digital voltage divider for high voltage(200v)

Thread Starter

ariemeir

Joined Jun 24, 2011
66
Hi all !

I want to be able to control a ~ 200v dc voltage level by a digital microcontroller.
All the digital potentiometers i find on digikey are either 3.3 or 5v (most of them)

how can i do it with a high voltage ?

thanks.
Lenny
 

crutschow

Joined Mar 14, 2008
34,283
One way is to use a digital potentiometer or DAC to control an op amp with a high voltage output. Here's an example of doing that with a standard op amp. What output current do you require?
 

Thread Starter

ariemeir

Joined Jun 24, 2011
66
thanks for the reply,
eventually i will use the divided voltage to charge a capacitor which will be used as power
source for the next stage.
~30-50ma would work for me.

I will look into the link you've sent , thanks. I remember trying to find an opamp that can take 0-200v on its supplies, but couldn't.
I haven't used bootstrapping method, so i will look into what it means.

Regards,
lenny
 

crutschow

Joined Mar 14, 2008
34,283
Note that the transistor can dissipate significant power, depending upon the output voltage and current. How long will the output current be on? Is it just a short time to charge the capacitor?
 

Thread Starter

ariemeir

Joined Jun 24, 2011
66
basically i have a high voltage (~200v) , low current (~30-60ma) voltage supply charging a large capacitor (~5000uf )
http://www.ebay.com/itm/320831453118?ssPageName=STRK:MEWNX:IT&_trksid=p3984.m1439.l2649

Then once a second roughly, i discharge the capacitor through a heavy (~10ohm) resistive load for 100-200us.

I looked at the bootstrapping method, i am a bit concerned about the power loss in the opamp circuitry, since my 60ma are not too much to play with.
Another method i thought of was using a voltage divider when one of the resistors is actually a mosfet in a triode region, but then again - i will be bleeding all this current to
ground, wasting from the precious 60ma budget.

Your feedback is appreciated,
Regards,
Lenny
 

crutschow

Joined Mar 14, 2008
34,283
A low power op amp, such as the LM321 typically uses less then a mA current, so I don't see that as a problem. For 200V you can bump up the bias resistor values from those shown in the schematic to reduce the current through those. And the emitter follower buffer will require no added current since all the emitter current goes into charging the cap.
 

crutschow

Joined Mar 14, 2008
34,283
Below is a simpler circuit to charge a capacitor, as I understand your requirements.

The gain of the circuit is +40 so +5V input will charge the cap to +200V. The gain can be changed as you want and is determined by R1 and R3.

The op amp and transistors were one's available in my model library. You can use a LM321 for the op amp if you prefer. You will need to select some higher voltage NPN and PNP transistors from a supplier such as DigiKey or Mouser. They should have a Vceo rating of 250V or higher.

D1 and I1 are just to simulate the current-limit of the supply and are not in the actual circuit.

HV Charge.gif
 

Thread Starter

ariemeir

Joined Jun 24, 2011
66
thank you so much for your help , I will play with this ckt in spice !


I was wondering if I could run something else by you:

I am looking into different components, specifically these large capacitors i am going to charge.

Now i'm wondering about the ripple current limitation, for example
this component:
http://www.digikey.com/product-detail/en/EET-UQ2E222LA/P13835-ND/1755500

It fits the bill in terms of voltage rating and price range, but the ripple current rating is roughly 5A, when i am planning to have short bursts of possibly 15-20A (say 10% duty cycle).

If i charge my cap with dc, and discharge it through a resistive load - should i be worried about ripple currents at all ? Or should i just grab a couple of big aluminum caps and i am good to go ?


Thanks again,
Lenny
 

crutschow

Joined Mar 14, 2008
34,283
The concern about ripple current is the heating it causes because of the capacitor ESR. The power through a resistor is proportional to the square of the current, thus the allowable power at 10% duty-cycle would be about 10 times greater, increasing the allowable peak current by √10 * 5 = 15.8A. To check, you can feel the capacitor temperature when operating to determine if it's getting too warm but I think you'll be OK.

Edit: If you have LTspice I can download my .asc simulation file.
 

Thread Starter

ariemeir

Joined Jun 24, 2011
66
Thanks for the file sharing offer.
I didn't want to be rude, so i just assembled my own ltspice file.

Thanks for your help.
Lenny
 
Top