Contact free DC voltage/electric field measurements
Employing a vibrating membrane measurement of DC voltages or electric fields from chrged items (ESD) are possile. The sensor can be homemade from household items. An Arduino type microcontroller is used to drive the measurement and proveide a readout.
Kind of reversing a condensor microphone by supplying "sound" as membrane vibrations it is possible to probe electric fields at the surface of the vibrating memebrane. More intedned as a kind of educational project I tried to realise this idea and explore the prospects of it.
The challenges are to detect the tiny induced currents (nA) while the sensing memebrane necesssarily is exposed to anvironmental fields. Also the vibration drive signal must be isolated.
The latter could be achieved by coupling the sound from a tiny loudspeaker resonantly coupled by a 20-30cm air column to the vibrating Al foil membrane, which serves es sensing element.
The electrical signal is conditioned by an RC netwoork, amplified an filtered to reject 50Hz humm via two operation amplifier stages. The amplified signal then is fed to two synchronous detection stages that operate with 90 degrees phase shift. The two detected and preaveraged signal voltages then are read by an 16-bit ADC into the microcontroller, which perforrms additional averaging and computation of amplitude and phase of the signal.
On the other hand the microcontroller supplies the frequency signal for the vibration actuator: a tiin ylogic stage creates the 90 degree separated sampling signals for the sync detectors. One of this signal is amplified by a simple transistor stage to drive the miniature loudspeaker used a actuator.
Upon start the microcontroller scans a range of frequencies to determine the proper resonance (between 400 and 750 Hz).
With this setup it was possible too measure the voltage of an adjacent plate at a distance. The sensitivty is sufficient to acess voltages in the 100ds..1000ds Volts range with 20mm distance, a few volts with 2mm distance, electroostaic charges in the several kV range with 10cm.
The sensitivity strongly depends on the distance (the coupling capacitance involved) and alos on the strength of the excitation and resosnance. Therefore to interpret the measure voltages in terms of potential at the test object, an offset voltage may be applied to the membrane, the observed difference
in reading reveals the actual sensitivity.
In the present version the offset voltage must be supplied by an externalk source.
In the moment the setup rather has educational value than presenting a ready-to-use meter. However, it can be particular usefult to montor electrostatic charges and test items -even the own body- for being charged, e.g. to identify/prevent ESD damaging items or actions.
NOTE: be cautious with high voltages! When experimentig with charged surfaces use voltage sources with sufficiently high inner resistnace repectively current limitation (max a few micro Amps)!!
The challenges are to detect the tiny induced currents (nA) while the sensing memebrane necesssarily is exposed to anvironmental fields. Also the vibration drive signal must be isolated.
The latter could be achieved by coupling the sound from a tiny loudspeaker resonantly coupled by a 20-30cm air column to the vibrating Al foil membrane, which serves es sensing element.
The electrical signal is conditioned by an RC netwoork, amplified an filtered to reject 50Hz humm via two operation amplifier stages. The amplified signal then is fed to two synchronous detection stages that operate with 90 degrees phase shift. The two detected and preaveraged signal voltages then are read by an 16-bit ADC into the microcontroller, which perforrms additional averaging and computation of amplitude and phase of the signal.
On the other hand the microcontroller supplies the frequency signal for the vibration actuator: a tiin ylogic stage creates the 90 degree separated sampling signals for the sync detectors. One of this signal is amplified by a simple transistor stage to drive the miniature loudspeaker used a actuator.
Upon start the microcontroller scans a range of frequencies to determine the proper resonance (between 400 and 750 Hz).
With this setup it was possible too measure the voltage of an adjacent plate at a distance. The sensitivty is sufficient to acess voltages in the 100ds..1000ds Volts range with 20mm distance, a few volts with 2mm distance, electroostaic charges in the several kV range with 10cm.
The sensitivity strongly depends on the distance (the coupling capacitance involved) and alos on the strength of the excitation and resosnance. Therefore to interpret the measure voltages in terms of potential at the test object, an offset voltage may be applied to the membrane, the observed difference
in reading reveals the actual sensitivity.
In the present version the offset voltage must be supplied by an externalk source.
In the moment the setup rather has educational value than presenting a ready-to-use meter. However, it can be particular usefult to montor electrostatic charges and test items -even the own body- for being charged, e.g. to identify/prevent ESD damaging items or actions.
NOTE: be cautious with high voltages! When experimentig with charged surfaces use voltage sources with sufficiently high inner resistnace repectively current limitation (max a few micro Amps)!!
Discussion (3 comments)
DE0068065ID 10 months ago
If you exchange the emitter with collector of it it should work too.
Brian Tristam Williams 10 months ago
Inch-o 10 months ago
The transistor as shown should be a PNP type... Then also the electrolytic capacitor across the loudspeaker... very likely acts as a short circuit to AC signals...
Otherwise very nice theoretical explanation of the principle of operation - Great!
monk99 10 months ago
thank you for the comment!
Yes the drive output is somewhar simplistic but works. The role of the parallel capcitor in combination with the 470 Ohm rsistor is to "smooth" the high frequencies of the rectangular switching pulse in order to minimize the croosstalk of the drive voltage to the input. Also already here the drive signal is tweaked a bit towards a simple sine excitation, finally more filtering towards a more or less "pure" sine is done by the acoustic resonanc e serving as a kind of filter.
The simple transistpor resistor combination is an "ineffective" amplifier (as emitter follower) but the resulting signal worked for mini loudspeakers as well with 8Ohms upt to 80Ohms and yields a reasonable amplitude (more power would create an even more unbaeringly annyoing soundfloor..). Any other contraption would also work....