# I am using a ZVS capacitor charger with an optocoupler

#### Rittter

Joined Dec 5, 2015
60
I am planning to use a zvs capacitor charger set at 250VDC to charge a capacitor. The capacitor, 330VDC, 95uF, is to be discharged via an optocoupler which is controlled by a 555 timer at 1Hz. The optocoupler (undetermined as yet) will discharge the voltage from the capacitor to a coil going directly to ground to generate an EMF. The opto. and the coil will have minimal resistance. I have a couple of concerns.
1) At start up of this circuit for example, the cap is at zero volts, and the opto. is energized. At that instant, could this create a dead short through the circuit and burn out both the zvs and the opto. ? I have seen circuits and schematics which have no provisions to address this concern.
2) Power and heat dissipation. It is my understanding that there is no current flow when the cap. is charging, however, at discharge, there is current. This would necessitate knowing what the current will be and the need for some type of heat sink for the opto. Again, I have seen circuits etc. delivering these levels of voltage through capacitors without employing heat sinks etc.

Any help would be greatly appreciated.

#### Alec_t

Joined Sep 17, 2013
12,744
The opto. and the coil will have minimal resistance.
Optocouplers are generally low current devices, but some can handle a few Amps. Do you have a particular one in mind?
At start up of this circuit for example, the cap is at zero volts, and the opto. is energized.
If the cap is at zero volts it can't drive any current through the opto. If the opto is energised then the cap won't charge.
It is my understanding that there is no current flow when the cap. is charging
There is charging current (but not through the opto).

#### crutschow

Joined Mar 14, 2008
29,528
1) At start up of this circuit for example, the cap is at zero volts, and the opto. is energized. At that instant, could this create a dead short through the circuit and burn out both the zvs and the opto. ?
A typical way to discharge a capacitor in such an application would be for the opto to fire a large SCR series with the capacitor and load.
To allow the SCR to turn off after the discharge you may need a small inductor in series with the zvs supply.
2) Power and heat dissipation.
Charging 95μF to 250V at a 1kHz rate requires an average current of 250*95μF*1kHz = 23.75 amps.
Is your zvs supply capable of delivering that?
That average current would mean the SCR would need to be mounted on a heat sink.

#### Rittter

Joined Dec 5, 2015
60
I agree, optocouplers are good for a few Amps. These opto's are just the nuts. I am looking at a few, but one in particular is CPC1988. Peak current is 10Amps, which when pulsing is a consideration. I try to stay away from the peaks.
If the cap is at zero and the opto is energized, the zvs turns on, doesn't that provide a open pathway to the coil and ground = Short ?

#### Rittter

Joined Dec 5, 2015
60
A typical way to discharge a capacitor in such an application would be for the opto to fire a large SCR series with the capacitor and load.
To allow the SCR to turn off after the discharge you may need a small inductor in series with the zvs supply.
Charging 95μF to 250V at a 1kHz rate requires an average current of 250*95μF*1kHz = 23.75 amps.
Is your zvs supply capable of delivering that?
That average current would mean the SCR would need to be mounted on a heat sink.
It is being run at 1Hz not 1kHz.

#### crutschow

Joined Mar 14, 2008
29,528
It is being run at 1Hz not 1kHz.
Sorry.
One of these days I'll learn to read.

#### Rittter

Joined Dec 5, 2015
60
Sorry.
One of these days I'll learn to read.
Not a problem. I do it myself all the time. Thats a nice equation for Average Current. What do you have for Peak Current ?

#### crutschow

Joined Mar 14, 2008
29,528
What do you have for Peak Current ?
The peak current is just the peak capacitor voltage divided by the circuit resistance (capacitor ESR, switch, wire, load coil).
If the coil has significant inductance, then that needs to be included in the equation.