Author Topic: Power Supply Short Circuit Detection and Shut-Off  (Read 7122 times)

0 Members and 1 Guest are viewing this topic.

Offline ShatteredRAMTopic starter

  • Contributor
  • Posts: 14
Power Supply Short Circuit Detection and Shut-Off
« on: May 06, 2014, 06:29:59 pm »
Hi all.

Over the last few weeks I have been playing with the idea of building my own bench power supply after see Dave's videos and looking at the rough price for a decent seeming bench power supply, after messing around with some samples of the LT3080 and deciding that I might want to use the LT3083, I was looking into other features that I might want, something that I was thinking about was a short circuit or overload detection circuit that could cut the power if it detected a short, now, I have the method that I am using to detect the short working perfectly, (I think.) all I need now is a way to shut off the voltage rail.


Mockup Short Detector Circuit; Please ignore crappy GENIE micro controller
Enviroment used is Circuit Wizard, I only used it because it is ok for quick testing as the simulation is completely graphical.
If the microcontroller detects that there is a dead short (<2Ohms), the light and the buzzer start pulsing. (Would be latched.)


I have experimented with using MOSFETs and other types of transistors but I haven't had much luck, my Electronics teacher hasn't been much help either.

so my question is, short (Pun not intended.) of using a Relay, what is my best be for switching off power to the rails when the micro-controller detects a short.

Any help would be appreciated, feel free to tell me that this is not a feature that would really be required, but I am just being paranoid.

Thanks in Advance.

ShatteredRAM

EDIT: For clarification, I am on my final year of my GCSEs, my electronics teacher always says that he knows the bare minimum, something that doesn't fill me with hope.
« Last Edit: May 06, 2014, 06:33:38 pm by ShatteredRAM »
 

Offline Thor-Arne

  • Supporter
  • ****
  • Posts: 500
  • Country: no
  • tinker - tinker, little noob.....
Re: Power Supply Short Circuit Detection and Shut-Off
« Reply #1 on: May 06, 2014, 06:47:38 pm »
IMO, if you have current limiting in your power supply you really doesn't need a short circuit detection.

Say you have a power supply with 3A output and the max control voltage for the current limiting circuit is set to 3A maximum output any short circuit will be limited to 3A or lower if you set it lower.

Any good bench power should survive a shorted output as they go into constant current mode.
 

Offline ShatteredRAMTopic starter

  • Contributor
  • Posts: 14
Re: Power Supply Short Circuit Detection and Shut-Off
« Reply #2 on: May 06, 2014, 06:57:00 pm »
If I was including a current limiting circuit, would it need to be adjustable?, or can I just have a limiting circuit set to 3A and no more?

EDIT: I'm a dummy, just read the datasheet again, the LT3083 includes a current limiting circuit, thanks for the help!
« Last Edit: May 06, 2014, 07:01:20 pm by ShatteredRAM »
 

Offline Jeroen3

  • Super Contributor
  • ***
  • Posts: 4195
  • Country: nl
  • Embedded Engineer
    • jeroen3.nl
Re: Power Supply Short Circuit Detection and Shut-Off
« Reply #3 on: May 10, 2014, 06:23:13 pm »
Yet still you might need Short circuit detection, or fusing characteristics.
I've designed a few credit card sized boards with two pairs of banana sockets and a 6 Volt power plug (for the extra C channel found on lab supplies) with a small Cortex M0 and a Relay.
The only thing the device does is comparing the voltage it measured before the relay turned on, to the voltage currently on the bus.
A typical short circuit condition gives 0 Volt with maximum Amps. So when the voltage drops below the % of the sensitivity knob for time set by delay knob, the relay turns off.
This method is capable of disconnecting load within 20 ms (too fast for inrush currents) from the time the voltage starts to drop. However, maximum current will be flowing for an unknown period of time due to capacitors used in lab supplies. But usually not enough to damage things.
They have tamed the magic smoke in my projects perfectly for over a year now.
 

Offline ShatteredRAMTopic starter

  • Contributor
  • Posts: 14
Re: Power Supply Short Circuit Detection and Shut-Off
« Reply #4 on: May 10, 2014, 08:58:42 pm »
Oh the joy of the complexity of electronics projects gradually rising.

