hard part, that is what my measurements of the OEM AFM show... It's worth checking again but either way that is easily handled in my code.

I have a work around for the increased current needed, don't believe it will be a problem, just want to dig into the original design some more to make sure i'm not missing anything.  As you can see i'm doing a lot of this in parallel (learning how the OEM works and making a new one), not ideal but it allows me to get things moving faster initially.  Appreciate the response, i'll go back through the documentation and let you know if i have any other questions.

Have a look at my experimental set up for measuring the voltage vs angle relationship. If nothing else you probably need to know this for your particular unit in order to programme it correctly.What I've called the dead zone is quite important to get right and the thermistor is non linear, and this is probably taken care of in the wiring of the resistor chain too.

If you put an NTC into a potential divider, you can make it linear over quite a wide range.

The laser cut resistors on the ceramic are a precise ladder of fixed points , the carbon track is effectively for interpolation between these reference points.

Some quite subtle engineering has gone into this unit. Again, think of it as an analogue computer and google that term if you think all computers are digital devices

No reason that this could not all be done digitally but you'll need to thoroughly understand how it was originally designed and what the physical relationships modelled in the electronics were, in order to correctly reproduce it.

There will be different strength springs on a car. AFAIK, all the K bikes use the same spring strength.

What is the magnetic device you are using to replace the carbon track, that could be useful.

I will devise a test setup that either controls the air flow or at the very least measured the force while rotating the flap.  In the short term I have plates of different designs that open the door at set intervals so i can at least match the OEM sensor from a flap opening vs voltage output standpoint.

Went back through your thread, some points/questions:

Your comment "It's very likely driven with a 5V regulated supply. The total resistor chain is a little over 510 ohms, drawing 10mA. "


I don't think this is true for these devices, V=I*R, seems more complication than that.  Point in case, here are some measurements I did on my desk, feeding the AFM a regulated supply. As you can see t he amperage is not anywhere near what one would calculate.



I see the schematic you made up now, it is interesting.. I should try and replicate it for mine for a fun comparison.  Although, I thought you really couldn't accurately measure a resistor that is in a circuit, you would want to isolate that resistor first, no?

Another interesting thing is when we "repair" the AFM (which I also do) we both determined that drilling the holes in the ceramic/metal board was the right way to go. 

When you were taking voltage measurements, were you doing so with a volt meter or an oscilliscope?  I found that using the oscilliscope is nice because I can quickly open the flap and see where this is a tendancy for the voltage to drop very quick which can be hard/impossible to find with a hand held volt meter.

Your comment regarding the Bosch computer being analog/digital... I don't see an Op Amp in the schematic i have for the ECU, it seems to clearly show the output of the AFM going into a microcontroller ADC port (i'm talking about the E30 Motronic 1.3).


With regards to the current measurements taken on the car with the stock units.  I don't have a good explaination for that, I wasn't there when the measurements were taken but I trust the folks who did it.  When I get back, I'll confirm that the AFM and ECU were the same... perhaps even switch them between cars.  If they are the same, that current different is huge to me (5x).

I have just realised that you started a new thread on this, rather than posting a comment on the flow meter restoration thread. It would be useful to at least have a link to what we are discussing. There was the initial thread I did as I was discovering how it all worked and another where I cleaned it all up into a tidier summary.  This is the summary which you should refer to , in case you are looking at the initial thread.This was pointed out in big red letters at the bottom of the last post in the initial thread.

https://www.k100-forum.com/t11055-bosch-air-flow-meter-restoration-summary

Your comment "It's very likely driven with a 5V regulated supply.

It seems you are referring to the earlier thread. I suggest you read the summary write up and ask questions about what I wrote once I had sussed the whole thing out. Please indicate where you are quoting from if it is not from the summary article ( eg with a link to a specific forum comment). That saves me having to spend time searching for the text you cite.

Circuit of the resistor chain.
https://servimg.com/view/19452982/32

The source of that cct. is linked my article. I suggest you verify the laser trimmed resistors for you self, this will help you program the relationship between position readout from the A1335 and the voltage you feed to the ECU.

I see the schematic you made up now, it is interesting.. I should try and replicate it for mine for a fun comparison.  Although, I thought you really couldn't accurately measure a resistor that is in a circuit, you would want to isolate that resistor first, no?

