Connect with us

LVDT question

Discussion in 'Electronic Design' started by Marco Trapanese, Apr 9, 2013.

Scroll to continue with content
  1. Take an LVDT (series-opposite), apply a sine wave on the primary coil
    and measure the output signal on the secondary coil with an oscilloscope.

    Do you agree you *must* see a change in the amplitude when you move the rod?
    At least you should expect a null signal when the rod is place at null

    What if this doesn't happen? I mean the output signal changes very
    slightly around it's maximum.

    Double checked the connections, changed the LVDT with new ones - no chance.

    What else?
  2. Jasen Betts

    Jasen Betts Guest

    have you got the terminals mixed up?

    swapping one of the primary terminals with one of the secondary
    terminals would give the effect you describe.
  3. Il 09/04/2013 13:03, Jasen Betts ha scritto:

    I checked the coils measuring their resistance.
    I'm pretty sure the connection is correct, anyway it's easy to swap two
    pins and retry. I'll do it in a while.

  4. Il 09/04/2013 14:29, Marco Trapanese ha scritto:

    Swapping two pins leads to a constant voltage with the same amplitude of
    the excitation voltage.
    With the correct connection the output voltage is quiet small.

    I don't think the sensors are damaged because we have several of them
    and all have the same behavior.

  5. Is it an AC LVDT?

    Is the secondary coil in fact two coils? This is usually the case. But
    possibly they have done this for you and there are only two wires

    In order to see a null you are supposed to connect the two secondary
    coils in series and in antiphase. So with the core in mid position, or
    absent, there is no net signal.
    I can see this happening if you are just using one of the two
    secondaries, and it is unloaded (like with a scope).
  6. Syd Rumpo

    Syd Rumpo Guest

    Ok, thats right.
    I'm suspicious about the core (rod). Are you sure it's the original rod
    and someone's not made a replacement from, say, Aluminium? Got a magnet

  7. Il 09/04/2013 14:40, John Devereux ha scritto:

    Yes, is a series-opposite configuration and the connection is made
    internally. There are only four wires.

    This is what I knew and what I expect.
    What about if the two coils are connected in series but not in antiphase?
    I think I should read a double voltage span, with no null of course.

    The secondaries are connected internally, and the two ends are connected
    to my amplifier board (which works fine with several other LVDTs -
    different models of course).

  8. Haha, yes, BTDT.
  9. Il 09/04/2013 14:53, Syd Rumpo ha scritto:

    I'm agree. I will search for a magnet...
    The rods were securely tied from the manufacturer along the sensor side.
    I opened the shipping box myself.

  10. Il 09/04/2013 15:09, Dennis ha scritto:

    Both frequency and excitation voltage are within the specification
    limits. They did the calibration at 3 kHz, I use 4 kHz... but it can't
    matter at all.

  11. What is the part number?
    Don't think it will work properly like that. Possibly if you put a load
    on, few hundred ohms perhaps.
  12. Should be OK then. Are you really really sure it is not a DC type?
  13. Bill Sloman

    Bill Sloman Guest

    It could, if somebody got cute, and tuned the primary. Self-resonance
    can be a bitch. Seems unlikely, but some peopled do really weird
  14. Il 09/04/2013 15:19, John Devereux ha scritto:

    Do you mean they could have an integrated amplifier?
    If so I cannot measure about 180 ohm on the primary and secondary coils
    in *both* directions...

  15. Il 09/04/2013 15:24, Bill Sloman ha scritto:

    The ds says from 2 to 10 kHz.
    Anyway I could try to change the frequency.

  16. Il 09/04/2013 15:57, John Larkin ha scritto:

    Was the first thing I tried to swap :)
    The behavior is identical.

  17. Il 09/04/2013 16:07, John Larkin ha scritto:

    Thanks John, I'll try to do this when I'll go back to the lab (next week).

Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day