Connect with us

Building a Cheap Tachometer

Discussion in 'Electronic Design' started by luki, Apr 9, 2007.

Scroll to continue with content
  1. luki

    luki Guest

    Hi everybody,
    I need to build a tachometer circuit to measure the speed of a motor.
    Any suggestion on the technique,circuit, problem that might exists,
    estimated cost of building such device,etc ? Any pointer would be
    highly appreciated


  2. We need more informaton. What kind/size/speed is the motor? What kind of
    power does it run on? At this point, I can only offer the advice that Don
    Lancaster would probably give you......use a PIC (and don't try to get rich
    by writing books). ;-)
  3. not really enough info, but you could start here

  4. luki

    luki Guest

    A small DC motor maybe, i don't know (i don't decide it yet) but would
    not be very big sized motor. Is there any circuit that would generally
    be able to measure motor speed?
    If not, what really caused that? Is there any special attention i must
    deal regarding each type/size/speed?
    This is the background\purpose of this tachometer :
    1. I was developing a motor controller, so i need a speed sensor to
    control my motor
    2. The motor would be a small DC motor, to be easy to control
    Any suggestion about the motor itself?
    A cheap frequency to voltage converter would be nice. That means i
    must measure the frequency change according to Doppler? What kind of
    signal that must be transmitted?
    What if I can't have LM2907, quite hard to find things like that in my
    country. Is there any alternative. Or could i made frequency to
    voltage my own?

    I really appreciate your posts. Looking forward to hear more.

  5. linnix

    linnix Guest

    How small? All CPU fans (three wires) have pulse sensors. Can you
    adapt one of those?
    If you need to control it anyway, why not a micro to count the pulse
  6. D from BC

    D from BC Guest

    Some Wild Ideas:
    Pimp up an optical mouse?
    Hook up the motor to a potentiometer. Measure delta ohms/sec.
    Put a magnet on the shaft and use a hall sensor connected to a timer.
    Connect the motor to a dynamo. Measure Vout to measure rev/sec.
    Drill a hole through the shaft. ...
    Beam a laser through the hole..Measure laser light interruption rate.
    Put a mirror on the shaft..
    Beam a laser at the mirror .Measure laser light interruption rate.
    Go to scrap yard...Get old tach from an old car for $5.00.
    Go to automotive store..get new tach.
    D from BC
  7. There are many, but you need to specify how. For example, RPM can be
    determined optically, mechanically, electrically even acoustically. Allot
    depends upon your accuracy requirements, the speed of the motor and how
    often you need updates. You could count pulses (magnetic, optical, etc.)
    over a defined time interval, or measure the time between pulses. What
    works best depends upon your application.
    Just the obvious things like hanging a magnet on the shaft of a pager motor
    probably won't work well etc... ;-)
    Is this for something like a robot where measuring the wheel speed/rotations
    might be easier/better? Can you be more specific about your end goals?

    You probably shouldn't mix responses from different people into one reply
    like this. :)
  8. scada

    scada Guest

    You could hack an old CD player, many of the motors have tach pulse outputs.
  9. luki

    luki Guest

    I am developing a motor controller, to have the optimum response of a
    motor. Like a PID controller, i will use the motor speed as an
    indicator of it's response and compare it to a desired response. A
    descent accuracy would be enough, what i need is that the sampling
    rate is high enough to catch enough information.
  10. luki

    luki Guest

    Hi D, from your wild idea i think i would like to try using one of
    your idea. Optical mouse, and dynamo seems interesting. I will search
    for infomation on that subject, or if you have some can you give me a
  11. luki

    luki Guest

    This would be interesting. I'll try to dig more information on that.
    I think the measurement should be done independently. The computation
    in the micro is quite heavy.
  12. luki

    luki Guest

    I think i have an old CD player in my house. I will look for it, i
    will be a good and cheap alternative.
  13. luki

    luki Guest

    I like the wild ideas, i never even think of that (mouse, CPU fan, CD
    player, etc). Great replies. I really appreciate the all of the
    response and attention. Thanks a lot, it really helps me. Hope to be
    able to finish my project with this.


  14. jasen

    jasen Guest

    decide first. techniquies that are applicable to brushless DC motors aren't
    applicable to other types.
    it's basically a frequency counter calibrated in per minute (or per
    minnute/100) instead of per second (if for display)
    that depends what you want to do with it, the speed sensors on VCR and
    floppy drive motors generate several pulses per rotation to allow closer
    control of the speed
    if you don't need great precision that's one way to go.
    getting repeatable accurate results from a design using discrete parts
    reqwuires more skill (more than I have), how much precision do you need?
  15. jasen

    jasen Guest

    the two wire fans have them too, they are just missing the third wire
    (the sensor is needed to switch the coils in the motor)

  16. jasen

    jasen Guest

    Put some numbers on those goals, make them measurable: how much is "enough".

  17. A resolver or a quadrature encoder with a large number of pulses per
    revolution will allow you the possibilty of getting very close to
    optimum response, but probably not "cheap".

    Best regards,
    Spehro Pefhany
  18. MooseFET

    MooseFET Guest

    You may also find that getting 1/RPM is an easier thing to measure.
    If you only need to run over a narrowish range of speeds, this may be
    a better feedback in your servo. Unlike some of the other measures,
    you don't have to wait for several rotations to get it.

    Once the motor is turning, all you need to do is count how many clock
    cycles go by as the shaft makes one turn. The trick to not waiting
    too long when the motor is stopped is to only wait long enough to
    create the full error signal if the motor is slower than desired.

    You can also make an analog ramp that you capture when the shaft
    completes the rotation. A DAC controls the slope of the ramp. The
    ADC grabs the voltage when the shaft makes a full turn.
  19. MooseFET

    MooseFET Guest

    Look up a "Digital pot" if you want a lower cost encoder.

    If the motor is turning a gear, you could place a hall device or other
    thing so that it picks
    up the teeth going by.
  20. joseph2k

    joseph2k Guest

    Your attitude could be better, just direct the slob to s.e.b where it
    belongs. Better yet, take the whole "conversation" there.
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