Reverse Polarity Protection for Open-collector Output

Discussion in 'General Electronics Chat' started by eepty, Apr 30, 2014.

  1. eepty

    Thread Starter New Member

    Sep 11, 2012
    6
    0
    I am designing a MCU controlled open-collector digital output (MCU control the base of transistor as a switch, output low). The output is assumed max. 26Vdc, 85mA.

    As transistor has max. reverse Vce of about 5V, I would like to add a reverse polarity protection for the output. In the other words, my device has no problem even the user collected the plus and negative wire wrongly.

    I have considered some solutions:
    1. Simply add a forward bias diode in series with the output. The problem is that it will drop ~0.7V at the output.

    2. Add N-channel mosfet at the low side of the output so that the mosfet will switch off when the polarity is reversed. The problem is the Vgs is not large enough to switch the mosfet fully on (When the transistor switch on, the collector will pull to ~0.1V)

    3. Add a reverse bias diode parallel to the output. The problem is it will short circuit when the output polarity reversed. The diode will burn eventually.

    (The schematics attached)
    (I have a T.V.S at the output)

    Could you suggest other solution?

    Thank you.
     
  2. ScottWang

    Moderator

    Aug 23, 2012
    4,853
    767
    I will choose diode.
    If the load is relay even you can try to use rectifier diode bridge, but it depends on what is your load, as relay, led, etc, ... it will related about the voltage range of load, since the relay can be accept a big range of voltage, but led can't, if the current too small then the led brightness will be too dark.
     
  3. ErnieM

    AAC Fanatic!

    Apr 24, 2011
    7,386
    1,605
    The MOSFET suggestion cannot work, look at it again and note where the gate is going. ;-)

    Either of the diode methods are fine. IC's use the parallel method (to both + and - rails) on most pins for ESD protection.
     
  4. inwo

    Well-Known Member

    Nov 7, 2013
    2,433
    315
    #3 with an in-line fuse seem best. IMO
     
  5. alfacliff

    Well-Known Member

    Dec 13, 2013
    2,449
    428
    why not just use a higher voltage rated transistor? 5v ce isnt a very robust transistor, a 2n2222 type has a much higher rating and is available in lts of styles.
     
  6. AnalogKid

    Distinguished Member

    Aug 1, 2013
    4,528
    1,248
    Alfa - as stated in the OP, the issue is the reverse Vce, not the forward Vce.

    eepty - You probably won't like this, but consider a bridge instead of the single series diode. Bad news, two diode drops. Good news, circuit doesn't care about output polarity, and works either way. If you can stand a small leakage current and more cost, use four shottkeys and you're back to the equivalent of one normal diode.

    ak
     
  7. alfacliff

    Well-Known Member

    Dec 13, 2013
    2,449
    428
    if you look up the specs on the 2n2222, you will find that the bvrce, reverse collector breakdown voltage is much above 5 volts., same as the breakdown voltage from collector to base, reverse.
     
  8. crutschow

    Expert

    Mar 14, 2008
    13,000
    3,229
    In what data sheet did you see that? :confused: The reverse colector-emitter voltage breakdown (where the collector-base junction is forward biased) is determined by the reverse base-emitter breakdown and that's typically 5V.

    Edit: I would use a Schottky diode in series for protection. Its forward biased voltage at low currents is only about 0.4V.
     
  9. alfacliff

    Well-Known Member

    Dec 13, 2013
    2,449
    428
  10. ian field

    Distinguished Member

    Oct 27, 2012
    4,413
    782
    For a parallel protection diode, a Shottky-barrier diode has lower VF, although I'm not sure if that would contribute anything worthwhile - and you have to watch reverse breakdown voltage, SB diodes start at 20V, 60V is getting towards exotic!
     
  11. alfacliff

    Well-Known Member

    Dec 13, 2013
    2,449
    428
    and who would deliberatly put reverse voltage on an open collector circuit with the output active? back emf from inductors would not do it, because the output would be grounded by the transistor till the output opened, at which time the transistor would not be biased on.
    a series diode would me the more realistic way to do it.
     
  12. crutschow

    Expert

    Mar 14, 2008
    13,000
    3,229
  13. alfacliff

    Well-Known Member

    Dec 13, 2013
    2,449
    428
    why would anyone have a reason to apply reverse collector to an open collector output when the transistor is turned on? the only posted voltage limits are for when the transistor is turned off. the circuit is supposed to turn on with the right polarity. the only open collector type circuits I have seen are the squelch in cheap scanners, the collector is connected between two caps in series with the audio before the volume control and the transistor is turned on and off to squelch the noise. but they use no collecor supply, just audio.
     
  14. eepty

    Thread Starter New Member

    Sep 11, 2012
    6
    0
    I cannot tell exactly what is the output, but the requirement I received is a "digital output". I assume it can be 3.3V or more. At logic low level it has to be close to ground (0V), so I think the diode voltage drop is not good enough. (For shottky diode although the typical voltage drop is 0.4V, it can be more when the forward current become high).
     
  15. eepty

    Thread Starter New Member

    Sep 11, 2012
    6
    0
    One solution seems good is using the parallel diode to protect the bjt from over reverse Vce, and add a polyfuse in series with the output to protect the circuit when the diode is forward bias (that is when the polarity is reversed).
     
  16. ronv

    AAC Fanatic!

    Nov 12, 2008
    3,282
    1,236
    How much current does the transistor need to drive?
     
  17. #12

    Expert

    Nov 30, 2010
    16,261
    6,770
    The OP says:

     
Loading...