High(ish) Voltage Regulator

[KingKrak]

Hey,
I'm building a circuit that is powered off a rectified 3 phase supply (when rectified is about 600V). The control part of the circuit is 5V.

I'm having a few problems figuring out how to easily provide a 5V power source to the control circuit. Normally I would just use a 5V regulator, but normally I have power supplies that are 12V to 24V.

I can't use a transformer (into one of the AC phases) because of tight size and cost restrictions. I thought about using a zener diode in series with a resistor across the 600V load, but the control circuit will probably draw around 100mA and the zener probably won't be able to handle that.

I'm sure somebody out there knows a better way to do this...

 

[crutschow]

For low currents you can use a high voltage, film capacitor in series with the AC power line feeding rectifiers and a shunt regulator. But this will not isolate the circuit from the line so is a hazardous circuit and it must have an interlock so no one can access the circuit while power is applied.

For isolation you must use a transformer.

 

[KingKrak]

Isn't a zener diode a shunt regulator?

I think I have figured it out. If I add an emitter-follower stage from a transistor I can use the transistor to drive the load - thus leaving the zener unloaded.

Thank you once again wikipedia...

Linear regulator - Wikipedia, the free encyclopedia

 

[Mr RB]

Hmmm.... 600v * 100mA.

I wonder what mr Ohm might think about your transistor heat.

 

[Styx]

Hey,
I'm building a circuit that is powered off a rectified 3 phase supply (when rectified is about 600V). The control part of the circuit is 5V.

I'm having a few problems figuring out how to easily provide a 5V power source to the control circuit. Normally I would just use a 5V regulator, but normally I have power supplies that are 12V to 24V.

I can't use a transformer (into one of the AC phases) because of tight size and cost restrictions. I thought about using a zener diode in series with a resistor across the 600V load, but the control circuit will probably draw around 100mA and the zener probably won't be able to handle that.

I'm sure somebody out there knows a better way to do this...

Use a FET as a regulator. A great circuit exists in "The Art Of Electronics"

The art of electronics - Google Books

I actually used a modified version of this circuit (components changed and the OPAMP referenced changed to be a Zener with some high-power/voltage resistors as a burn). I also added a crude close-loop feedback system to ensure tighter regulation of the output 15V (via another zener and a bjt to shunt the main zener)

It had to provide 15V from a "supply" of 30V -> 750V (nominal was 350V).
This works very well. I was pulling ~15mA from it and a TO220 FET was bolted to some aluminium nearby and it experienced a temprise of ~7C settled out after 20min at 600Vdc

For higher power you might want to concider paralleling up the FET's to spread the power disipation between 5 or 6 such devices

 

[crutschow]

If you don't mind dissipating the power, you could also use a power resistor in series with a shunt regulator (yes, a Zener diode to ground makes a shunt regulator). This avoids having to use high voltage transistors.

 

[tcmtech]

So how do you plan to dissipate 60 watts of heat with a power resistor in a space smaller than a transformer that has to only handle a .5 watt load?

A 1.5 VA transformer is very small and would easily fit inside a 1.5 inch cubed space. A 60 watt resistor and or heat sink is many times larger than that!

 

[giftiger_wunsch]

I agree with tcmtech; trying to dissipate 60W in that way is absurd; consider for a moment the amount of heat a 60W lightbulb produces...

 

[ronsimpson]

This product is a PWM with 700 volt FET. It could buck 600 volts down to 5 volts 100mA.
https://www.electro-tech-online.com/custompdfs/2009/09/lnk302_304-306.pdf