Non-contact DC Voltage detector

[AndreyG]

How to build Non-contact DC Voltage detector/sensor?

 

[berkeman]

How to build Non-contact DC Voltage detector/sensor?

SEM, that's pretty much it.

 

[berkeman]

Well, that, or having your assistant make the DVM contact, and yelling the numbers to you...

 

[davenn]

SEM, that's pretty much it.

SEM ? (only thing I could put to that was scanning electron microscope ??)

Non contact AC V/A meters easy have one myself. DC ones seem much more difficult
I didn't really find anything useful when I did a google search for circuits for oneDave

 

[AndreyG]

Dear Berkeman, you over-simplifying it. At least one more method I know myself - field mills.
It should be something better though, more modern.

 

[berkeman]

SEM ? (only thing I could put to that was scanning electron microscope ??)

Yes, sorry if it wasn't obvious.

 

[berkeman]

Dear Berkeman, you over-simplifying it. At least one more method I know myself - field mills.

Well then you should have posted that. And you need to post about your thoughts to avoid getting an infraction for being obnoxious...

 

[AndreyG]

Dear berkeman I found your replies confusing sorry. I did not mean to be obnoxious or anything personal. I am new here, asked plain question, got pointless disrespectful sarcasm from you. It does not help.
Have good day.

 

[berkeman]

Dear berkeman I found your replies confusing sorry. I did not mean to be obnoxious or anything personal. I am new here, asked plain question, got pointless disrespectful sarcasm from you. It does not help.
Have good day.

If you know of a solution, and are asking for additional solutions, post what you know. Please do not waste our time here. We do not appreciate that. We enjoy helping people, but do not appreciate lazy people wasting our time.

 

[Baluncore]

You can measure the local voltage gradient by moving a FET gate connection backwards and forwards along a line towards the unknown voltage. That will produce an AC signal proportional to the field strength.

You could bounce a hollow metal sphere capacitor from the unknown voltage. It will come back charged. When you catch it, discharge it to zero and knowing it's capacitance and charge, apply V = Q / C.

 

[AndreyG]

I am not lazy berkeman. I do know few things but do not consider myself an expert in electrostatic measurement techniques. That's why i posted question hoping somebody more experienced will share his/her opinion with me.

 

[AndreyG]

You can measure the local voltage gradient by moving a FET gate connection backwards and forwards along a line towards the unknown voltage.

Do you mean to move it mechanically? Interesting. This I can try using small linear motor. Do you know if anybody produces equipment like that?
Thank you.
Andrey

 

[Baluncore]

Do you mean to move it mechanically?

Yes. But we do not know the magnitude of the voltage or the energy involved. That may destroy a FET. It would take exceptionally low leakage diodes to protect the gate.
I would use a slow geared motor with a crank to move the sensor as that would prevent the twisting of cables. The FET would be connected as a voltage follower so the low impedance voltage on the source would follow the field. The sensor might be a small sphere on the end of a hollow teflon or glass rod. You will need very low leakage and field disturbance.

 

[sophiecentaur]

It would depend on the actual voltage and its source impedance but I think you could try an Inductive Charging (link) method. Put a pair of (well insulated) parallel plates in the region to be tested. They will become polarised. Short them together, briefly and then take them away. The two plates will have equal and opposite charges. There should be a charge on this small capacitor and a PD between them which would be the original PD across the plates when they were in the test position. The volts you see would start off at this value and gradually decay.
Comments, anyone?

 

[anorlunda]

 

[sophiecentaur]

DC?

 

[anorlunda]

DC?

Yes DC. Put a DC charged wire near the bulb and see distortions is the patterns.

 

[sophiecentaur]

Yes DC. Put a DC charged wire near the bulb and see distortions is the patterns.

Good thinking young man. Sensitivity and calibration should be soluble.

 

[anorlunda]

I forgot an even more obvious non-contact sensor, the Van de Graff generator. The presence of lightning indicates a minimum voltage.

Message to the OP. If you want useful answers, you must be much more specific in your question. Otherwise, you merely provide us with a way to amuse ourselves.

 

[AndreyG]

Thank you for all the suggestions, serious and playful.

"... you must be much more specific in your question. "

More specific - ok. I need to detect presence of 400VDC in two-conductor cable with no load from ~25cm away. I do not need to measure the actual voltage only detect it. The detector should be reasonably small, likely hand-held. 25cm is approximate distance can vary +/-20%.

 

[anorlunda]

That's better. If you can put a suitable voltage divider between the two conductors, and power an LED, then a photo receptor could pick up the light.

 

[AndreyG]

Thank you for the reply. However I can not put anything on the cable. I need to sense the voltage in it without interfering. I need to come up with sensing method. After two days of studying I concluded that I need to sense static electric field, discovered Kelvin sensors, field mills. Idea of using FET with spatial modulation as suggested by baluncore is scheduled for tests.

 

[anorlunda]

. Idea of using FET with spatial modulation as suggested by baluncore is scheduled for tests.

If you mount the FET eccentrically on a rotating shaft, it would generate an AC voltage with frequency proportional to RPM when in a static field. With a notch filter tuned to that frequency, that signal should be very easy to detect. A counter balance would reduce vibrations.

 

[AndreyG]

Thank you, this is good insight. Modulating the input should help to eliminate-limit drift due to accumulation of stray charges. Similar to what other sensors do. Kelvin sensor is capacitor with oscillating plate, field mill uses shutter. My biggest concern now - will there be any electrical field to detect? (This is outlined in another post I placed). I still do not have clear picture what will be field distribution from two-conductor cable. If the charges on wires will be equal and opposite the fields at a distance from cable will likely cancel leaving nothing to measure. I will be setting up experiment next week.

Also found clever example of FET implementation:
http://amasci.com/emotor/chargdet.html