I was just wondering if anyone knew of methods for high power circuit protect. Specifically regarding capacitor discharge circuits. I'm assuming I would want some type of diodes placed between the capacitor and the load to prevent a feedback current damaging the capacitors but I am unsure of how to select the correct diode as the amount of current I would theoretically expect is much high than any affordable diode would be able to handle according to their data sheets.

Considering the circuit with the protection diodes would only be active during the discharge period, which would be a very short amount of time, can I get away with using a diode with a much smaller current rating? It seems like most of the data sheets only list the constant current rating. Is there a way to calculate how much current the diode can handle during a short pulse period?

 

Impulse generators that discharge capacitors at rates of hundreds of volts per micro-second do not normally employ diodes within the discharge path – but it is important that the rate of voltage change does not exceed the specification for the capacitors.

Looking at a data sheet for 1N400X diodes (with a nominal current rating of 1A), gives a figure of 30A for 8.3ms.

While many diode data sheets may not list such information – clearly diodes can withstand being subject to overcurrents an order of magnitude greater than their nominal rating for short periods. I suspect much like many components, the limiting factor is the resultant power dissipation – with the heat destroying the component.

 

Thank you Hymie!

So if I knew the current through the discharging part of the circuit would be around 1kA, could I place like 20 diodes rated for 50 surge amps in parallel to distribute the power? In reality I would probably use more to have a safe tolerance if this is something that would actually work.

Also, I'm only worried about feedback current because the capacitors would be discharging through a large coil with a core and I've been told by an EE professor that a return current could pontentially be an issue.

 

If there is ringing in the circuit – causing a back emf that you are concerned at, optimising the circuit values to minimise any back fed voltage maybe the way forward.

One problem with paralleled diodes is that those with a lower forward voltage drop will pass a greater share of the current. This will result in those diodes heating more, reducing their forward voltage drop further – passing an even greater share of the current. This will result in those diodes heating even more …..and so on.