High Voltage and High Current Load Switch/Relay

[virtualeast]

Hello All,

I'm trying to find a Relay, preferably Solid State, that can handle delivering and switching high voltage/High current to and from a bank of capacitors.

To simply put it, i am trying to charge multiple capacitors (concurrently) with one set (3 bank) of relays and switch between each capacitor bank to discharge into a Flash Lamp with another set of relays (3 bank). Just like charging "Low Charge" batteries and choosing between them for controlling a device.

The spec's are these:

Charge side:

500Vdc Power supply to charge each capacitor and shut off
when desired voltage reached.
7Amp charge current into each capacitor until shutoff.

Discharge side:

Max 500Vdc on the capacitor into the flash lamp head.
Large Surge current for short time. No more then ~25A/Sec.

So essentially, there are 3 capacitors, 3 charge control relays, and 3 capacitor bank selectors. The control system is a microcontroller based system which monitors the charge and bank selector with either 5 volts or 12volts though a transistor. I've found and tried to use a lower dc voltage based contact relay, but when the voltage exceeds 300vdc, the discharge feeds back into the control coil from the surge of energy rushing through and resets the control system and with prolonged use, destorys the relay. I also have an inline fast-blow fuse of 20A with the discharge into the lamp head and does not break (shows an inrush of less then 20amps/sec).

I've searched the internet far and wide and found little to nothing of these specs. Although I have found Mosfets that support this voltage and current but am having difficulties trying to find a soltion in finding a way to make Mosfets work like a relay or load switch. FYI, there are two relays per capacitor, one for charging and one for bank selection, so i cannot do lowside driving with any switching circuit (Seperating ground from the capacitor). Highside driving seems to be the solution but seems to be difficult as the capacitor voltage rises the source voltage rises which means the gate voltage has to follow (rated voltage of +- 30VGS).

I'll post up a few ideas i have with the mosfets once I draw them out.

I'm hoping someone has anyidea of how to do this unique idea of switching high voltage/high current and speaks out. Thanks for your help!!

 

[ccurtis]

My first thought, off the wall, is to use SCRs (which many solid state DC relays are made from), but perhaps use opto-isolators to isolate the control section from the capacitor section. The same for isolating the feedback from the capacitor section to the control section. Those tremendous surges are perhaps causing great ground loop currents, wrecking havoc on your control section. SCRs are capable of handling surge currents much greater than the continuous rating. Barring that, maybe a very substantial ground plane, or single point ground.

A detailed diagram, including how ground connections are made, speaks a thousand words.

As far as SSRs go, this may work:

https://www.crydom.com/en/Products/Catalog/s_sc.pdf

 

[virtualeast]

This is the idea that i had (See Picture). There is a voltage tripler that take the voltage Ac, Rectifies, and Triples it to Aprrox 500Vdc. I setup the gate to the mosfets so that the gate to source voltage doesnt exceed threshold. (+-20Vgs) though the voltage divider. Only problem is, when testing, it doesnt work. The voltage is there but its as if the capacitor or something is preventing it from being turned on. Another idea i had is to use a zener diode that's 12volts from the gate and the cap and a resistor as required but that also didnt work (you'd figure the cathode of the zener would follow the capacitors charge and the anode would always be 12v). Is there something that I am missing? Any thoughts, Ideas?

**broken link removed**

 

[ccurtis]

As far as your particular circuit goes, I can't offer much help. I have little top-of-the-head knowledge of biasing the various types of FETs. I see others here who can help with that easily, whereas I would have to hit the books.

What I do know is that you are dealing with high surge current and voltages which I would isolate from low voltage, sensitive control circuity as much as possible, if not entirely. SSRs are a good way to go for that reasons. Not to mention that the surge may actually be greater than you might expect if the capacitors are of good quality. The ionized gas of a flash lamp has low resistance.

Also, if the flash lamp is used for a photographic application the turn-on time is probably important, and can make-or-break a design approach. Gate capacitance, and inductances become important parameters.

My two cents.