I would like to make an indicator to show that DC bus (700VDC) capacitors are charged (be careful!).

What is the best way to make a LED indicator, which will work for a long time from 40VDC to 1000VDC without additional power supply and with minimum power losses?

  • \$\begingroup\$ "work with" how? What is the intention? \$\endgroup\$ – Eugene Sh. Apr 5 '19 at 16:53
  • \$\begingroup\$ at what? constant current? \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Apr 5 '19 at 16:54
  • \$\begingroup\$ it should show us that DC bus capacitors are charged \$\endgroup\$ – Nikolay Apr 5 '19 at 16:54
  • 3
    \$\begingroup\$ Welcome to EE.SE! Keep in mind that questions about optimization (i.e., "What is the best ...?") require a definition about what problem dimensions are to be optimized for your application, such as size, speed, energy consumption, user experience, etc. Since these can't be optimized all at once, you need to have a good idea of which ones are most important to you, and be able to articulate that clearly to us. \$\endgroup\$ – Dave Tweed Apr 5 '19 at 16:55
  • 1
    \$\begingroup\$ As interesting as an answer would be, my strong recommendation is that if you have to ask the question, you don't have the skills to work safely with HVDC. Mains is risky but you'll usually survive. At 700V though, one mistake and you're dead. Or one mistake in assembling your project and someone else is dead when the case goes live. You aren't safe to do it, and it would be dangerously idiotic to try. There are easier ways to commit suicide which probably won't kill other people too. \$\endgroup\$ – Graham Apr 5 '19 at 21:50

Connect a moving coil analog voltmeter across the power bus.

Either a voltmeter as shown with internal series resistor or an external resistor and scale calibrated for the desired range. Photo from this useless site.

Old-school suppliers such as Crompton should be able to supply a meter with the markings you need, if not a turnkey solution.

enter image description here

  • 1
    \$\begingroup\$ GMTA .........great alike \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Apr 5 '19 at 17:39
  • \$\begingroup\$ How does this answer the question? \$\endgroup\$ – bot47 Apr 11 '19 at 17:01
  • \$\begingroup\$ @MaxRied The meter needle indicates the voltage. It works down to 40V and up to 1000V. The OP asked for an indicator so this answers the question. The scale can be marked with red above 40V to make it even more obvious. It's much safer than other options because it's obvious when it's not working so the possibility of shock is significantly reduced over something like a neon bulb. \$\endgroup\$ – Spehro Pefhany Apr 11 '19 at 17:14
  • \$\begingroup\$ @SpehroPefhany Didn't the OP asked especially for a LED solution? \$\endgroup\$ – bot47 Apr 11 '19 at 19:34
  • \$\begingroup\$ @MaxRied Yes, and indeed I first thought of an LED relaxation oscillator but really the obvious solution isn't always the best and safest solution. \$\endgroup\$ – Spehro Pefhany Apr 11 '19 at 20:28

Something else to consider is to build a relaxation oscillator using a DIAC, capacitor, LED, and couple of resistors.


DIACs are still readily available, although Digi-Key wants to sell them in full reels. They can be found at most electronic suppliers as well as places like eBay, AliExpress, Banggood, and DealeXtreme.

The advantage of using a relaxation oscillator rather than driving the LED with a large-value resistor is that the LED remains visible (flashing) with low voltages applied. It will stop flashing when the input voltage drops below the sum of the DIAC trigger voltage and the LED forward voltage:

  • \$\begingroup\$ Probably the best solution. Only maybe R1 should be 10 M or even more, and the cap under 1uF. Also there are LEDs with 100X light output than the listed one, so R2 can be increased too, which would save power. \$\endgroup\$ – Ale..chenski Apr 5 '19 at 19:46
  • \$\begingroup\$ I normally use very-high-brightness LEDs. The part number on my schematic is simply what the built-in CAD package has as a default. \$\endgroup\$ – Dwayne Reid Apr 5 '19 at 20:08
  • \$\begingroup\$ You have to keep the discharge current high enough to ensure that the diac switches cleanly from conducting to not conducting. In other words, R2 has to remain fairly low value. But higher brightness is better anyway. \$\endgroup\$ – Dwayne Reid Apr 5 '19 at 20:10
  • \$\begingroup\$ Wouldn't using an optocoupler here be preferred to decouple and shift to lower voltage? Signal to user/operator shouldn't be powered by high-voltage if they don't have to. \$\endgroup\$ – Mast Apr 6 '19 at 5:59
  • \$\begingroup\$ Original poster wants something that doesn't require external power. Plus: you still have the problem of driving the LED in the opto from that high-voltage power supply. That requires either a large power resistor (or two or three to handle the dissipation) or a relaxation oscillator as shown above. \$\endgroup\$ – Dwayne Reid Apr 6 '19 at 19:28

An example of a reliable economical no-power-supply (less than US$10) indicator solution that does not cost more than US$200, like the other meters. :(

Since this coil draws 50 µA full scale it is equivalent to 10 V/50 µA= 200 kohm. And thus at 1 kV the R load is 50 mW full scale with 1 kV/50 µA = 20 Mohm 1% or +/-200 kohm.


simulate this circuit – Schematic created using CircuitLab

It also draws the least current and is readily available.

Enter image description here


Another potential option is to use a neon bulb or lamp. The common neon indicator that I used to use is the NE-2H - this has fairly-wide current capability and would be able to handle the current range of caused by the supply voltage changing from less than 100V up to 1000V.

The downside is that a neon indicator does not match your requirement of indicating down to 40 Vdc. The NE-2H extinguishes (after being lit) at about 60 Vdc.

NE-2 & NE-2H indicators are still readily available. There are also much larger neon bulbs and lamps but they may not be readily available any longer. But you can check.

Final downside of a neon indicator is that they do die after an extended time. You have to weigh the consequences of the indicator failing some time in the future. Do note that they fail "gracefully" - they don't fail completely at one time, but rather, degrade. You would use that degradation as an indication that the lamp needs to be replaced.

  • \$\begingroup\$ I had considered this but discarded it due to the V range must be <<1mA with 1M series R is barely visible at 50uA , but then when it wears out you can reverse it and use the other electrode to illuminate. ha. \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Apr 5 '19 at 18:32
  • \$\begingroup\$ I had considered suggesting a simple neon relaxation oscillator but the downside is that the minimum operating voltage would be about 90 Vdc. But it doesn't get much simpler than that - and remains quite visible even with low supply voltage. \$\endgroup\$ – Dwayne Reid Apr 5 '19 at 18:49
  • \$\begingroup\$ Will the neon work on DC power? \$\endgroup\$ – Harper - Reinstate Monica Apr 5 '19 at 23:43
  • \$\begingroup\$ Yes. Neon works well on DC supply. However, only one electrode illuminates. Neon relaxation oscillator works only with DC power. But minimum operating voltage for relaxation oscillator is the firing voltage of the neon bulb - about 90 Vdc for NE-2H. \$\endgroup\$ – Dwayne Reid Apr 5 '19 at 23:55

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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