Would I be able to get similar results using something like a digital (serial) voltmeter and the main control microcontroller linked to a relay, or would that be to slow? Would the circuit I had in the first post actually work better with less compexity?
 

Offline Bored@Work

  • Super Contributor
  • ***
  • Posts: 3932
  • Country: 00
Re: Power Supply Short Circuit Detection and Shut-Off
« Reply #5 on: May 10, 2014, 11:04:02 pm »
EDIT: I'm a dummy, just read the datasheet again, the LT3083 includes a current limiting circuit, thanks for the help!

The LT3083 also contains a fatal flaw. The difference between the output and the set pin must not exceed 10V. This becomes a problem for any "lab" supply designed for > 10V output, especially during power up, power down and fast load changes.

The LT3080 is even worse. I personally don't trust any regulator from that LT308x and LT309x family. Even if Linear markets them as a revolution in linear regulators.
I delete PMs unread. If you have something to say, say it in public.
For all else: Profile->[Modify Profile]Buddies/Ignore List->Edit Ignore List
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17211
  • Country: us
  • DavidH
Re: Power Supply Short Circuit Detection and Shut-Off
« Reply #6 on: May 12, 2014, 06:51:14 pm »
EDIT: I'm a dummy, just read the datasheet again, the LT3083 includes a current limiting circuit, thanks for the help!

The LT3083 also contains a fatal flaw. The difference between the output and the set pin must not exceed 10V. This becomes a problem for any "lab" supply designed for > 10V output, especially during power up, power down and fast load changes.

The LT3080 is even worse. I personally don't trust any regulator from that LT308x and LT309x family. Even if Linear markets them as a revolution in linear regulators.

This behavior, foldback current limiting, is deliberate and exists in other integrated voltage regulators like the LM7805 and LM317.  It restricts the power pass element to operating within its safe operating area and the LT308x and LT309x datasheets include the output current limit versus input to output voltage differential showing this.  Normal transistors have the same restriction to one extent or another and respond by failing unless protected.

It is easy enough to fix this with three external parts by adding a cascode to the input.
 

Offline Bored@Work

  • Super Contributor
  • ***
  • Posts: 3932
  • Country: 00
Re: Power Supply Short Circuit Detection and Shut-Off
« Reply #7 on: May 12, 2014, 08:31:29 pm »
No, it is not normal, and it has nothing to do with current limiting, foldback or otherwise.  |O If you think so, you work with a rather strange definition of current limiting: An IC blowing itself up in the name of current limiting. But really, may I recommend you don't play buzzword bingo with us, because it has all to do with crap IC design and Linear deciding to not telling the full truth.
I delete PMs unread. If you have something to say, say it in public.
For all else: Profile->[Modify Profile]Buddies/Ignore List->Edit Ignore List
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 17211
  • Country: us
  • DavidH
Re: Power Supply Short Circuit Detection and Shut-Off
« Reply #8 on: May 12, 2014, 09:51:35 pm »
If it is not normal then why do other integrated regulators do the same thing?  The LM340 series, 78xx series, 79xx series, and LM317 all do this to some degree or another.  At high input to output voltages, they limit their maximum output current to protect themselves via foldback current limiting.

The circuit function to restrict a pass element to operation within its safe operating limit operates the same way that foldback current limiting does.  In a discrete design, usually only the output voltage is used because the input voltage will be fixed but in an integrated design, the voltage cross the pass element is measured.

As to Linear Technology not telling the truth, it is right there in the data sheet characteristic curves and it is a well known feature with integrated regulators.
 

Offline Bored@Work

  • Super Contributor
  • ***
  • Posts: 3932
  • Country: 00
Re: Power Supply Short Circuit Detection and Shut-Off
« Reply #9 on: May 13, 2014, 04:42:07 am »
You still haven't got it, and continue to play buzzword bingo.
I delete PMs unread. If you have something to say, say it in public.
For all else: Profile->[Modify Profile]Buddies/Ignore List->Edit Ignore List
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf