Connect with us

Help with MOSFET switching circuit

Discussion in 'General Electronics Discussion' started by FlexedCookie, Oct 2, 2014.

Scroll to continue with content
  1. FlexedCookie

    FlexedCookie

    11
    0
    Oct 2, 2014
    Hi guys,

    Needing some help in getting my MOSFET switch circuit to work.
    Currently im using IRFZ34NPBF power mosfet, and im trying to use it to switch a DC motor on and off with PWM from an arduino.

    My current circuit is attached below.
    At the moment, I've got it wired up as shown, and am just trying to control the gate voltage with a power supply outputting 3V DC.

    My problem is that whenever I do this, the mosfet never turns on. I checked the voltage between the resistors and the gate, and it seems like there is no voltage going to the gate of the mosfet.

    Also interestingly, i can take out the wire leading to the gate from the resistors, and then just put a resistor from the gate of the mosfet to ground, and all of a sudden the motor works. However, i can do this nonstop until i turn off the power supply, which makes me think the mosfet isnt turning off.
     

    Attached Files:

  2. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,412
    2,779
    Jan 21, 2010
    Check the datasheet.

    This device needs a VGS of at least 4.5V, preferably higher.

    Your circuit will give only 1.5V which will likely not turn the mosfet on at all.

    You will need to use a transistor or similar to use the 12V supply to provide a higher voltage to the gate. Unfortunately, this may result in the input needing to be inverted (unless you use yet another transistor).

    If it's your software then changing the logic should not be difficult.

    md.png

    Here is an example of a gate driver that may work for you.

    In your case, you may also need further gate protection if your 12V supply is subject to voltage spikes. (i.e. this is conceptual, not a direct drop-in solution)
     
  3. FlexedCookie

    FlexedCookie

    11
    0
    Oct 2, 2014
    Hmmm i see.

    I've just read another thread which mentioned logic level mosfets. Do you think they would be useful/ easier in this application?
     
  4. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,412
    2,779
    Jan 21, 2010
    Logic level mosfets might be a solution, however you would have to increase your R1 (by a factor of 10 perhaps) and/or reduce the value of R2 by a factor of 10. Switching speed may be an issue.

    What frequency does your PWM operate at?
     
  5. KrisBlueNZ

    KrisBlueNZ Sadly passed away in 2015

    8,393
    1,270
    Nov 28, 2011
    Yes, a logic-level MOSFET is probably the answer. There are some pretty impressive ones around now, like the Alpha & Omega AO4402: http://www.digikey.com/product-detail/en/AO4402/785-1549-1-ND/3621495

    It comes in an SO-8 package with the standard pinout, so there are lots of drop-in replacements, and it has a guaranteed ON-resistance of only 0.007Ω with just 2.5V of gate voltage! Actually it's overkill, but at only USD 0.97 in 1-up quantity, for a one-off project, who cares!

    You won't find a logic level MOSFET like this at Radio Shack or any other small store; you'll have to get it from a big company like Digi-Key (see that link) or Mouser or similar. Also it's SMT so it's a bit harder to work with than a through-hole part, but SO-8 MOSFETs are not that hard to work with - you can solder them directly to stripboard with no trouble.

    You can maximise your gate voltage by connecting R1 at the left end of R2. And 10k is a very large value for a gate resistor; it will form an RC delay circuit with the gate-source capacitance of the MOSFET. This will slow down the turn-ON and turn-OFF of the MOSFET, and beyond a certain PWM frequency, the gate voltage will be smoothed out so badly that the MOSFET will remain in its linear region constantly, and will dissipate a lot of power. That's why Steve asked you what your PWM frequency is.

    Also can you show us or tell us where the control signal comes from.

    The diode you have across your fan load should not be a zener diode.
     
  6. FlexedCookie

    FlexedCookie

    11
    0
    Oct 2, 2014
    Thanks so much for your help so far guys.

    The control signal will be coming from an Arduino, so hopefully just PWM.

    Couple of questions:

    - Is the IRLZ34NPBF a suitable logic level mosfet?
    - What value resistors should I have for the best operation?
    - If not a zener, what else should i use for snubber/ inductive kickback protection?

