measuring current (microamp range) for a supply voltage of 300V - suggestions

[riforifo]

Hello,

I have a setup where I need to measure the current out of a 300V supply (current is supposed to be in microamp range)
I think that I can use uCurrent gold in a low side mode but that will disturb the GND.

As far as I see, I can't use uCurrent in high side mode when the supply is 300V.

I would be happy to hear your suggestions for (high side) measuring current in uA range when the supply voltage is 300V.

thanks
rifo

[IanB]

300 V is not a very high voltage. You can insert a current meter anywhere in your circuit as long as the measurement part is properly insulated and isolated to avoid short circuits or current leaks. If you were dealing with 3000 V it would be a different story...

[riforifo]

Hello,

Thanks for the answer. In my initial post I forgot to add that this should be an automated setup hence using a current meter is not possible.

I came across  HV7802 from microchip which supports up to 450V in high side configuration but would be happy to hear alternative options.

[Ice-Tea]

An automated setup does not exclude an off-the-shelf meter. Many have some for of connectivity (GPIB/LXI/Ethernet/USB/Serial/...)...

[IanB]

Anything that measures current is a current meter. The uCurrent is a current meter.

I am not sure why you keep talking about "high side" configurations? Maybe you can explain more about what you are trying to do? Are you trying to do laboratory test instrumentation, or are you trying to design something with built-in current regulation? Why do you think you might need a uCurrent in the first place?

[HKJ]

A standard bench meter can be used, including computer connection. My Keysight & Keithley meters are rated for 500V from measurement minus to Earth.

[tautech]

A standard bench meter can be used, including computer connection. My Keysight & Keithley meters are rated for 500V from measurement minus to Earth.

Exactly.
Any reasonable bench DMM can accomplish uA measurements. Model will depend on the resolution and accuracy required by the OP.

[riforifo]

Hello,

Thanks again for the answers.

I would like to keep the costs down so would prefer to use a cheap IC and a uC. This setup will be replicated so the costs quickly builds up when one uses an off the shelf equipment with (GPIB/LXI/Ethernet/USB/Serial/..)

I would like to use high side configuration and connect my current meter to 300V directly so that I don't disturb my ground (which will happen in low side configuration)

thanks

[HKJ]

The DIY solution:
Use a microprocessor that as a ADC with build in preamplifier to sense current on a series resistor and drive a optocoupler with digital data. The other end of the optocoupler can be feed to a serial to usb converter. For power use a isolated DC/DC converter.
This means you need a optotcoupler and a DC/DC converter that can handle 300V, this is very easy to find.

[Ice-Tea]

Well, as the saying goes: your funeral. But my guess is you should be able to get some sort of connetivity on a meter for well under 100$. Is it worth developping your own solution to avoid that? All depends on volume, topology etc but still...

[Alex Nikitin]

I would be happy to hear your suggestions for (high side) measuring current in uA range when the supply voltage is 300V.

Hi Rifo,

What is the current range, how accurate do you need to measure the current and how much voltage drop at maximum current is OK?

Depending on the acceptable accuracy and voltage burden, you could get away even with a reasonably simple current mirror (~3 BJTs + some diodes and resistors).

Cheers

Alex

[ocw]

You said that you did not want to break the ground path.  But, while there are all sorts of things which can go wrong if you are not careful, I used my version of EEV's uCurrent Gold to measure the current on a 1,000 volt supply down to the pA range with 100 aA resolution by taking the measurement on the ground path.  My unit uses an ADA4530-1 IC and has virtually zero shunt insertion resistance.  That made the voltage seen by my current to voltage converter to be very low and it survived happily ever after.

User beware.  Incorrect connection, or inserting something with too great of an insertion resistance, might cause it to activate its smoke module and fail.  I was willing use my ADA4530-1 in a proof of concept test and did not have any problems.

[Kleinstein]

For a 300 V supply the small voltage drop on a low side shunt (e.g. 100 mV range) might not be a problem.

Another option might be to use a kind of trans-impedance amplifier at the low side. However here the AC performance can become tricky.

There is the option to build the classical higher side current sens circuit with an OP that works at the positive rail and an FET. The only difficulty is getting a low current P-MOSFET suitable for a higher voltage. It also needs some supply at the high side, though some OPs can be rather low power (e.g. 10 µA range).

[David Hess]

Another option might be to use a kind of trans-impedance amplifier at the low side. However here the AC performance can become tricky.

This is what I would do but it does require a low voltage negative supply.  The low side voltage drop can be arbitrarily small.

There is the option to build the classical higher side current sens circuit with an OP that works at the positive rail and an FET. The only difficulty is getting a low current P-MOSFET suitable for a higher voltage. It also needs some supply at the high side, though some OPs can be rather low power (e.g. 10 µA range).

Low power operation will limit bandwidth but a low voltage floating supply could be used to avoid this.

A high voltage zener diode in series with the output to ground can lower the voltage requirements of the p-channel MOSFET.

[tkamiya]

I really don't know what the concern is.....

Voltage may be 300V referenced to ground, but two points where meter or whatever will be inserted, voltage differential is minimal.  Since it has to be automated, cutting a wire or a trace probably isn't an option.  Then find a point where there is a constant resistance, and measure a voltage across that.  Make sure your measuring apparatus is FLOATING.  Then it should be fine.


300V is not that high of a voltage, especially if the current is in micro amp range.  It can sting you but that's about it.  Unless OP can go into detail of what kind of automation it is, I don't think anyone can give a concrete advice.  But this should be pretty basic.

[IanB]

This is the usual problem where the poster is completely unable to describe what they are trying to do, in what context, with what objectives. It seems to be that when someone can see the whole picture in their own mind, they somehow imagine everyone else is a mind reader and can see the picture too.

This question cannot be answered without more information about the application.