Author Topic: Crowbar Circuit  (Read 8956 times)

0 Members and 1 Guest are viewing this topic.

Offline RerouterTopic starter

  • Super Contributor
  • ***
  • Posts: 4700
  • Country: au
  • Question Everything... Except This Statement
Crowbar Circuit
« on: August 04, 2012, 12:49:23 am »
My circuit under normal conditions feeds up to 35V to this circuit, and the components behind it have a maximum voltage of 40V,
over 36-39V, i am hoping this circuit will pull the voltage down by dumping a lot of current, sort of like a crude shunt regulator, and trigger a polyfuse at the same time,

what i want to know is if this seems feasable to you as an SCR crowbar would not be suitable for a polyfuse, as it would be jerking on and off,

the mosfet in question will have a 2.5V threshold, and i am well aware of body diodes,


« Last Edit: August 04, 2012, 12:54:17 am by Rerouter »
 

Offline PeteH

  • Supporter
  • ****
  • Posts: 117
  • Country: ca
Re: Crowbar Circuit
« Reply #1 on: August 04, 2012, 02:43:02 am »
Why not build a protection scheme that either clamps the voltage with a pass FET (kind of dangerous) or switch off the input above a certain voltage to protect your sensitive ICs?
 

Offline amspire

  • Super Contributor
  • ***
  • Posts: 3802
  • Country: au
Re: Crowbar Circuit
« Reply #2 on: August 04, 2012, 03:28:12 am »
The FET circuit is not a real crowbar circuit. The crowbar works by turning hard on, and the crowbar SCR can sustain this as it has a low voltage drop across it. The FET circuit will have over 33V across the FET, so it can dissipate a large amount of power. It will help limit external inductive voltage spikes, but if there is a chance of a sustained overvoltage (like protecting the load from a shorted power transistor in a power supply), a true crowbar circuit + fuse/polyfuse will do a much better job then this FET circuit.
 

Offline RerouterTopic starter

  • Super Contributor
  • ***
  • Posts: 4700
  • Country: au
  • Question Everything... Except This Statement
Re: Crowbar Circuit
« Reply #3 on: August 04, 2012, 03:36:44 am »
for reference its an isolated linear section of a power supply,

if someone hooks on a larger voltage at the output, the capacitors of that section will charge past the 40V, while only pulling a few mA, until a few components begin to breadown,

though the fet on the output isnt too bad an idea, actually quite ideal, as when running input is a few volt above output, to lift it up with, and when its off, its too low so it stays open circuit... hmm, even if it leaks, the other circuit can help it along,

edit: only major downside is i loose output accuracy due to rson at higher currents,
« Last Edit: August 04, 2012, 04:07:49 am by Rerouter »
 

Offline T4P

  • Super Contributor
  • ***
  • Posts: 3697
  • Country: sg
    • T4P
Re: Crowbar Circuit
« Reply #4 on: August 04, 2012, 04:03:11 am »
There's a reason SCR's were always used ... they are so called switchable "diodes" or part thereof their main uses
because they latch on
 

Offline SeanB

  • Super Contributor
  • ***
  • Posts: 16362
  • Country: za
Re: Crowbar Circuit
« Reply #5 on: August 04, 2012, 06:03:48 am »
You would probably be better off using a transorb of 39V. It will clamp a lot faster, as the one you give is unfortunately very slow to turn on, due to the gate capacitance. That will let spikes and such through.
 

Offline RerouterTopic starter

  • Super Contributor
  • ***
  • Posts: 4700
  • Country: au
  • Question Everything... Except This Statement
Re: Crowbar Circuit
« Reply #6 on: August 04, 2012, 08:05:14 am »
hmm, you where very correct with the gate capacitance,

i will have to respin my design but i think i will have to settle for an scr, was hoping of avoiding one, but now that the first posters suggestion of the mosfet has helped me isolate the output, i only having to worry about input voltage so i thank you all for your time, and your knowledge :)
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf