Connect with us

tapped trimpots?

Discussion in 'Electronic Design' started by John Larkin, Oct 4, 2007.

Scroll to continue with content
  1. John Larkin

    John Larkin Guest

    Does anybody make tapped trimpots?

    What I'd really like is a surface-mount part with lots of taps, not
    just a ct. Six or so would be nice.

    Google is quiet on the issue.

  2. crominator

    crominator Guest

    Series a bunch ............

  3. crominator

    crominator Guest

    If each tap needs full range - parallel 'em

  4. All I could find were huge wire wound rheostats. Not exactly surface
    mountable or miniature.

    I have no idea what you're designing, but with lots of taps, it might
    be easier to switch a mess of fixed resistors, and trim the final
    value with an ordinary pot. Kinda like a "range" switch, and a "fine
    tune" pot.
    Actually, Google search does talk:
  5. John Larkin

    John Larkin Guest

    I want to delay a digital signal from, say, 0 to 5 ns in one turn of a
    pot. Ideally, it would behave like a true variable delay line, and not
    lose bandwidth as the delay increases, and it would delay both edges
    equally. This implies that it must store more than one bit of

    So, this sort of works, but of course nobody makes the pot:

    I could use a tiny rotary switch to pick off the course delays, then a
    simple fine trimpot-based delay generator to interpolate between
    steps, but it would take two gadgets, and I'd like to put eight
    adjustments in a small box, and 16 widgets is a bit much.

    I recall that somebody used to make a variable delay line, sort of
    like a wirewound resistor, but they were huge. The programmable
    silicon delay lines aren't very good... coarse steps, large min delay,
    bad tc's, lots of jitter.

  6. JosephKK

    JosephKK Guest

    John Larkin posted to
    I have seen them with one and two taps. Though i have not seen any for
    over 20 years. Unless you need about a million of them, they would
    be considered a very custom part with a corresponding price.
  7. I just looked at a linear slide potentiometer in my junk pile. It
    kinda looks like some silver conductive paste can be smeared on the
    edge of the carbon area and brought out with a wire. Obviously, this
    is not a production concept, but it can work for a small number of
    It also eliminates using an all pass filter which was going to be my
    next suggestion. Basically, you have to maintain waveform integrity
    which means your circuit can't have any group delay and amplitude
    rolloff with frequency. It also eliminates using a variable inductor
    coupled with tracking variable capacitors due to the same reasons.

    So, that leaves a digital solution, assuming you don't mind
    regenerating the digital input signal.

    in---------------| \
    | | NAND O------- Out
    |---R-----| /
    | |------/
    \ /

    Yech, but it should work for small delays. Either R or C can be
    adjustable. A cleaner version could be done with a J-K flipflop, but
    the minimum delay with substantially larger than with a simple gate.

    Got numbers for frequency range?
    Duz the minimum deleay really need to be zero? That's not easy.
    Hmmm, you could build one. Take a soda straw like coil form, and wrap
    as many turns of magnet wire around the soda straw. Add taps at
    convenient places. It might work with just a single powdered iron
    core, moving in and out. However, if it doesn't, break the core into
    small pieces where the length is about 1/2 the space between taps. The
    other half, stuff in a plastic insulator. When you move that in and
    out, you change the inductance of each coil section equally.

    I vaguely recall that there was a commercial version of this
    abomination sold for SSB quadrature phase shift modulator designs.
    You're right, they were big because they had to work at audio
    frequencies. I'll see if I can find one, but I'm fairly sure you'll
    hate it.

    Disclaimer: I have't done this for a very long time.
  8. In the olden days *varicaps* (variable capacitance diodes) were used
    to make a variable analog delay line.
    And then 2 opposite each other .
    This was the way to shift couple of nanoseconds to even microseconds
    in TV.
    Ampex colortec and amtec time base correctors.
    So the C in you LC, replace it by varicaps, take signal from end.
  9. Rich Grise

    Rich Grise Guest

    "Sort of" works, OK, I'll grant you that, but my techie pea-brain
    is thinking, you can't pick up a fine delay with a pot across
    a stage of a lumped line - you'd get a signal with two humps -
    one at point A, the other at point B, with varying amplitude as
    you go along.

    Or am I missing something, as usual?

    And anyway, from the hacked trimpots I've seen, it would be almost
    impossible to construct one without a little dead band on top of
    each tap, but I'm thinking of wirewound. It could probably be done
    with cermet or something, but does that circuit _really_ act like
    an infinitely variable delay line?

  10. Penny & Gile$ may do a custom, or maybe somebody who does those force
    sensitive resistors may have something.

    then I found this

  11. JosephKK

    JosephKK Guest

    JosephKK posted to
    On the other hand if you can tolerate a dip package 10 taps about 1 ns
    each is perfectly feasible.
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