Driving a led from high-voltage source

[Peter Benkö]

Hi!

I`m pretty new to electronics (so please be gentle), but came across a problem with which I need some help.

I have an AC power source with variable output ranging from 150V to 5kV. I would like to branch this circuit to be able to use a LED with it (at will with a switch).

I`m pretty sure I will need a rectifier and probably a capacitor too, but how can I regulate the voltage for this additional circuit to supply only the required voltage for the LED. The whole thing should be pretty small (IC size).

Is it possible at all?

 

[poor mystic]

Hi
I hope you don't do this thing. It freaks me out, and I'm brave and tough.
5kV is too high a voltage.
Please do not feed current from your source to any switch - you or someone else might so easily die as a result.
When you've accepted that, I'll listen more, but right now I want to be sure you're safe.

 

[Peter Benkö]

Hi, and thanks for the fast reply. Don´t worry, it all exist only on paper (screen). I won´t do anything before simulating and preparing for the outcomes (and besides I´m only doing a proof-of-concept).

 

[poor mystic]

Hooray!
I don't know what this voltage source of yours can be but it's lethal. There's no way to safely regulate 5kV, and there is certainly no IC which is capable of channeling currents at such voltages.
What's it for, please?

 

[Peter Benkö]

Unfortunately I´m not allowed to give you too much details about the voltage source, but it is used in the healthcare. The initial idea was to "steal" a part of the current, but I don´t know, if that is possible.

 

[poor mystic]

I consider the result you want to be impossible.
Sorry, but at least you'll survive.

 

[Harald Kapp]

Unfortunately I´m not allowed to give you too much details about the voltage source

Unfortunatley this will limit our possibilities to help you.

You really should follow poor mystic's advice.
Your being "pretty new to electronics", dealing with 5 kV and all this in a healthcare related circuit scare me, too. Requirements on safety are extremely high in healthcare products. What you try to do does not sound safe.

If you need an indication of the presence of the high voltage, I suggest a different method: Use a high-voltage safe transformer to sense the voltage of your source. E.g. a 5kV:5V transformer or similar. With this ratio the transformer will give you between 0.15V and 5V for 150V to 5kV on the input. And it will create an isolation barrier for safety.
Use an amplifier, rectifier and limiter to create an ON/OFF signal for the LED. Power the LED from a safe low voltage (e.g. 5V).

 

[Peter Benkö]

OK. Thank you!

 

[BobK]

A standard LED takes 20mA. If you were to run it from a 5KV source, that is 100W. It is unlikely that your 5KV source has an extra 100W available. Most likely, it cannot product 100W total. Not to mention the heat you would have to get rid of.

Bob

 

[Peter Benkö]

Thank you very much for your help. It became clear to me, that there is no really good solution to this problem along my ideas.

I would like to thank you very much for your time, and maybe it would be the best to delete this thread, as I cannot (more precisely not allowed to) give more information, and the problem cannot be solved without it.

And don't worry, I won't electrocute myself.

Peter

 

[Harald Kapp]

I don't think we'll delete this thread. It can still serve as a lesson to others.

Harald

 

[Arouse1973]

Hello Peter
Being AC you could possibly use a sense coil to pick up the signal from the cable. A small inductor wrapped around the wires might work.
Adam

 

[Colin Mitchell]

This is a very simple application.
You just need a capacitor, zener, diode and a few other components.

 

[(*steve*)]

I would take a look at how non-contract voltage detectors work. This would allow you to detect the electric field using a probe placed close to the (very well insulated) high voltage output.

 

[Externet]

High voltage presence monitoring is not done with a LED, but with a neon bulb. And there are sub-miniature ones as used in photo cameras flash circuits. A proper resistor in the MΩs range should provide brilliance in the high voltage range.

And a LED does not 'take 20mA' as stated. Can be driven from ~1mA for high efficiency types.

A solar cell array canibalized from an old wristwatch is small as an IC and could provide ~1mA. If attached to neon bulbs, it could turn on a real led, as a generating optocoupler.

As another resource, as there are no neon optoisolators that I know of, I had to make them with cyanoacrilate glueing a neon to a photodiode, inside black shrinktube, and works perfectly.