Shop OBEX P1 Docs P2 Docs Learn Events
solid state relay + pwm — Parallax Forums

solid state relay + pwm

science_geekscience_geek Posts: 247
edited 2012-02-28 19:55 in General Discussion
i need to control a motor in one direction, due to the amp draw and the price of a commercial motor controller, i want to make a controller. the motor im using has a max amp draw of about 390 amps. i was thinking of using a megafet, but cant find a good one, i was wondering if setting up a solid state relay and then using a pwm would be ok, or if it would wear it out

Comments

  • Chris SavageChris Savage Parallax Engineering Posts: 14,406
    edited 2008-11-21 15:35
    Inherently an SSR is not prone to wearing out like a mechanical relay and many devices, including reflow ovens use them with PWM to maintain relative output. What can happen is that depending on the device connected to the SSR and whether or not it switches on the zero-crossing boundary and affect the life of the equipment connected to the SSR.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Chris Savage
    Parallax Engineering
  • TeslaTesla Posts: 55
    edited 2008-11-21 20:59
    Check out a company called Crydom. On there web site they have a bunch of SSR controls that could work for you. They have some that take an anolog in and do a phase angle firing that could work for you. Also all of there SSR's have TVS built in. I use them for all my industrial controls for 2kw+ heater controls(at my job).

    A controled phase angle fired SSR(two SCR's back to back) has great resolution and a couple DC motor controlers i have laying around also use the same technology.
  • science_geekscience_geek Posts: 247
    edited 2008-11-21 22:57
    ok, so now say i had 10 fets rated for 50 amps, and i was trying to drive the same 390 amp load, say i parallel them up so they can handle the current, 10 for safety reasons in case one fires early, would i be better off paralleling fets, or getting one or two big ssr.
  • kwinnkwinn Posts: 8,697
    edited 2008-11-23 21:57
    The problem with putting 10 fets in parallel is that they are not perfectly matched, so one may carry much more (or much less) current than another and the one carrying too much current will blow. For multiple fets (or bipolar transistors) you would need a resistor in series with each one to balance the currents. For 50 amps a .001 ohm 50 watt to .002 ohm 100 watt resistor would be required. Yes, I know those resistors are twice the wattage they are dissipating, but they will not last long if they run at their full rating.

    Easier to have one device for the full load current if possible.
  • stamptrolstamptrol Posts: 1,731
    edited 2008-11-24 00:56
    Multiples of the same model FET's will inherently current share because their temperature coefficient is opposite to bipolars.

    You stiil have to take care that the traces (or wire leads) are pretty equal in resistance. You do have to make sure you've got a robust source for the gate drives because they are quite capacitive and you want to drive the gate through the linear region as fast as possible.

    One question: you say maximum draw is 390 amps. Is that full load current? If it is, and your going to on-off control, your inrush is going to be several times the full load current.

    PS don't forget the fuses!

    Cheers,

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Tom Sisk

    http://www.siskconsult.com
    ·
  • 4x5n4x5n Posts: 745
    edited 2012-02-28 11:20
    Hate to say it but there are a number of good reasons why a motor controller that can feed 300 amps costs a lot of money! :-) That's a LOT of current and things get critical. Is the motor a DC or AC motor? In any case you're going to want to put snubbers on the motor to protect the electronics from the high voltage and high current spikes reverse voltage spike you get when the motor is turned off. Then there's the high current draw when the motor is first powered on. Your thyristors will need to be able to supply a lot more then 390amps for at least a short period of time.
  • localrogerlocalroger Posts: 3,452
    edited 2012-02-28 11:46
    What 4x5n said. Using PWM on a DC motor that size will generate some serious switching spikes. Design of the snubbers will be nontrivial. If it's an AC motor you need a frequency drive, which is fairly complex. Trust me, as expensive as you think commercial controllers are they are cheap compared the time it will take you to learn enough to engineer a solution that's unlikely to blow up either your homemade controller or the motor.

    I have worked in industry, including fairly heavy industry, hacking heavy-duty stuff for 25 years. I wouldn't try this.
  • PJAllenPJAllen Banned Posts: 5,065
    edited 2012-02-28 17:43
    Guys, the subject went cold over three years ago.
  • 4x5n4x5n Posts: 745
    edited 2012-02-28 19:55
    Interesting. Wonder how I saw it at the top of the list on the forum.?
Sign In or Register to comment.