In the attachment the capacitor will charge to 1250V. Applying a 5V pulse (or whatever the SCR wants) to the SCR gate will dump the capacitors charge into the transformer. When the cap has fully discharged I'm presuming the SCR will self-reset at 0V or when the SCR forward current is 0. If so, how do I repeat the process immediately? I could apply another pulse to the gate but I don't know exactly when to do that. How do I know when the pulse has completed and the SCR has reset? I would like to automate this somehow.

 

http://nptel.ac.in/courses/108105066/PDF/L-19(SSG)(PE) ((EE)NPTEL) .pdf
give several examples.

 

In the attachment the capacitor will charge to 1250V. Applying a 5V pulse (or whatever the SCR wants) to the SCR gate will dump the capacitors charge into the transformer. When the cap has fully discharged I'm presuming the SCR will self-reset at 0V or when the SCR forward current is 0. If so, how do I repeat the process immediately? I could apply another pulse to the gate but I don't know exactly when to do that. How do I know when the pulse has completed and the SCR has reset? I would like to automate this somehow.

I very much doubt this circuit will work as you expect. An SCR must reach 0V for it to turn off, and your circuit will discharge most of the capacitor but will never allow the SCR to reach 0V... either you change the principle of operation, or use forced commutation.

 

forced commutation?

 

forced commutation?

Check the attached document

 

Thank you, I'll give both a read. I thought this might be simple...

 

Wouldn't a IGBT of the correct voltage work? No matter what you use it needs to be of the correct voltage, not just the voltage your switching.

 

are you making an ignition system, like CDI??

google cdi triggering circuit.

 

The correct voltage would be the voltage I'm switching. My cap banks voltage is about 1200V, so I need an SCR rated for 1200V which I have. I'll take a look at a CDI, it might be close but I'm not experienced enough to understand it right now. I thought there would just be a general answer for repeating a pulse from an SCR and there kind of is with different commutation solutions, just nothing ever fits directly with what I'm doing.

 

only two ways to turn off an Scr, reduce current to zero, or short out the A/K terminals, cdi triggering uses the cap in series with the transformer, and dumps the cap across the coil, so the cap discharges and turns off the Scr.

 

only two ways to turn off an Scr, reduce current to zero, or short out the A/K terminals, cdi triggering uses the cap in series with the transformer, and dumps the cap across the coil, so the cap discharges and turns off the Scr.

Not quite true, if the current through the SCR falls below the SCR's holding current it will automatically turn off.

 

The correct voltage would be the voltage I'm switching. My cap banks voltage is about 1200V, so I need an SCR rated for 1200V which I have. I'll take a look at a CDI, it might be close but I'm not experienced enough to understand it right now. I thought there would just be a general answer for repeating a pulse from an SCR and there kind of is with different commutation solutions, just nothing ever fits directly with what I'm doing.

You could do it that way if you want, but what ever you use as a switch won't survive long. Components are never used at their maximum rated data sheet level. You always give some 'headroom' to them. Switching devices and capacitors at ~1 1/2 to 2 times working voltage.

Since your switching a coil, what about the snubber to absorb the kickback voltage? Formed when the coil secondary collapses. The coil in your car has ~12V on the primary from the battery when starting, but when it is running it has ~400V in the primary from the kick back.

 

Since your switching a coil, what about the snubber to absorb the kickback voltage? Formed when the coil secondary collapses. The coil in your car has ~12V on the primary from the battery when starting, but when it is running it has ~400V in the primary from the kick back.

I was going to loop back one of the secondaries as shown in the attachment. Otherwise I don't know how to choose a diode to put across the primary of T2. Its 1200V being stepped up 40 times by T2. Max amps will be about 1A from the capacitor bank being discharged into T2. I did use a UF4007 which helped a little but for some reason it did not work properly when I increased the size of the capacitor bank.

 

You can trigger the coil like this by putting the capacitor in series , same as CDI firing, as the cap discharges it turns off the SCR ready for the next pulse..

 

How would the circuit or SCR in general know when to figure again at the exact same voltage level? Thats the part I'm confused on. I'm looking to get a clean sawtooth wave like in the attachment.

 

You need to mesure the time it takes to discharge the cap ,then your firing circuit can retrigger after that time. Bigger voltage and bigger capacitor = bigger delay.

 

New question for you, kenw, What are you using for 'R6'? Standard resistors are limited to ~200V. If you are simulating this in software your values on your components may be giving you your problems. All components have limits of use, this is what I'm trying to get across to you. Your software is aware of this and you should be too.

 

Yes, just a standard resistor. Doesn't matter, it just drains the cap bank.

 

Yes, just a standard resistor. Doesn't matter, it just drains the cap bank.

Does matter when the cap bank is at 1250V !! If you build this circuit, I see flames or worse in your future.