Folks,

I don’t understand SCRs very well. That’s why I’m posting to this thread, you guys are well-versed at SCRs. I’m simulating a circuit that will be generating high-voltage pulses – sharp negative spikes from 100V to 0V and recharge back. Unfortunately, I can’ explain the simulation results that I’m getting:
* The red trace is the output. The green trace is the gate of the SCR. Why the output doesn’t fall at the first gate pulse?
* Why the output falls to 40V instead of all the way down to 0.7V?
* What could be the cause of these problems?
* Am I using the wrong component for the SCR?

Any enlightenment would be appreciated!

Thanks,
- Nick

 

Your R2 is 1 Ohms. The SCR is only rated for 1.5A. If you increased R2 to 67 Ohms, the SCR would be able to sink enough current to get the voltage down.
R=E/I, so R2 = 100V/1.5A = 66.6666667 Ohms.

However, you will not be able to turn the SCR off without turning off the power supply.

Once the gate of an SCR is triggered, it stays conducting until the current through it stops.

You should instead consider using a P-channel MOSFET.

BTW, with 1.5A flowing through R2 when it's 67 Ohms, it'll be dissipating 150W of power. Right now it should be at around 100W. The other 50W is being dissipated in the SCR. Were this build from real components instead of a simulation, you would've heard a loud bang, seen a bright flash and lots of smoke.

 

What a coincidence, I was just wondering how a pulse-triggered SCR setup worked.

If you look at patent #4455510, on the first page, you see a schematic that sends a series of pulses at an SCR gate to turn it on and off rapidly during the second half of the part of the AC cycle where it is positive. I haven't the foggiest idea how it works, but we've got one of these power supplies here, right in front of me, and I put a scope on the SCR. It definately does work. It triggers the SCR on and off repeatedly some how during the lamp startup. Any thoughts? I'm curious as to how it works.

 

BTW, here's a waveform recorded from a gate of an SCR in a 1977 Panametrics pulser. The width of the spike in the beginngin of the hump is 9ns or less.

- Nick