Connect with us

Help needed on simple short circuit protection

Discussion in 'General Electronics Discussion' started by mattomatto, May 18, 2016.

Scroll to continue with content
  1. mattomatto

    mattomatto

    8
    0
    May 18, 2016
    Hi all,

    I'm trying to add some simple short-circuit protection to my design to make it safer. Some contacts are exposed which could potentially short, and I'm looking for an effective way to prevent that from damaging components or being dangerous.

    I have attached a picture of a simplified schematic. The control board has three potentiometers attached to power and ground, with their wipers being read by a micro controller. The switch (sw) represent where the short-circuit could happen and the '?' is where I'm looking to place a solution.

    I have looked into using a resistor in-line with the power (something like 10k so that it doesn't get too hot) however, this affect the potentiometer reading to the point where it is unusable.

    I'm also looking at an opamp buffer based solution, but so far what I have tried has also had an effect on the potentiometer ready, perhaps because ether voltage output is never fully 3v, as the source supply is.

    I hope that all makes sense and let me know if you need any more info. Help much appreciated!
     

    Attached Files:

  2. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,501
    2,841
    Jan 21, 2010
    insulate things so you don't accidentally short it out.
     
  3. mattomatto

    mattomatto

    8
    0
    May 18, 2016
    That's not an option in this case, due to the product design.
     
  4. Alec_t

    Alec_t

    3,050
    830
    Jul 7, 2015
    That's not what your diagram shows :confused:.
     
  5. mattomatto

    mattomatto

    8
    0
    May 18, 2016
    The diagram shows three pots, each connected to ground and to 3v through the '?' which as stated above represents where I think the short circuit protection should go.

    Does that make sense? Not sure how I can present it more simply, sorry
     
  6. hevans1944

    hevans1944 Hop - AC8NS

    4,646
    2,167
    Jun 21, 2012
    What happens to this ill-designed product if the power to the potentiometers is "accidentally" shorted to ground? Does it fall out of the sky? Will people be injured? Will there be a gigantic high-current arc through the short from the unspecified 3 V DC power rail? How fast must the circuit in the "?" box operate once a short is detected?

    Easy way to do this is to place a relay wired with self-latching, normally-open, contacts between the power applied to the potentiometers and GND. Wire a normally-open push-button switch across the normally-open relay contacts. Connect one side of the switch to the 3V rail and the other side to the relay coil. Connect the potentiometers to the relay coil. Connect the other end of the relay coil to GND. Use a relay with a 3 V coil. Reset the circuit and power up the potentiometers by momentarily pressing the push-button switch to energize the relay coil, closing the normally-open contacts and latching 3 V power to the coil and potentiometers. If a short occurs (and lasts long enough), the relay will de-energize, the normally-open contacts will open, and the open contacts will remove 3 V power to the potentiometers. Reset after clearing the short by momentarily pressing the push-button switch.
     
  7. Herschel Peeler

    Herschel Peeler

    401
    66
    Feb 21, 2016
    At 3 Volts there isn't much room for a current limiter circuit. A fuse?
    Do you have a cost limit in mind? There are current sensor circuits and chips. If you had that turn off a MOSFET in series with the power line that would get the loss down. How much current is normal? At what point would you like it to turn off?
     
  8. mattomatto

    mattomatto

    8
    0
    May 18, 2016
    Thanks for your replies

    A little rude, lots of products have open contacts. For example: This, this, and this. The worst case scenario is that some components on the main board are damaged. This is about CE/FCC certification and making a product as safe and resilient as possible. A relay is and reset switch is overkill for this simple circuit in my opinion.

    A fuse would need replacing so I'm keen to avoid something like that. Ideally it would consist of a few small smd components that don't add more than a pound or two to the BOM, hence looking at opamps and resistors. I'm not aware of current sensor chips - I'll check those out, thanks! Current is around 10mA normally.

    Nothing involving an opamp based buffer spring to mind? I'm sure it's easily achievable, just escaping me at the moment.

    Thanks,
     
  9. Herschel Peeler

    Herschel Peeler

    401
    66
    Feb 21, 2016
    Starting rough draft. Design details depend on current levels that are normal and what you want to call normal. Too many parts?
     

    Attached Files:

    Arouse1973 likes this.
  10. mattomatto

    mattomatto

    8
    0
    May 18, 2016
    ooh, Nice one! So that's a current sensor and a comparator that switches a mosfet on/off.

    That's very logical. I do feel there's an even simpler way but that's a good starting point.
     
  11. Herschel Peeler

    Herschel Peeler

    401
    66
    Feb 21, 2016
    Not intended to be anything other than a starting point. Another problem is the source of power to the chips. Is 6 Volta available?
    A simpler but slower solution is a self resetting fuse. At a certain current (heating point) they pop into a high resistance, then return to normal low resistance when the problem goes away. The problem is they do not respond very quickly. See "Poly Fuse".
     

    Attached Files:

  12. hevans1944

    hevans1944 Hop - AC8NS

    4,646
    2,167
    Jun 21, 2012
    Rude? I am sorry you took it that way, but this sounds like a re-fit to an existing product, ergo, the original design is in some manner defective, or least incomplete. I am aware that even though you are talking about minuscule currents (10 ma?) driving the three paralleled potentiometers, depending on what they control some serious damage could occur elsewhere if any of their active terminals (supply and wiper) were shorted to ground. Does that about sum it up? You want power to the potentiometers to go away if their excitation happens to get "accidentally" shorted to ground?

