0
\$\begingroup\$

I was wondering whether I could use a simple multimeter to measure small currents (<1mA) at high DC voltages (~30KV). From a cursory internet search I got three answers which were repeated ad nauseam:

  1. Don't do it you'll kill yourself, your family (past and future), and most of your friends.
  2. Don't even attempt to do it with a standard multimeter.
  3. Introduce a resistor and measure the voltage drop over it.

Okay, so

  1. I promise to be careful.
  2. When you measure current with the old analog ammeters you simply measure the voltage drop over a small shunt resistor (i.e. 0.1-10 Ohm) connected in series with your circuit. So if I connect an ammeter to my circuit, which has resistance > 30MO so that the maximum current is 1mA, the voltage drop over the shunt resistor is at most 10mV. So the current is small, the voltage drop on the resistor is small (the resistor is not subjected to overvoltage), why not do it then? Is it something that has to do with the way digital ammeters are implemented?
  3. Sure I can do this, but isn't this what the ammeter does anyway?
\$\endgroup\$
4
  • \$\begingroup\$ One word: arcing \$\endgroup\$
    – PlasmaHH
    Commented May 3, 2017 at 19:18
  • \$\begingroup\$ Don't miss the related questions on the right hand side of this page. \$\endgroup\$ Commented May 3, 2017 at 19:39
  • 3
    \$\begingroup\$ If you are asking this kind of questions, I would strongly advise you to stop doing what you are doing with 30 kV source. Obviously if you place your shunt at the ground end of your 30 MOhm load, you can use milllivoltmeter relatively safely, but you didn't mention this. Are you using a grounded cage around your source? Do you have a crowbar discharge device in the cage? So, the best advice is to stop what you are doing. \$\endgroup\$ Commented May 3, 2017 at 20:01
  • 1
    \$\begingroup\$ You seem to disregard all safety warnings you've already found in the internet, but at the same time your questions imply you don't seem to know where you're stepping in. Good luck. \$\endgroup\$ Commented May 3, 2017 at 20:14

5 Answers 5

0
\$\begingroup\$

Since the voltage drop across a DMM current mode shunt is very low voltage there is no risk when sensing current from the 0V or return side of the HV.

Sensing on the high side exposes the meter to you if not careful and is not the preferred way.

I see no problem if careful and done as I described.

Care included shorting the HV after power removed to ensure, all stored charge continues to bleed safely to ground, especially from e-caps with memory.

\$\endgroup\$
4
  • \$\begingroup\$ Thank you very much, this is exactly the answer that I was looking for. To the point, while also highlighting the dangers and the safety precautions to be taken. \$\endgroup\$
    – Slava P
    Commented May 4, 2017 at 6:45
  • \$\begingroup\$ (Unfortunately I have to sense on the high side as well due to the corona discharge which is left unaccounted for when I measure on the ground side). \$\endgroup\$
    – Slava P
    Commented May 4, 2017 at 7:04
  • \$\begingroup\$ Can you please explain why you (think you) have to measure high side? \$\endgroup\$ Commented May 4, 2017 at 17:01
  • \$\begingroup\$ I guess the corona to ground is a moving target.. see my second answer. THis is a SWAG since the DUT is undefined. \$\endgroup\$ Commented May 4, 2017 at 17:03
0
\$\begingroup\$

Yes, it could be done. But why?

If you try it, never get close with power connected. Make sure you can pull the plug remotely. Use a non-contact voltage sensor every time you come near it. And make sure you aren't going to cause issues on other parts of your house/building/office's grid.

\$\endgroup\$
0
\$\begingroup\$

Standard battery-powered multimeter? Yes, it'll float up to the level of the circuit it's connected to.

However, none of its insulation is going to be rated for 30kV! So the only way to do this safely would be to put it and the cables in an insulated enclosure rated for 30kV.

Obviously touching the multimeter while it's connected could be fatal, so you have to set the whole thing up, retreat to safe distance, and then turn it on.

