2
\$\begingroup\$

This should be fairly simple but I can't find any satisfactory answer on the web (I'm quite new to this things and I don't really know where to look for). I have two sensors in my circuit, one working at 12VDC and the other at 5VDC. The power supply is 12VDC and in my circuit I have a DC/DC converter bringing my 12VDC to 5VDC:

  • Murata Power solutions OKR-T/3 Series

I would like to add some protection for the sensors (they are quite expensive). I would like to protect them from over-voltage supply and negative voltages. The only thing I have found is:

  • Voltage detector Linear Technologies LTC4360ISC8
  • Voltage detector Linear Technologies LTC 4365HTS8

But they only have SC-70 packages and it's simply to hard to solder them and test them

\$\endgroup\$
13
  • \$\begingroup\$ What about a crowbar? \$\endgroup\$
    – PlasmaHH
    Commented Nov 13, 2015 at 12:41
  • \$\begingroup\$ How can you wire a sensor with inverse polarity? Even if you build a protection circuit, you still have to wire the sensor correct onto the protection circuit, because the circuit can't detect if you swaped wires. As @PlasmaHH suggested a crowbar circuit for overvoltage, still I had never faced a seroius PSU from outputing overvoltage, so... \$\endgroup\$
    – Marko Buršič
    Commented Nov 13, 2015 at 13:25
  • \$\begingroup\$ Interesting, I was not aware of this kind of things. I looked it up very quickly on wikipedia and they say: "It operates by putting a short circuit or low resistance path across the voltage output". But I don't really like having a short circuit on my power supply either. \$\endgroup\$
    – Worldsheep
    Commented Nov 13, 2015 at 13:28
  • \$\begingroup\$ @Worldsheep: Fuses, you know... \$\endgroup\$
    – PlasmaHH
    Commented Nov 13, 2015 at 13:30
  • \$\begingroup\$ I am not worried about the power supply but mostly about the "Murata Power solutions OKR-T/3 Series". But if you tell me that there is no reason to be worried, I could just drop the protection circuit \$\endgroup\$
    – Worldsheep
    Commented Nov 13, 2015 at 13:30

1 Answer 1

Reset to default
2
\$\begingroup\$

If you really want to protect your sensors, then shorting the output of the Murata DC/DC converter is both the easiest, simplest, and safest way. If there is an overvoltage condition, then there are capacitors charged to that potential on the output of the power supply, and that energy is either going to into your sensors and destroy them, or it can go partly into your sensors and something else (and still destroy your sensors) or it can be consumed by a dead short so quickly that the over voltage condition doesn't have time to hurt your sensors.

A buck converter like the one you're using tolerates output shorts very well, and it simply has to not turn on one MOSFET (the top one) to survive a short indefinitely. Which it certainly will do. Most buck converters with over voltage protection in fact short the output through the synchronous MOSFET until the voltage falls so it is back within regulation, so it's definitely the standard and most reliable way of doing this.

Use a thyristor/SCR crowbar circuit. It's thousands of times faster than the fastest fast blow fuse there is. You can use a zener or zener reference like the TL431 to set a very sharp trigger voltage. The DC/DC converter will quickly shut down as it's overcurrent protection is triggered before any damage comes to your sensors, it, or the crowbar.

As for negative voltage, put a diode in reverse across the power rails. If instead the power supply is hooked up backwards, 5V to ground and ground to 5V, the diode will turn on and conduct like a short, which will engage the Murata DC/DC converter's overcurrent protection yet again and hopefully save the sensors. A fast schottky diode would work best here.

\$\endgroup\$
5
  • \$\begingroup\$ Thank you very much for the very detailed question, I'll have a look in detail in the next days. At the moment one question comes to mind. What happen if the over-voltage is not a transient condition but rather a constant over-voltage? The current will rise until the fuse in the power supply breaks, right? \$\endgroup\$
    – Worldsheep
    Commented Nov 13, 2015 at 16:36
  • \$\begingroup\$ Second question, do you have a link showing a SCR crowbar circuit using the TL431? \$\endgroup\$
    – Worldsheep
    Commented Nov 13, 2015 at 16:56
  • \$\begingroup\$ ti.com/ods/images/SLVS543O/ai_crowbar_lvs543.gif \$\endgroup\$
    – metacollin
    Commented Nov 13, 2015 at 19:09
  • \$\begingroup\$ Also, the over voltage will always be transient, whether it likes it or not. That's another reason that a short is employed - a dead short will trigger whatever fault protection is available further upstream, or in the case of a synchronous buck converter like the one you linked, it actually CAN'T maintain any useful amount of voltage on the output if it is shorted. So I really see no way it could not be transient. Even if you intentionally hooked up a bench power supply set to 30V, the OVP would trigger and short it, and it would have no choice but to drop its output to a few hundred mV. \$\endgroup\$
    – metacollin
    Commented Nov 13, 2015 at 19:13
  • \$\begingroup\$ What I don't understand is how come I don't get huge currents when I have a short circuit. Could I do the same just after the power supply? (I have a second sensor plugged without buck). And thank you for the circuit image. One last question, how do I choose the resistor values? \$\endgroup\$
    – Worldsheep
    Commented Nov 13, 2015 at 21:04

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.