That is true as a generality but in this case ( with the AFM in isolation , not connected ) we know it is a simple resistor chain with no external source of current so we can measure in situ.  I used a DVM for accuracy on my test measurements. You will see photos of the test setup in the article. You will notice that I am not measuring resistance directly, I am measuring the output voltage at the external connector under power conditions similar to running on a bike. Resistance measurements jump all over the place and what is relevant is the output voltage of the AFM.

Your comment regarding the Bosch computer being analog/digital... I don't see an Op Amp in the schematic i have for the ECU, it seems to clearly show the output of the AFM going into a microcontroller ADC port (i'm talking about the E30 Motronic 1.3).

Obviously my comments relate to the electronics on the K series BMW motorbikes. If you are using this AFM  with different electronics you will need to take that into account. Sorry, I had not seen the detail of the Motoronic unit you provided. This is different to the units on K bikes.

What is the provenance of that circuit? Some of it does not make sense to me.

I'm currently using an Allegro A1335 for my sensor,

Thanks, that's interesting. A magnetic pickup should be more robust over time than these finely made carbon track pots. Though they do seem good for about 20y of service which I would not have thought possible if you had shown me one. They really are beautifully made instruments.

here are some measurements I did on my desk, feeding the AFM a regulated supply. As you can see t he amperage is not anywhere near what one would calculate.

Since you don't say where you connect the power and where you take the measurements, I can not really comment. What connections are you calling input and output? Where are you measuring the current? If this is your primary question, you need to explain exactly what you are doing for me to make any sense of it.  If you are following the cct you posted, I suspect it is misleading you. Again , read my article and you will see my interpretation of how it works. See whether that makes sense to you. It is possible that the Motronic is using the same AFM in a different way but I don't think the connections in your diagram are correct.

Beamer wrote:I have just realised that you started a new thread on this, rather than posting a comment on the flow meter restoration thread. It would be useful to at least have a link to what we are discussing. There was the initial thread I did as I was discovering how it all worked and another where I cleaned it all up into a tidier summary.  This is the summary which you should refer to , in case you are looking at the initial thread.This was pointed out in big red letters at the bottom of the last post in the initial thread.

Fair, the quote came from post 13 here:  https://www.k100-forum.com/t10996-flowmeter-restoration. 

Beamer wrote:Hmm, I'm not sure that a simple resistance measurement is the right way to test this unit. 

There is a resistor chain with carefully, laser-cut resistors on the ceramic support. Each interval is of the order of 10 ohms but they vary progressively. 

It seems that the carbon track is much higher resistance. This implies that the hard resistor chain is a potential divider that is fed with a fixed, controlled voltage. The carbon track is a high resistance path whaich probably draws little current and whose resistance may rather variable. 

I think this needs to be tested with a voltage applied to the chain and testing resistance at the wiper with a voltmeter ( ie drawing negligible current ). This probably nearer to the way the EFI unit is driving/using it. 

It's very likely driven with a 5V regulated supply. The total resistor chain is a little over 510 ohms, drawing 10mA. 

There is clearly a very light contact force on the wiper, presumably carefully calibrated for maximum lifetime. The contact resistance may not be critical in that context as long as the track is not worn through.

Yes, that did it. Feeding it with 5V between pin 5 and E with a bench power supply I get a nice stable output between the wiper ( pin7 ) and E. It goes from 4.45V down to 1.5V when in the fully open position. 


My guess is that these are factory filled with an inert gas or 'dry' air before being sealed. So subsequent to further testing I'll find a can of water-free air and seal up with silicon paste. ( This air can be bought in electronics or computer shops and techy bits of some supermarkets - for a silly price, of course. )

The point was, you stated that you had a total resistance of 510 ohms and with a 5V supply that would yield 10mA... I would of thought the same, but when i measured the current using my multi-meter I got no where near the V=I*R relationship (reference my "desk calculations" which are one response up).  I think i'm going to try another multi-meter, it just doesn't make sense that the math wouldn't work out... I'll report back.

Beamer wrote:That is true as a generality but in this case ( with the AFM in isolation , not connected ) we know it is a simple resistor chain with no external source of current so we can measure in situ.  I used a DVM for accuracy on my test measurements. You will see photos of the test setup in the article. You will notice that I am not measuring resistance directly, I am measuring the output voltage at the external connector under power conditions similar to running on a bike. Resistance measurements jump all over the place and what is relevant is the output voltage of the AFM.

