Motor current sensing - Transient suppression

Discussion in 'The Projects Forum' started by JonHopper, Aug 12, 2010.

  1. JonHopper

    Thread Starter New Member

    Jun 28, 2010
    9
    0
    Hello everyone. I have a 24VDC bidirectional brush motor with a 24VDC 15A pwm controller, functioning quite well. The motor draws about 250ma no load with a stall current of a few amps.

    I am attempting to create a current sensing circuit using TI's INA195 highside current sensing shunt monitor chip. I have a 0.022ohm 50W sense resistor for voltage sensing.

    My main problem is that the transient voltages produced from the start / stop forward / reverse action of the motor has destroyed the first INA195 chip i've tested. Here is a basic circuit diagram of my test circuit so far.

    [​IMG]

    Vload is the motor side, Vsupply is the controller side. Vout is 0-5v analog signal.

    I am unsure of which transient suppressor to use (moreso what ratings). The INA195 can handle common mode voltages up to 80V so I am assuming the transients have been in excess of that. Do I need a clamping voltage of about 50v?

    Basically I'm worried if I go for a transzorb with 50v clamp rating, a transient voltage of 200v may destroy the transient supressor.

    All of the transient suppressors I have looked at haven't had a maximum voltage rating.

    Any help would be appreciated... thanks
     
  2. Bernard

    AAC Fanatic!

    Aug 7, 2008
    4,176
    397
    Just speed controlled one direction 5V @ 1A , just use a fast recovery diode or schottky 40V @ 1A in shunt with motor cathode to + side.
     
  3. marshallf3

    Well-Known Member

    Jul 26, 2010
    2,358
    201
    If you ever expect to reverse direction at some point then a single diode won't do it, you'll need to go with two zeners wired back to back as in K-K or A-A. In this manner you can choose zener votages just above the highest expected drive voltage and get some pretty good protection. For the sake of higher frequency transient response I'd parallel 0.1 uF or better capacitors across each of the zeners, many aren't known for having fast switching responses. Even run in a single direction as I think you may have stated its still a good setup to use.
     
  4. JonHopper

    Thread Starter New Member

    Jun 28, 2010
    9
    0
    Sorry for the confusion, I am looking for bidirectional operation.

    I've created a PCB for TVS devices and the current sensor, and i've tested it.

    I used some vishay TVS devices rated for 40v , but they do not allow any signal to pass through (I used bidirectional tvs chips). I measured the voltages directly across the terminals, the foward voltages change slightly when power is applied to the sense current resistor but they do not allow the 24v common mode signal to pass.

    The only other issue is the forward voltages. They mess with the current sense voltage between the common mode voltage.

    Anyways, I will try Zeners but I fear the forward voltages will still keep my readings going askew.
     
  5. JonHopper

    Thread Starter New Member

    Jun 28, 2010
    9
    0
    The main issue is this: The motor I previously referenced is merely a test motor, I have recently acquired the prototype motor, 24V gear motor with a no load current of 4A (this is a 77W motor with a planetary gear head).

    I can place capacitors across the motor terminals, however most zener / schottky diodes (all, that i have come across thus far) are rated for far less power than is required (about 10A @ 24v, 240W). I am worried that merely capacitors will not protect sufficiently. Transient voltages will be quite high with this powerful of a motor.

    Putting these components on the other side of the sense resistor, rather than directly across the motor terminals, messes up the sense voltage and gives me 100% innaccurate readings.

    I have ordered a number of zeners, schottkys, and a wide variety of ceramic capacitors and will test them each.
     
  6. gootee

    Senior Member

    Apr 24, 2007
    447
    50
    If the zener diodes are only conducting during the transients, is it possible that you are thinking the wrong way about their ratings? Just something to think about.

    And I thought that TVS (transient voltage suppression) zener diodes are only usually worried about how much energy they have to dissipate, not voltage per se. I could be wrong but I thought that their upper voltage spec is just how high their voltage might go at max current, while the lower one is where they start conducting and trying to limit the voltage rise. You have to keep the lower voltage spec above your normal operating voltages.

    Here is a pretty good paper about transient suppression in general, and also about TVS zener diodes:

    http://www.radiocad.com/_downloads/LoadDumpPaper-final.pdf

    I'm not sure if it would be applicable for your situation or not, but the P-FET switch is a cool idea (last schematic)! There are also commercial ICs that take the place of the rest of that circuit, and give added features (see maxim and linear.com; google "transient suppression" or somesuch). I think you can also put another p-fet upstream, by itself, with gate tied to gnd, to get reverse protection, in the last circuit shown.
     
    Last edited: Aug 23, 2010
  7. marshallf3

    Well-Known Member

    Jul 26, 2010
    2,358
    201
    I see your confusion about the zeners - remember that they will just have to handle the transient peaks and most are quite good at that - especially the ones designed to do so.
     
Loading...