VFR and Digital RPM meter on drill press, "mysterious" behaviour...cannot find answer
Close
Login to Your Account
Results 1 to 7 of 7
  1. #1
    Join Date
    Feb 2021
    Country
    NETHERLANDS
    Posts
    4
    Post Thanks / Like
    Likes (Given)
    3
    Likes (Received)
    0

    Default VFR and Digital RPM meter on drill press, "mysterious" behaviour...cannot find answer

    Goodday,

    I've upgraded my 220V drill press with a VFD and a 3x220V motor. That works great.

    Because the transmission ratio's has changed i've ordered a simple digital RPM meter to get an accurate RPM readout.
    I've mounted a magnet on top of the rotating main shaft and made a bracket/mount for the sensor.

    First there seemed to be no problem. Digital readout was good when manually tested.
    But, after assembly on the drill there was no readout as soon as I switched the machine on.

    I tested and tested and finally found out that when the sensor touched the metal of the drill the display only showed stripes.



    When I held the sensor in my hands though it showed the RPM, one step forward!
    So I isolated the holder and thought that would solve the issue.

    img_3584.jpg

    Unfortunately not. When the sensor is in the holder the RPM readout switches continuously...




    I thought of vibrations that maybe could trigger the sensor? So I welded the support for extra rigidity. No luck sofar...

    Does any body know a solution?

    Thanks

    Rob

  2. #2
    Join Date
    Mar 2005
    Country
    UNITED STATES
    State/Province
    Indiana
    Posts
    6,543
    Post Thanks / Like
    Likes (Given)
    2350
    Likes (Received)
    2542

    Default

    Is the drill press grounded properly? Check with an ohmmeter and be sure. Try moving the sensor closer to the magnet, and also try flipping the magnet over so the polarity is reversed. If none of that works you may have a faulty tach. Or maybe interference from the VFD.

  3. Likes neanderthal mach liked this post
  4. #3
    Join Date
    Feb 2021
    Country
    NETHERLANDS
    Posts
    4
    Post Thanks / Like
    Likes (Given)
    3
    Likes (Received)
    0

    Default

    Hello eKretz, Thanks for your reply.

    I will answer on your remarks:
    * The grounding, that's okay.
    * I've adjusted the magnet to several different positions, very narrow Gap and further away.
    * The magnet and sensor work fine when I rotate the drill axle by hand (when the sensor is in the bracket) there is a RPM readout then, but as soon as I switch the machine on it starts to fluctuate.
    * The main test I did was to hold the sensor in my hand and close to the same magnet/setup, then the readout is fine and constant when the machine is switched on! So the device seems to be able to work properly.

    The last mentioned point is what mainly puzzles me; why is everything fine and stable when I hold the sensor in my (gloved [and therefore isolated]) hand, when I hold it close to the identical magnet in the identical setup??

    I would presume that mounting it in the insulated holder would do the same trick, but then it goes "rogue". I first thought that vibrations could be an issue, but there are no mechanical parts in the Hall sensor I discovered...

    Interesting puzzle....

  5. #4
    Join Date
    Oct 2012
    Country
    UNITED STATES
    State/Province
    Iowa
    Posts
    433
    Post Thanks / Like
    Likes (Given)
    166
    Likes (Received)
    102

    Default

    The other day when I was trying to find the vibration issue with my tool post grinder, I had an accelerometer and oscilloscope to view the signal. On an adjacent lathe, I had left the VFD on, I was getting hash spikes at 100KHz, with that VFD just powered, and not running the machine.
    I found that the noise was coming from a common mode noise via the ground wire, and the ground to the machine chassis. The VFD was injecting that into the shop ground system.

    I have no suggestions on fixing your specific proximity type sensor EMC issue.
    If you leave your belt settings on the drill press fixed to a constant reduction, you can program the VFD to display the RPM via it doing math with what the Hz setting is, and a constant multiplier you program into the machine. But I'm believing that's not your intent. I use a VFD on 3 belt driven machines, and can leave the belt settings to their mid range and program the VFD for 0-120Hz operation. This works if your wanting enough torque to machine, but not needing hellacious spindle RPM, with that fixed belt setting.