You state that you aren't measuring resistance directly, I don't think that is the case when you made the schematic, correct?  You were trying to measure each resistor separately and that is how you came up with the individual resistances shown in the diagram https://servimg.com/view/19452982/32.  The resistance measurements you state that are "jumping all over the place" refer to the taking measurements on the I/O connector on the outside of the AFM, which I agree.  My point was can you really measure the individual resistances of the laser trimmed artwork while it is in the voltage divider network?  Meaning, if you took a razor blade and at each printed resistor cut the link to the rest of the circuit and then took the resistance measurement, would you get the same result as you did?

Beamer wrote:Obviously my comments relate to the electronics on the K series BMW motorbikes. If you are using this AFM  with different electronics you will need to take that into account. Sorry, I had not seen the detail of the Motoronic unit you provided. This is different to the units on K bikes.

What is the provenance of that circuit? Some of it does not make sense to me.

That circuit is supposed to be an official schematic for the ECU.  Really wish it had some more detail like resistor and capacitor values.  

Beamer wrote:Since you don't say where you connect the power and where you take the measurements, I can not really comment. What connections are you calling input and output? Where are you measuring the current? If this is your primary question, you need to explain exactly what you are doing for me to make any sense of it.  If you are following the cct you posted, I suspect it is misleading you. Again , read my article and you will see my interpretation of how it works. See whether that makes sense to you. It is possible that the Motronic is using the same AFM in a different way but I don't think the connections in your diagram are correct.

The chart I had posted in my last response directly correlates to the picture in the beginning of the thread in terms of signal names.  To ensure that there is no confusion, I re-did the chart to add pin #'s, that should make it more clear. 

 In the desk measurement setup I utilize a wall plug which converts down to 5V (5.163V exactly) and that has spade terminals that allow it to directly interface with the AFM connector.  I hook up the ground directly to the wall plug cable and then depending on which pin I want to read (12 or 7) I hook up the AFM to the wall plug in through my multi-meter such that I can read the current.  The point here is, if V = 5.163V and R = 563 ohms then why doesn't I = .009mA (top row of chart below)?

OK, now I have some pin numbers, it's a bit clearer where you are connecting.

What you are calling "input" is labelled "reference output" and what you are calling "output" is labelled "airflow input".

If you are grounding pin 26, the air temp pin is effectively grounded through the thermistor and will never produce any non zero input. However you state you see 50 ohms between 26 and 7. Obviously your diagram is incomplete.  Again, look at my description , I laid all this out , there are more resistors than the loop around the wiper.

 I photographed the connector pin layout and included it since there seem to several different numbering systems for basically the same unit. No idea why but be warned.

I suggest you check the resistor layout against what I show, maybe they have a different configuration for the Motronic.

BTW, as I note above and in the article, the resistor measurements were not mine. Again I suggested you check what you have. If the wiper is at rest ( closed ) and there are no external connections you should be able to measure each resistance in the chain. Connect probes to the metal each side of the green strips, taking care not to touch or damage the green part : it's just carbon too and laser trimmed.
You will need this info for your software , so you need to get it first hand on the unit you are using.

The thing which tends to jump around is measuring the total resistance at the connector when you move the wiper. This is why you need to test under a simulation of working condition by measuring voltage output , not resistance. The fixed resistances will not jump.

 In the desk measurement setup I utilize a wall plug which converts down to 5V (5.163V exactly) and that has spade terminals that allow it to directly interface with the AFM connector.  I hook up the ground directly to the wall plug cable and then depending on which pin I want to read (12 or 7) I hook up the AFM to the wall plug in through my multi-meter such that I can read the current.

I'm sorry I'm having trouble following all that. Words are rarely good for describing such things accurately, that's why we invented circuit diagrams. You should be connecting + and - from the DC supply and if you want to measure the current being drawn, put the ammeter in series with the voltage supply. 

Ohm's law should apply so clearly there is some misunderstanding of what you are measuring or how you are doing it.

I think i'm going to try another multi-meter, it just doesn't make sense that the math wouldn't work out... I'll report back.

How are you getting? Have you managed to make any more sense of it yet?

Oh well, I guess he realised that he did not know how to use an ammeter and slunk away in embarrassment.

Don't mention it, glad to be of help.