Op Amp Driven HV meters

[95se5m]

I am in the process of building a High Voltage power supply for a CO2 laser, it's 15KV current limited @ 30ma. I need a circuit that will monitor the voltage and current of the output across the discharge tube. Remember this circuit must handle HV without letting the magic smoke out of the chips.

The finished circuit will be incased in an oil filled unit for safety.

 

[IRstuff]

Resistive dividers are commonly used

TTFN

 

[95se5m]

I am trying to avoid resistive because of a noise issue, I tried a resistive ammeter and it fluttered between 1ma and 5ma, I need something more reliable and less jumpy

 

[Skogsgurra]

The divider shouldn't be jumpy per se. It should be perfectly OK to use for voltage measurements.

For current measurements, you could try a magamp circuit (aka Kraemer transducer or current transductor). Or use a low current LEM transducer. You will need to use a primary winding with several turns to get a decent resolution.

If you can put a sensing resistor in the current path near ground potential, it will be a simple matter to just measure tha voltage drop across it.

 

[buzzp]

Since your buying high dollar resistors for voltage readings and/or current, is a transformer out of the question for measuring the voltage? Otherwise, your kind of stuck with a resistor network.

The current measurement could be done using torroids, resistors, or even a little IC chip (where the current to be monitored is passed under the chip via a trace on a PCB). I would opt for the torroid or a little hall effect chip (if you are using a PCB) simply because of the isolation provided.

In addition, if a transformer is used for voltage measurements you will also be isolated.

They do make resistors designed specifically for current measurements (better temp coefficient and tolerance).

 

[logbook]

95SE5M

It sounds as though you are looking at unusual solutions to normal straightforward problems.

High voltage measurements are usually done with resistive dividers. It is true that direct measurements are possible with spark gaps, but for your application a resistive divider is the correct solution. If it is too noisy then that problem needs to be fixed. Is it some high frequency signal getting onto the high impedance attenuator? If so, then put a capacitor across the output of the divider and/or put an electrostatic shield around the divider. If you are creating the EHT from an unshielded EHT transformer then put an electrostatic shield around the transformer to minimise the stray capacitive couplings.

If you are looking at the DC value of the EHT supply then all those transformer proposals are obviously inappropriate.

Skogs is right about measuring the current in the low potential part of the circuit, ie in the ground return conductor. This trick makes the measurement much easier.

 

[OperaHouse]

About those noisy resistors!!!!! I suggest that you look at the voltage specification of common resistors. Small 1/4 to 1/2 watt resistors have a voltage rating of well under 500V. That means they will internally arc and create the noise you have seen. By the rules, you would need about 50 in series. You can cheat and probably get by with a little over a dozen if you pot them. I've seen a lot of companys do this. The higher the wattage, the higher the voltage they can take. Hope the 1 ma is only a test and you are not using that much current just to monitor.

 

[Warpspeed]

I would go out and buy a Fluke 20Kv EHT probe, that will give you 1 volt per Kv when working into a 10meg ohm load, and it will be safe to use, and reasonably accurate.

These probes also work into an oscilloscope, and I have never become aware of any serious HF noise problem when doing so.

 

[electricuwe]

Regardless of the circuit you choose please keep in mind that many of the currently used plastic materials are not able to withstand hot transformer oil for a long time.

We used to have a small heated vessel with transformer oil and everything that was supposed to be used under oil was tested there for at least half a year ( and lots of components failed !).

 

[95se5m]

Well, it's for monitoring. Here's the deal. I am building a HV power supply for a CO2 gas laser. The Plasma cavity is 24" long. I need the power supply to monitor the voltage across the cavity, the current through the cavity, and the voltage and current before the HV transformer. Obviously before the transformer is easy, resistive network will work well.

I am trying to have the highest impedance possible, I found a circuit online that uses an 80M and 82K divider circuit as the starting point of the OP-AMP driven meter movement. I am just leery about running 15KV @ 30MA directly through a shunted meter movement in a metal case. If it’s what I need to do, so be it. Let me know.

 

[buzzp]

Spacing is critical. This includes the spacing of your resistor leads so select a properly rated resistor. Even more attention should be given to the spacing in your metal box. There are 'guidelines' out there for spacing based on potentials (UL has some too). Keep in mind that elevation and humidity will affect the spacing requirements a great deal. So you will definately want a above normal safety margin in here for this reason. Better find more than one source of spacing tables to compare too as well. Perhaps a potting compound should be used to reduce the spacing requirements and to eliminate the environmental considerations. Since your probably dealing with low heat dissipation requirements, this could be real solution to your spacing problems.

I dont understand why the voltage measurement before the HV transformer is not adequate. This number could factor in the transformer ratio to come up with the secondary voltage (HV). This is typically how HV measurements are done.