Connect with us

Increasing PWM output voltage

Discussion in 'Electronic Basics' started by Dave, I can't do that, Sep 18, 2008.

Scroll to continue with content
  1. Hello,

    I need to boost a PWM signal at 5KHz and 450mV to 5V at 5KHz and 100mA

    Could someone please give me a list of the components I need and a
    schematic to do this.

    A suggested idea was this...

    But as you can see the guy was unsure of N-channel or P-channel and he
    never got back to me.

    Thank you.

  2. Do you need to pull both up and down? What are you driving with the
    PWM signal?

    At 450mV, not many mosfets will work directly in any case. The easiest
    thing would be to use a comparator, and use the output to control a
    driver chip like an SN754410:

    Bob Monsen
  3. Hi Robert, thanks for the reply.
    I do not know what that means. I am trying to drive a 25W CO2 Laser
    marking machine's Laser intensity.

    I was given the dead Laser and installed one I bought surplus. I now
    have everything working except the intensity. It is supposed to put
    out 5v PWM at 5KHz but it is only putting out 450mV. There are no
    schematics available as it is such an old model and there are several
    EEPROMs on board and I have no idea what they are coded to do.

    The PWM is working just not at sufficient voltage to trigger the
    Laser. I want to use the existing 450mV PWM output to drive something
    that will output the 5v I need.

    I have been told that this Laser requires "Output Current, max 100mA
    through a 50 Ohm CMOS driver" and that's all I have to go on.

    If I can't do that then I will tear out the electronics and install a
    2-axis CNC stepper driver board and run it under EMC2. Obviously
    getting it to work would be better since it is configured as a printer
    for Corel and a simple matter to "print" the part. If I go the CNC
    route I will have to add in two more programming stages for every

  4. Jamie

    Jamie Guest

    Maybe what they were referring to is that it must be current regulated
    to not exceed the 100 ma or set point since a Laser current is critical."
  5. Hi again,

    I don't know about rise times etc as the PWM is already being
    generated, but at 450mV and I need to boost that to 5v. I can see the
    pulses on a scope and the frequency is around 4.980KHz and they are
    nice clean square waves that I can vary by software.

    I just need something that can be driven by the existing PWM output
    and it in turn outputs the same signal but at 5v.


  6. Ok, the output is probably too low resistance for the driver you are
    using. If you disconnect the thing you are driving, then it probably
    will go up to 5V, right? If so, all you need is a follower, which is
    an NPN transistor, connected to the PWM output and the input, like

    5V ---------------.
    | collector
    base |/
    PWM --------|
    | emitter

    (created by AACircuit v1.28.6 beta 04/19/05

    This will decrease the required current at the PWM to 1/50th of what
    was required before.

    At 100mA, if your PWM drives it the base up to 5V, then the transistor
    will only dissipate about 40mW, so any little NPN transistor will
    work. Radio Shack will have them. 2N4401, 2N2222, 2N3904, etc.

    This is probably worth a try.

    Bob Monsen
  7. neon


    Oct 21, 2006
    100ma 5amps it all related to time and cooling cycle. if i would have designed something like that would use a LM117 regulator as a current source and pulse it at interval. at hi watts it becomes dangerous for your eyes so be carefull. Can you pass 5 amps you bet. it is the power dissipation of the device that you have to worry about.
    Last edited: Sep 21, 2008
  8. Hi Bob,

    Thanks for the reply. No, this is 450mV with or without the Laser
    connected. The output remains the same into load no load.
  9. Hi John,
    Yeah, you are right. Here I am retired and living on SS of $1200 a
    month. I am trying to make a few extra bucks to brighten my lifestyle
    a touch. I bought the replacement Laser about 7 years back when I was
    working and it has been sitting waiting to be used for some project or
    other. When I was given the dead marking machine, I thought I was on
    the way to a little more fiscal independence.

    That seems a little ways off still, but I am hopeful and will keep
    plugging away at it.

    I am sorry, but there seems little point in emailing you as I have
    little enough to spend as it is.

    Thank you for the offer.

  10. Ok, then you'll need to go the comparator route. Here is a simple
    circuit that might work, using a TL431 as a comparator:

    | | |
    | | |
    | .-. |
    10k | 1k | | |
    ___ | | | |
    Input -|___|------' '-' |e
    A | b |<
    | o-------| PNP
    | | |\
    | .-. |c
    | 1k | | |
    | | | |
    | '-' o-------Output
    | | |
    | k |
    | - TL431 .-.
    '----------------r- ^ | | 10k (probably not needed)
    a | |
    | '-'
    | |
    | |

    (created by AACircuit v1.28.6 beta 04/19/05

    The TL431 is a shunt regulator. If the voltage on the 'r' pin goes
    above 2.5V, it'll suddenly become very low resistance between the k
    and a leads. So, if you set the pot so that the r lead goes a bit
    above 2.5V when the input is 400mV, and a bit below 2.5V when the
    input is 0, it'll cause the PNP transistor to turn on and off.

    Look at the datasheet link above for more information. You can get
    these at electronics surplus places online, or of course mouser or

    I'm sure there are simpler solutions, but this one will give you a
    nice crisp square wave with a minimum of parts.

    Bob Monsen
  11. Hi Bob,
    Thanks for that. I'll keep you posted

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