    QUOTE="mattomatto, post: 1694173, member: 44314"]I'm sure it's easily achievable, just escaping me at the moment.[/QUOTE]
    When you capture that easily achievable solution please post that bad boy here.

    Is that a pound sterling or a pound weight? There is a current sensing SMD component that consists of wire traces passing over a Hall detector. The output of the Hall detector is conditioned to produce an analog output signal proportional to the current and its polarity. This is the analog equivalent of a magnetically operated reed switch, except it produces a continuously variable analog output instead of a contact closure. Not as simple as a reed relay but smaller and more versatile.

    These Allegro devices (there are other manufacturers) are intended to measure significant currents (amperes) in power applications. One member here is using them in his woodworking shop to power-on a sawdust vacuum collection system whenever he turns on any of his power woodworking tools. The analog output of the current sensor interfaces to a Microchip PIC that then controls a solid-state relay to energize the vacuum system, and also to provide a timed turn-off delay to clear sawdust from the collection pipes when the powered woodworking tool is turned off.

    Depending on how much short-circuit current is available from your 3 V rail, one of these Allegro devices could probably sense the fault condition, convey that information to a PIC in an SMD package, which PIC would then remove the voltage to the potentiometers by turning off an SMD MOSFET series pass transistor, positioned between the 3 V rail and the potentiometers. The whole circuit would fit in an area about the size of your thumbnail. Response time would be on the order of a millisecond or less.

    We get a lot of newcomers here who have absolutely no practical knowledge of electricity or electronics, yet they want to do some incredibly dangerous things. In this case, I needed to know what the implications of a short circuit would be. Short circuits are normally considered to be dangerous things. Apparently the worse-case scenario is some other components might get damaged because of the lack of control from the shorted potentiometer(s), or perhaps overloading and loss of the 3 V supply rail. Regarding your posted images, IMHO safe products do not have exposed electrical contacts. Safe encompasses danger to people as well as danger of damage to equipment. Although a reed relay is about as simple as it gets, it would require manual resetting and occupy more space than the solution outlined above. So, I would investigate the Allegro device with a PIC and a MOSFET, all in SMD packages mounted on your circuit board with programming to auto-magically reset according to criteria you establish. No op-amps, and only a few SMD resistors and bypass capacitors required.:D
     
  13. BobK

    BobK

    7,682
    1,688
    Jan 5, 2010
    Which short are you worried about? If it from 3V to ground, that can be handled in the power supply by current limiting. If it is between the wiper and 3V or ground, that can be handled by a fixed resistor on one side or the other.

    Bob
     
    hevans1944 likes this.
  14. mattomatto

    mattomatto

    8
    0
    May 18, 2016
    Nice, I will look into an SMD poly fuse :) I had noticed those other devices run from 6v, but perhaps there are low power versions.

    Thanks, there is an arm chip already on board with a couple of spare pins, I'll look into a suitable allegro device :)

    It's between 3v and ground. The power comes from 2xAAA batteries, so I have tried a resistor in line with the 3v coming onto the control panel, but it's been affecting the pot readings, due to the ADC reference being different, I presume. Perhaps two resistors, one inline with incoming 3v and other inline with the incoming ground, to keep the voltage the same?

    Thanks all
     
  15. BobK

    BobK

    7,682
    1,688
    Jan 5, 2010
    I think you are over worrying. A temporary short from 2 AAA batteries is not likely to do much harm. If anything, it is the batteries that would be damaged.

    Have you seen cases where this short has destroyed something, or are you just speculating?

    Bob
     
    (*steve*) likes this.
  16. mattomatto

    mattomatto

    8
    0
    May 18, 2016
    Thanks for the reply Bob

    I'm used to working with LiPoly batteries, in which case a short would cause a spark and would blow a couple of components. I'm trying to take the same precautions for AAA to ensure I can meet CE/FCC regs. I plan on covering all components and contacts in conformal coating, except for these few exposed ones that the design relies on being exposed.

    Do you have experience in CE/FCC and think I'm being over cautious?
     
  17. BobK

    BobK

    7,682
    1,688
    Jan 5, 2010
    No, I have no such experience. But I have never seen anyone worry about a possible short across the terminals of a potentiometer.

    If this is a product, it is going to be enclosed no? How do expect something to short inside your closed box?

    And, LIPOs are a different story. They need short protection because they will catch fire otherwise. The short protection is typically built right into the battery pack.

    Bob
     
    hevans1944 likes this.
  18. mattomatto

    mattomatto

    8
    0
    May 18, 2016
    It will not be enclosed, (Similar to this product) hence trying to make it as safe as possible so it will conform to regs.
    I decided to abandon lipo's due to other CE/FCC regs to do with battery protection being an issue.
     
  19. Proton

    Proton

    5
    2
    Sep 8, 2009
    What about a 2 cell flashlight bulb? A PR6 would seem a good choice. Resistance is quite low when cold, so wouldn't affect your values. if dead shorted, worse case is bulb lights up - could even be labeled "Overload! RUN for your life"
     
    CDRIVE and hevans1944 like this.
  20. hevans1944

    hevans1944 Hop - AC8NS

    4,646
    2,167
    Jun 21, 2012
    Without knowing anything about CE/FCC certification, and the testing protocols for compliance with same, it is impossible to recommend a solution. However I am sure there are laboratories you can hire who will do just that and explain why your product passes or fails. I would take my fishing gear and go troll in those waters for a solution.
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-