Announcement

Collapse
No announcement yet.

AFM Bench Testing

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    AFM Bench Testing

    I’ve been having a problem with my 1990 325ic where it is breaking up at 4,000-4,500rpm. Acts like it’s hitting a rev limiter, drives fine otherwise. Changed cap, rotor, wires, plugs and no change.

    I decided to pull the AFM and did the voltage test. Voltage seems to climb normally until the last 2-3% of travel in the flap. At about 97-98% I get 7.5V while at 100% I get 5.2V. It seems that the higher voltage lines up with a line across the carbon board and when the door goes passed that is when it drops. Wondering if this is how it normally functions or if something is wrong with my AFM.

    Essentially — .43V——>7.4V—>5.2V

    Any help is much appreciated!!

    #2
    There is something wrong for sure.
    Try moving the contact point to higher place i did mine by unlossing the board and get it lower (to spring wheel direction).

    Comment


      #3
      How are you measuring voltage on the AFM?

      You should never see any voltage higher than 5V so something is already wrong.
      Owner - Bavarian Restoration
      BMW and European Electronics Repair and Restoration
      www.BavRest.com
      My Feedback Thread
      Our Facebook!
      Follow our Instagram!

      Comment


        #4
        Did you have to modify the screw holes to slide the board? I pressed down the contacts by bending the arm a bit and noticed no change. I put it on the car with the arm exposed and it is breaking up way before the spot I noticed the issue. Like I said it’s literally the last 2mm or so it drops off.

        I pulled the meter and used a 9V and some wires to fire it up on the bench for testing.

        Comment


          #5
          Originally posted by IceyVert View Post
          Did you have to modify the screw holes to slide the board? I pressed down the contacts by bending the arm a bit and noticed no change. I put it on the car with the arm exposed and it is breaking up way before the spot I noticed the issue. Like I said it’s literally the last 2mm or so it drops off.

          I pulled the meter and used a 9V and some wires to fire it up on the bench for testing.
          You tested it all wrong. Here is why.

          First you should have never used a 9V battery. The DME powers the AFM with a highly regulated 5V supply. This means that if your AFM has a voltage drop (which it most likely does,) that voltage drop may be apparent with a 5V power but go away with a 9V power. This means you are experience a false positive test.

          Second, a meter takes and displays an average of the voltage measured. So when you use a meter to "test" the AFM, you are literally viewing the average of the voltage, not the actual voltage. Plus voltage drops can occur on the microsecond scale, something that a meter would never be able to see. Only an oscilloscope should be used to measure voltage because it will show you true signal. See my AFM thread for more info on AFM testing and how the output curves should look like.

          I've rebuilt many thousands of AFM's and never seen one enter our shop that I would consider "still good to use." Almost every single one has 30+ years of wear and buildup, causing it to be a failed unit. Unfortunately, every article I've read on the internet and even the Bentley repair manual are wrong because they are using 9V batteries or DMM to test them.
          Owner - Bavarian Restoration
          BMW and European Electronics Repair and Restoration
          www.BavRest.com
          My Feedback Thread
          Our Facebook!
          Follow our Instagram!

          Comment


            #6
            Alright, take my money.

            Comment


              #7
              I always do the resistance test to find out the carbon track worn area, the resistance should go up smoothly as the flap moves.

              What is the difference with resistance test and the voltage test?

              Comment


                #8
                The resistance test is not as accurate (or so I hear obviously I’m not an authority on this). Something about the meter only providing stepped averages.

                Comment


                  #9
                  Originally posted by Gregs///M View Post

                  You tested it all wrong. Here is why.

                  First you should have never used a 9V battery. The DME powers the AFM with a highly regulated 5V supply. This means that if your AFM has a voltage drop (which it most likely does,) that voltage drop may be apparent with a 5V power but go away with a 9V power. This means you are experience a false positive test.

                  Second, a meter takes and displays an average of the voltage measured. So when you use a meter to "test" the AFM, you are literally viewing the average of the voltage, not the actual voltage. Plus voltage drops can occur on the microsecond scale, something that a meter would never be able to see. Only an oscilloscope should be used to measure voltage because it will show you true signal. See my AFM thread for more info on AFM testing and how the output curves should look like.

                  I've rebuilt many thousands of AFM's and never seen one enter our shop that I would consider "still good to use." Almost every single one has 30+ years of wear and buildup, causing it to be a failed unit. Unfortunately, every article I've read on the internet and even the Bentley repair manual are wrong because they are using 9V batteries or DMM to test them.


                  Doesn’t the DME only see an “average” signal from the AFM? Do you have your scope set at the same sample rate as the DME? If you have your scope set at too high of a sample rate, that could explain why all of your tested AFM’s test bad.

                  Comment


                    #10
                    Why would the DME see an average? There is no pre-processing before the DME and it sees raw signal from the AFM. The source voltage is a regulated 5V. See my thread on AFM testing to learn more about AFM testing and how the output curves should look. The bad AFM's we see not only test bad on our scope but they are reported bad by the many thousands of customers that send them to us and report failures in their vehicles. These bad AFMs show microsecond voltage drops, which result in a misfire type symptom. Once refurbished, there are no more voltage drops and clients report a significant improvement in drivability and power and no more misfire type symptoms. This is just one of the many ways an AFM can fail, temperature sensor failure, mechanical failure, calibration drift are all very common issues we deal with.
                    Owner - Bavarian Restoration
                    BMW and European Electronics Repair and Restoration
                    www.BavRest.com
                    My Feedback Thread
                    Our Facebook!
                    Follow our Instagram!

                    Comment


                      #11
                      The DME is a digital device. Like all digital devices, it only samples data at the clock speed of the microprocessor. This means that it can only read the value of a signal when it is told to by the microprocessor. In between those clock cycle reading times, the value of the sensor is NOT used.

                      if your oscilloscope is sampling at a faster rate than the microprocessor, you are seeing values that the DME doesn’t see, and therefore doesn’t use.

                      Comment

                      Working...
                      X