  6. #5
    Join Date
    Feb 2021
    Country
    NETHERLANDS
    Posts
    4
    Post Thanks / Like
    Likes (Given)
    3
    Likes (Received)
    0

    Default

    Quote Originally Posted by ignator View Post
    The other day when I was trying to find the vibration issue with my tool post grinder, I had an accelerometer and oscilloscope to view the signal. On an adjacent lathe, I had left the VFD on, I was getting hash spikes at 100KHz, with that VFD just powered, and not running the machine.
    I found that the noise was coming from a common mode noise via the ground wire, and the ground to the machine chassis. The VFD was injecting that into the shop ground system.

    I have no suggestions on fixing your specific proximity type sensor EMC issue.
    If you leave your belt settings on the drill press fixed to a constant reduction, you can program the VFD to display the RPM via it doing math with what the Hz setting is, and a constant multiplier you program into the machine. But I'm believing that's not your intent. I use a VFD on 3 belt driven machines, and can leave the belt settings to their mid range and program the VFD for 0-120Hz operation. This works if your wanting enough torque to machine, but not needing hellacious spindle RPM, with that fixed belt setting.
    Dear people,

    The problem is solved, the mystery not yet.

    I switched the power supply for the tachometer from a lab power supply to a 15V connection on the frequency controller, so I could bring the tachometer closer and film tests better.

    And, that was the solution ... A completely stable speed display appeared.

    Design and stable display:


    Changing the speed


    So that solves the problem, but I don't yet understand why that is the case. I had also tried different voltages on the lab power supply, that didn't matter.

    Just like before, when I touch the metal with the sensor, the display only shows dashes. But as long as the sensor is mounted in the insulated holder, there is no problem anymore.

    It is strange that in the many films that I see on youtube no one has shielded / isolated the sensor from the machine. I have to insulate the sensor from the metal of the machine. I think that part has to do with the mystery, but I don't know for sure.

    Anyone who knows the real total solution may say it ...

    In any case, thank you very much for your comments!

  7. #6
    Join Date
    Nov 2004
    Location
    Webster Groves, MO
    Posts
    7,985
    Post Thanks / Like
    Likes (Given)
    2077
    Likes (Received)
    3909

    Default

    Working with sensitive electronics, I have seen similar effects many times. One of the causes is having a common ground for a sensor and something that generates spikes. You can build up an amazing voltage on a ground wire, just from its resistance and inductance. People would ask what were the rules. To an extent, there are no hard and fast rules. You have to look at the setup and consider the possible effects. One time we had a simple, somewhat regulated supply in an instrument we were developing. The original, built on Vector board, worked fine. When it got translated to a PC board, it had noise pips that seemed to be incurable. It turned that the person who laid out the PC ran the voltage sensing and charging current through a couple of inches of the same trace to common. The voltage built up in that short trace was added to the output voltage sensed by the regulator and it did its thing, adjusting the output to what it thought was the correct value. If you have an oscilloscope, you may be able to find a similar noise somewhere like in the ground wires.

    Bill

  8. #7
    Join Date
    Feb 2021
    Country
    NETHERLANDS
    Posts
    4
    Post Thanks / Like
    Likes (Given)
    3
    Likes (Received)
    0

    Default

    Quote Originally Posted by 9100 View Post
    Working with sensitive electronics, I have seen similar effects many times. One of the causes is having a common ground for a sensor and something that generates spikes. You can build up an amazing voltage on a ground wire, just from its resistance and inductance. People would ask what were the rules. To an extent, there are no hard and fast rules. You have to look at the setup and consider the possible effects. One time we had a simple, somewhat regulated supply in an instrument we were developing. The original, built on Vector board, worked fine. When it got translated to a PC board, it had noise pips that seemed to be incurable. It turned that the person who laid out the PC ran the voltage sensing and charging current through a couple of inches of the same trace to common. The voltage built up in that short trace was added to the output voltage sensed by the regulator and it did its thing, adjusting the output to what it thought was the correct value. If you have an oscilloscope, you may be able to find a similar noise somewhere like in the ground wires.

    Bill
    I would love to break down this problem with a oscilloscope, unfortunately I don't own one (yet! ��). Both the lab power supply and the VFR are on the same plug of the garage, share the same ground. But it seems that the VFR indeed has an influence somewhere. I first thought of a high frequency disturbance because I don't use shielded wires. But it seems to be something different. Anyhow, I can build further now!


Tags for this Thread

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •