Shop OBEX P1 Docs P2 Docs Learn Events
the right transistor to drive a motor — Parallax Forums

the right transistor to drive a motor

agfaagfa Posts: 295
edited 2008-09-13 19:53 in Propeller 1
i want to drive a motor /variable speed /one direction /using PWM.· motor supply is 14v but will be 12v. stall current is 8 A.

using the components i have on hand i tested the circuit attached but used 5v (continuous)·as a test i/o. seems to work ok, but i have·a few questions that i couldn't find answers to.

*i question whether or not i am getting saturation.· i measure approx. .9 v at collector, and transistor (tip3055)·runs very hot.

*will the transistor operate cooler at full saturation?· I will be using a heatsink.

*will the darlington arrangement i have be able to handle PWM ok?· i seem to remember posts saying there could be an issue with some PWM freqs.

*how can switching speed be determined with a darlington?· tip3055=2MHz (min) & kn2222=250MHz.

*how can total gain be determined with hFE specs, tip3055@ IC-1A= 55 to 150 & kn2222 IC-150mA 100 to 300? seems like·a broad range.
transistor.jpg
421 x 376 - 59K

Comments

  • Mike GreenMike Green Posts: 23,101
    edited 2008-09-10 22:37
    The TIP3055 looks like a bad transistor for this application. If you read the datasheet from a manufacturer, you'll see that the saturation voltage can be as high as 1.1V at some voltages, even higher at high currents.

    Things that contribute to heat dissipation: saturation voltage, load current, base current. The base voltage is fixed by the physics.

    PWM ... the issue is that Darlingtons aren't as fast as single transistors. You determine the switching speed from the datasheet or from experimentation.

    hFE is the gain. It's normally given as a broad range because that's what it really is. If you see a narrower gain range, the devices are probably tested and selected and the cost reflects that. The gain also varies somewhat with the current.
  • agfaagfa Posts: 295
    edited 2008-09-10 22:47
    why is that saturation voltage bad?· i·saw that spec.· i thought that the darlington arrangement would compensate.


    Post Edited (agfa) : 9/10/2008 10:55:39 PM GMT
  • Mike GreenMike Green Posts: 23,101
    edited 2008-09-10 23:08
    Here's a nice power transistor. It's only rated at 6A, but you can probably find similar ones rated a little higher. It can certainly handle the voltage you're using and the saturation voltage is guaranteed to be under 380mV with a base current of 300mA (at 6A collector current).
    www.st.com/stonline/products/literature/ds/12350.pdf
  • agfaagfa Posts: 295
    edited 2008-09-10 23:56
    thanks Mike, i do appreciate it.
  • BadgerBadger Posts: 184
    edited 2008-09-11 13:13
    Hey agfa

    hey if you go out to i think the spelling is pololu.com you will have to check on that . you can google the following micro dual motor controller and find it.
    it is .9 x .45 in size will handle 2 motor with 127 speed settings forward and reverse. Will connect to an external power sourse and only use to pin on the pic board. it has built in brown out circuits and such and a few more things i can remember. the partial assembled kit delivered to my address here in mid ohio is 22.90 it is a good buy for the price.. oh by the way if you wish to have more than on. i think you can daisy chain several together and control them from the pic board .. it is the best choice all around to building and designing your own... if you wish to build your own they offer all the indivigual parts for this as well.

    badger
  • BadgerBadger Posts: 184
    edited 2008-09-11 13:14
    i did not read your origanal post i dont think you can use pwn it a serial device.. your choice
  • Beau SchwabeBeau Schwabe Posts: 6,560
    edited 2008-09-11 15:12
    agfa,

    Another reason not to use a Darlington for high current demands is that between the C-E (Collector-Emitter) junction, you have a 1.2V drop where as with a standard bi-polar transistor the voltage drop is only 0.6V. In either case however with an 8 Amp load, I would consider using a MOSFET with a low Rds value.

    Here is an Example with a 12V supply under an 8 Amp load....

    With a Darlington, since it will automatically drop 1.2V, it will need to dissipate 9.6 Watts of wasted heat (P = I*V = 8A * 1.2V = 9.6 Watts)
    With a bi-polar, since it will automatically drop 0.6V, it will need to dissipate 4.8 Watts of wasted heat (P = I*V = 8A * 0.6V =·4.8 Watts)

    If a MOSFET (IRLZ44Z as an example) with an Rds of 0.0135 Ohms is used, you have a voltage drop of about 0.108V (V = I*R = 8A * 0.0135 Ohms = 0.108V)
    With a 0.108V drop across the MOSFET, it will need to dissipate only 0.864 Watts of wasted heat. (P = I*V = 8A * 0.108V =·0.864 Watts)


    http://www.irf.com/product-info/datasheets/data/irlz44z.pdf

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Beau Schwabe

    IC Layout Engineer
    Parallax, Inc.
  • agfaagfa Posts: 295
    edited 2008-09-12 12:27
    Badger, thanks for the tip.· i actually have one, if it could handle the curent i'd use it.· i would just buy a motor controler, but can't right now.· so was hoping to use parts on hand and what i could salvage.· since the project is just for fun, why not use the opportunity to learn.

    Beau, thanks for the explaination.· i have a better understanding now.

    btw, i did try it with an actual 5v pwm.··to big of a hurry to·use a heatsink or a protection diode.· it worked great for about 30·seconds.

    thanks again guys.

    agfa
  • Alex StahlAlex Stahl Posts: 4
    edited 2008-09-13 02:20
    Beau, thanks for suggesting the irlz44z. It's specs are fine for my application. Can the gate be driven directly from a propellor output pin?

    thanks

    Alex
  • Alex StahlAlex Stahl Posts: 4
    edited 2008-09-13 02:23
    Its. smile.gif
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2008-09-13 02:28
    Alex,

    For some reason your turtle avatar requires QuickTime on my computer and doesn't show up unless I load it into a separate page. Can you change it to a JPEG or GIF?

    -Phil

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    'Still some PropSTICK Kit bare PCBs left!
  • Alex StahlAlex Stahl Posts: 4
    edited 2008-09-13 02:46
    Sure, thanks for letting me know.

    -Alex
  • evanhevanh Posts: 15,545
    edited 2008-09-13 07:11
    Alex Stahl said...
    Can the gate be driven directly from a propellor output pin?
    Yes, but depends on switching frequency vs slew rate. And still need a series resistor to reduce possible latchups and it's a good idea to throw in a clamping zenor between gate and source as well.
  • Beau SchwabeBeau Schwabe Posts: 6,560
    edited 2008-09-13 18:43
    Alex Stahl,

    "Can the gate be driven directly from a propeller output pin?" ... Not a good idea. Transistors are current devices, the B-E junction of a transistor likes to stay around the band-gap voltage to operate. For a bi-polar transistor this band-gap voltage is typically 0.6V. For a Darlington this band-gap voltage is typically 1.2V or two band-gap voltages. A resistor is used to adjust the current applied to the transistor at the B-E junction. The voltage near the B-E junction will generally stay near the band-gap voltage or near two band-gap voltages for Darlington transistors. If you exceed the band-gap (or two) voltages, then you can "burn-out" your transistor.

    http://forums.parallax.com/attachment.php?attachmentid=37701

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Beau Schwabe

    IC Layout Engineer
    Parallax, Inc.
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2008-09-13 19:14
    Alex Stahl said...
    Can the gate be driven directly from a propellor output pin?
    For some MOSFETs, under certain circumstances, yes. A complete answer depends on the MOSFET's gate capacitance and the required switching speed. MOSFETs with a high gate capacitance will pose an instanteous short upon the driving pin when it changes state from low to high or high to low. If the gate capacitance is low and/or the switching is infrequent, this probably won't hurt the Propeller, although a small series resistor will help to alleviate current surges. If the gate capacitance is high and being driven with a PWM signal, you will want to use a driver chip that can handle high transient currents, such as the ones made by Micrel. In any event, make sure to add a pulldown from the Propeller's driving pin to ground, so that the MOSFET isn't inadvertently being driven when the Propeller is in reset.

    Also of concern is the Propeller's 3.3V "high" output. Many MOSFETs, including some "logic" MOSFETs won't turn completely on with such a low gate voltage. Be sure to check the RDS vs. VGS graphs in the datasheet. The gate threshold, by itself, is not a reliable indicator of the gate voltage required to drive a load.

    -Phil

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    'Still some PropSTICK Kit bare PCBs left!

    Post Edited (Phil Pilgrim (PhiPi)) : 9/13/2008 7:21:36 PM GMT
  • Beau SchwabeBeau Schwabe Posts: 6,560
    edited 2008-09-13 19:53
    Alex Stahl,

    Sorry I misread the post, Phil is right. The 'gate' of a MOSFET is capacitive and can be directly connected, but you need to observe the in-rush current effects mentioned in the previous posts or you can destroy the capacitor. With a bi-polar transistor however (including a Darlington) you must have a resistor on the 'base' to limit the current.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Beau Schwabe

    IC Layout Engineer
    Parallax, Inc.
Sign In or Register to comment.