    Also im still alittle confused about the schematic; should I just move R2 to the left end of R1 and then the circuit should be fine, or do i need to change more than that?\

    Thanks again guys
     
  7. KrisBlueNZ

    KrisBlueNZ Sadly passed away in 2015

    8,393
    1,270
    Nov 28, 2011
    PWM at what frequency?

    Do you have a higher voltage supply rail available?
     
  8. FlexedCookie

    FlexedCookie

    11
    0
    Oct 2, 2014
    My apologies. I think it will be variable, considering we want to do speed control with it. I'm sorry if this isnt making much sense, I'm the guy in charge of electronics and my team mate is looking after software, and hes not here at the moment.
    As for a higher voltage supply rail, there is a 12v rail, that goes to a regulator which pushes it down to 5v for the arduino.
     
  9. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,412
    2,779
    Jan 21, 2010
    Assuming you can get 4V or more to the gate you'll be OK in terms of being able to (eventually) turn it on sufficiently. (but that's an absolute minimum). You would also be advised to determine the short circuit current of the motor.

    What is the frequency?

    Just use a normal 3A rated diode.

    R1 and R2 are in the right place, but R1 needs to be at least 10 times the value of R2. I would recommend R2 being 100 ohms or less, but this will depend a lot on frequency.
     
  10. FlexedCookie

    FlexedCookie

    11
    0
    Oct 2, 2014
    The IRLZ34NPBF is logic level so it should only need 1v right?
     
  11. KrisBlueNZ

    KrisBlueNZ Sadly passed away in 2015

    8,393
    1,270
    Nov 28, 2011
    You can take it from here Steve
     
  12. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,412
    2,779
    Jan 21, 2010
    Check the datasheet.

    You're probably looking at VGS(th)

    vgsth.png

    This tells you (sure enough) that the mosfet begins to turn on at between 1 and 2 volts. Note that the current that can flow through the mosfet is 0.00025A, and you need at least 3A, possibly significantly more.

    What you need to look at is this graph:


    xfer.png

    This tells you that you need at least 3V to get the current you require. However I wouldn't recommend anything less than 4V, and if it is available I would recommend 6V or higher.

    The critical issue from here is the FREQUENCY (<-- find this out!!!!!) so we can determine whether the arduino can supply sufficient gate current to turn the device on and off fast enough.
     

    Attached Files:

  13. KrisBlueNZ

    KrisBlueNZ Sadly passed away in 2015

    8,393
    1,270
    Nov 28, 2011
    Steve, he has a 12V rail available, so he can use a gate driver. Do you have a recommendation?
     
  14. FlexedCookie

    FlexedCookie

    11
    0
    Oct 2, 2014
    The max supplied current from the source will be 3A (it's fixed unfortunately).
    The software guy is a bit hesitant to give me an exact number, but he said the minimum would be ~62Hz for an 8 bit timer? I hope that helps.
    Also, since the arduino can output like ~4.5V, the only factor for this mosfet, is making sure it does it quick enough right?

    Also space is a big issue, so the least amount of components the better
     
  15. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,412
    2,779
    Jan 21, 2010
    Ask your software guy if he really means PWM. For PWM, typically the frequency does not change. In any case, we're interested in the MAXIMUM frequency.

    For a driver, you could use a TPS2812, or a TC4427. I only mention these because I happen to have some of them on hand and I would consider them if I were doing this. They are both non-inverting which is an advantage. I haven't done a lot of investigation, you should refer to the datasheets to see if they're applicable for you.
     
  16. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,412
    2,779
    Jan 21, 2010
    Oh, and if your power supply is limited to 3A then you may have problems driving a motor unless that 3A is the current at the full rated load (and especially at startup).
     
  17. FlexedCookie

    FlexedCookie

    11
    0
    Oct 2, 2014
    Apparently the max from the 8bit output is 62kHz. And the motor is spec'd for 3A and 8v, but will happily run for long enough at 12v... :)

    So is there any chance of just being able to achieve this with a mosfet or am I looking too small and should consider more parts?
     
  18. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,412
    2,779
    Jan 21, 2010
    I'll do some calculations and get back to you.
     
  19. FlexedCookie

    FlexedCookie

    11
    0
    Oct 2, 2014
    I owe you a footrub. Thank you.
     
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

-