\$\endgroup\$
3
  • \$\begingroup\$ Just to be clear, you're worried about arcing between the multimeter/cables and the hot/ground parts of the circuit? \$\endgroup\$
    – Slava P
    Commented May 4, 2017 at 6:47
  • \$\begingroup\$ Yes, and the user! \$\endgroup\$
    – pjc50
    Commented May 4, 2017 at 8:59
  • \$\begingroup\$ Yes, of course. However, that hazard is eliminated if the equipment is set in advance, before the circuit is energized, as you wrote. \$\endgroup\$
    – Slava P
    Commented May 4, 2017 at 10:55
0
\$\begingroup\$

Simple:

  1. Use a coil meter;

  2. Use a high linearity opticoupler;

  3. Use a digitally isolated meter;

  4. Use a module to convert it locally and transmit the readings digitally via couplers;

...

The sky is the limit.

\$\endgroup\$
3
  • \$\begingroup\$ 30 kV optocouplers are not exactly standard, and will be very hard to find. Of course, you can roll your own using fiber optics, but that's not what you call a beginner's project. \$\endgroup\$ Commented May 4, 2017 at 1:05
  • \$\begingroup\$ Huh? I'd call playing around with 30kV alone not a beginners project. \$\endgroup\$ Commented May 4, 2017 at 4:24
  • \$\begingroup\$ 1. So, basically an analog ammeter. Thanks for 2+3. \$\endgroup\$
    – Slava P
    Commented May 4, 2017 at 6:59
0
\$\begingroup\$

Corona starts first as a Partial Discharge due to surface contaminants and oscillates like a Unijunction RC cct. Are you trying to measure AC,DC or corona current? I would use a 1 turn current loop around high tension insulated wire with 50 Ohm coax and 50 termination into a schottky diode FET buffer to measure DC or a LeCroy scope to see current pulse. As I have done before.

What is your end-game?

To understand corona or fix it>?

To fix , connect jumper to short any HV charge, clean thoroughly offending surfaces with good IPA (isoprop.alc.) and coat surfaces with silicone RTV.

To understand? ask a better question and don't use DMM's You will get false readings from high L fly leads and antenna effects.


Don't worry about past and future life.

This is the only one that counts.

\$\endgroup\$
7
  • \$\begingroup\$ I'm using a non-isolated PS, 0V side earthed. An electrode is held at high voltage. The electrode might reside at physical distances from the earth comparable/smaller than its distance to the 0V electrode. Certain levels of corona at the HV electrode (either to the 0V electrode or to earth) are desired. At this stage I just want to measure the total average current requirements of the circuit, and their dependence on the distance & orientation of the electrode w.r.t. earth. I asked a simple question to which I knew the answer in order to be sure (HV risks and all). Thank you for your help. \$\endgroup\$
    – Slava P
    Commented May 6, 2017 at 12:54
  • \$\begingroup\$ A small singe turn current loop to coax terminated with 50 Ohms is best method with 1GHz DSO as rise time is <<1ns to peak current,. \$\endgroup\$ Commented May 6, 2017 at 15:02
  • \$\begingroup\$ Could you please elaborate a bit more on the ''small singe turn current loop''? Do you mean a closed-loop hall effect sensor such as LTS 25-NP or something else entirely? \$\endgroup\$
    – Slava P
    Commented May 7, 2017 at 5:43
  • \$\begingroup\$ No I meant exactly as I said. 50Ohm Coax with a short loop from centre to shield around current carrying discharge into a 50 OHm LeCroy scope or equiv. \$\endgroup\$ Commented May 7, 2017 at 13:15
  • \$\begingroup\$ What's the principle? According to Faraday's law an induced EMF is equal to the change of magnetic field flux through the loop, which in turn is proportional to the current in the wire. If I understand correctly this setup will measure the changes in current. Are you suggesting to use integration on the scope? What am I missing? \$\endgroup\$
    – Slava P
    Commented May 7, 2017 at 17:46

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.