Field Mill Voltmeter Update

About 5 years ago I made this video:

I have toyed with the idea of doing an update on this project, to make it into something that others could build easily.

I have noodled around in CAD a bit considering how this could work, I cooked up an idea for a design that uses 3 PCB’s - Main Board, Rotor, and Display Board.

The housing and a few internal parts would be 3D printed.

Display and UI are still TBD- The original prototype just uses a bunch of RGB LED’s
as a simple bi-color bar graph display - RED for positive voltages and GREEN for negative.

The rotor and main analog board PCB’s could get gold plated for sexy factor!

A belt drive from a small brushed DC motor spins the rotor.
A flag on the pully creates the angular phase reference via a slotted optical switch.
The mount for the switch can be adjusted to get the phasing perfect - while it’s running.

One tricky thing is to prevent any charges from accumulating on the surfaces of the housing or the rotor- this is not trivial if the housing is plastic- maybe we need to print the housing in conductive plastic?

The hollow section of the red bracket would contain a small compression spring and a graphite rod (pencil lead) that would be lightly preloaded against the rounded end of the rotor shaft, the spring rests on a grounded pad on the display PCB to keep the rotor at ground.
This setup acts as a low-friction grounding brush.

Any electric charge on or near the rotor totally kills the function, the external signal gets totally swamped by the field. I learned this building the first one, without the grounding brush, it goes haywire.

The first tests would need to validate the idea of a PCB rotor.
The bare edges of the PCB might hold enough charge to make it not function correctly?
Maybe a bit of conductive paint could fix this?

Some other thoughts:

Add a distance sensor, or some parallax lasers to position the head at a known distance- so the actual surface voltage can be determined?

Display and UI? what would be cool and most practical?

Batteries? Internal LiPo or AA cells?

External Power? a jack to plug in some external source? USB power?

For sure a 1/4”-20 female thread somewhere so you can stick it on a tripod?

Data output? via what means? What format?

The GND reference? a banana jack?

Let me know if you have any other ideas- and PLEASE VOTE for your favorite project idea!

I just came across this device and your build is the only one i have found so far. I have a couple questions 1st, there appear to be 2 sensor contacts yet the video shows 1 signal path are there 2 signal paths or are both sensor contacts 1 and the same (electrically the same point)? what is the shutter made from is it 2 sided PC board if so are both sides grounded? or is it copper sheet? when the carbon contact grounds the shutter what is considered ground the negative side of the power supply or some earth ground? Is the schematic for the electronics available? and of course can i get a copy of your code for the micro processor?

i like the cans used for the case as it avoids issues with plastic case building a charge! also a lot simpler and easier to get than a plastic case or a 3d print with post processing to make it useable! any answers better than none thanks in advance!

I am travelling at the moment, but I will try to help as i can.

The unit shown in the video has 4 electrodes cross-connectted into 2.

These each feed a preamp - which then connects to a differential amplifier.

The whole circuit uses a virtual ground circuit at 1/2 Vcc- the ground brush connects here.

The shutter is 2 sided PCB and I connected both sides with solder.

I built this so long ago- I think it’s coded in assembly.

When I get back from NY trip I will post more details about this project.

Some tips from experience-

Make sure that everything near the electrodes is conductive and grounded.

Grounding the rotor is absolutely critical- it will instantly accumulate a charge and corrupt your signal.

Make sure your brush is low-friction and that the mount applies constant gentle pressure without bounce or vibration.

I found that a pencil lead rubbing on the axle tip is a good solution.