Shop OBEX P1 Docs P2 Docs Learn Events
Is there a normally closed mosfet or switch? — Parallax Forums

Is there a normally closed mosfet or switch?

On a brushless dc motor there are 3 windings. If you short the windings(3 wires) together the motor becomes like a brake. On my motor driver board there is a method to short the windings in Brake Mode. However if there is no power to the controller the can be no braking. I am looking for an option to short the 3 wires together if there is No power but really do not want to use relays as I have seen failures, and a stuck relay would cost downtime. A “mosfet” - like switch would have less of a chance of failure IMO over many years of use. Any ideas on how this can be done? Ideally a prop with 3.3v would turn the brake off as needed.

Comments

  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2018-01-06 19:22
    Here's a circuit (untested) you might start from to experiment with:

    auto_brake.gif

    Without power, one MOSFET gate is biased high whenever the motor tries to turn, limiting the source-drain voltage to the gate threshold. The body diode in the other MOSFET conducts. (If the body diode can't handle the braking current, you will want to add diodes in parallel to the MOSFETs. ) A high on the NPN from the Prop keeps both gates low, preventing braking. For this to work, the motor coils must be logic-ground-referenced when the motor is running.


    -Phil

    400 x 206 - 2K
  • jmgjmg Posts: 15,140
    edited 2018-01-06 21:10
    T Chap wrote: »
    On a brushless dc motor there are 3 windings. If you short the windings(3 wires) together the motor becomes like a brake. On my motor driver board there is a method to short the windings in Brake Mode. However if there is no power to the controller the can be no braking. I am looking for an option to short the 3 wires together if there is No power but really do not want to use relays as I have seen failures, and a stuck relay would cost downtime. A “mosfet” - like switch would have less of a chance of failure IMO over many years of use. Any ideas on how this can be done? Ideally a prop with 3.3v would turn the brake off as needed.

    What current ?
    There are NC Solid state relays, usually with modest currents, digikey goes as low as 90mOhm : Circuit SPST-NC (1 Form B ) 90 mOhms Load Current 3.25A

    There are depletion mode mosfets, that conduct with 0VGS and are turned off by a negative voltage.
    Search Digikey for MOSFET then filter using depletion. There is a IXTH16N20D2 with 73 mOhm @ 8 A, 0V spec. (not cheap)

    You could try a design that uses the motor back EMF, to provide some gate drive, to standard cheap Logic level MOSFETS, but you might want to avoid too-drastic brake action at power off ?
    - or even a storage-cap approach, so motor drive cell does not loose power as quickly as everything else.
  • Dynamic braking would be the best bet. DC brushless is an ac motor with a magnetic rotor. See if there are any braking options in your motor control. You could go old school and use a 3 phase shorting contactor with high watt low ohm resistors. Or an IGBT module STGI1F10CH60TS-L to drive it in the opposite direction like the mosfet theory already mentioned. Careful not to overheat the motor.
  • The motor driver does have braking, this is for a case of fail secure so that if there is a power failure the motor will be very difficult to turn having a gearbox on it. It would be unlikely that a powerfailure occurred while running. It rarely runs. I can’t imagine needing more than a few amps since at a standstill it will be almost impossible to spin it up anyway. I’ll study these ideas above.
  • You just might want to add a friction brake for power failure fail safe. Look at Anaheim Automation BRK-12H, BRK-18H etc.
  • DigitalBob has a very good suggestion of having something that requires power to DEACTIVATE, not the other way around.
  • DC braking can lead to actual breaking if it is accidentally applied during running operation. I remember I was told to apply safety braking to a big grinder wheel when the emergency stop was pressed but that proved to be more of a danger than just letting it spin down as the grinding wheel would suddenly unthread itself and fly across the room like lightning. One came very very close, and that was that, I let it spin after that and the power it generated was used to light up warning LEDs.
    The friction brake sounds more like what you need.
  • Phil Pilgrim (PhiPi)Phil Pilgrim (PhiPi) Posts: 23,514
    edited 2018-01-07 04:34
    Peter,

    The wisdom behind electronic braking depends upon the application. I've had several where electronic braking was not only the best and simplest solution, but also completely safe. So, without knowing the OP's app, it would be wrong to prejudge the safety of his desired braking scheme.

    -Phil
  • Peter,

    The wisdom behind electronic braking depends upon the application. I've had several where electronic braking was not only the best and simplest solution, but also completely safe. So, without knowing the OP's app, it would be wrong to prejudge the safety of his desired braking scheme.

    -Phil

    "can" vs "will" :) merely relating an experience and a word of warning in there.

  • kwinnkwinn Posts: 8,697
    DC braking can lead to actual breaking if it is accidentally applied during running operation. I remember I was told to apply safety braking to a big grinder wheel when the emergency stop was pressed but that proved to be more of a danger than just letting it spin down as the grinding wheel would suddenly unthread itself and fly across the room like lightning. One came very very close, and that was that, I let it spin after that and the power it generated was used to light up warning LEDs.
    The friction brake sounds more like what you need.

    Good point. High deceleration forces can be hazardous for both safety and equipment, however the deceleration can be controlled by placing a load on the motor instead of a dead short.
  • I already use an Anaheim automation 34 frame motor but you can’t just stick the brake mentioned above on it. My motors have custo
    Dual shafts for encoder on top. To add the break you have to special orde the motors with the brake and special extended shaft for the encoder to sit on it. I have an in-line electro mag shaft brake but i don’t want to deal with the extra hardware. I also have a toothed sprocket mod I can add to the output of the gearbox with a solenoid drive latch. This works well but long term the power for the solenoid can become a source of failure or the solenoid itself can fail. Since I am looking for the most maintenance free solution the motor winding short method (no moving parts) is still something I want to try. There is no concern as above mentioned about things flying apart. I will look up a NC SS relay and see what can be found. Thanks for the input guys.
  • T Chap,

    What about a UPS or some kind of other power backup that when activated shuts everything down and requires a manual restart.
  • One of the safety concerns I had with testing a system with a 3MVA generator and load, was if my electronics shorted the generator then that could lead to the destruction of the windings as they flew apart if the safetys didn't kick in on time. So from small 12V 10A grinder motors to 3MVA generators there are concerns about "electronic braking" but of course there are plenty of motors where that should not be a problem. Once again, just be aware of it. In fact the grinder motor needed soft starting as it appeared as a dead short when it was stalled and even the 14V/14A PSU complained.
  • T Chap wrote: »
    On a brushless dc motor there are 3 windings. If you short the windings(3 wires) together the motor becomes like a brake. On my motor driver board there is a method to short the windings in Brake Mode. However if there is no power to the controller the can be no braking. I am looking for an option to short the 3 wires together if there is No power but really do not want to use relays as I have seen failures, and a stuck relay would cost downtime. A “mosfet” - like switch would have less of a chance of failure IMO over many years of use. Any ideas on how this can be done? Ideally a prop with 3.3v would turn the brake off as needed.

    I lean toward the relay over a Mosfet just because a relay would not be taken out by a spike or surge, and if the system power drops the relay will drop. Everything can fail, just a matter as Peter implies, what are the risks for your various solutions. I also would not use a short, especially with electronics, depending on the surge your device will generate, you could likely punch your FETs rendering the braking function inoperable. Go with an appropriately sized power resistor. Heck, gang and step them if need be, kinda an inverse of softstart, a soft-stop if you will. Worse case on a relay, the surge welds the NC contacts during the braking time and the fault manifests before the system can be started up again. Just my two cents from repairing systems where paranoia was left behind long ago!
Sign In or Register to comment.