Electrostatic Current Monitor

The neon won't fully discharge the cap, so the rate will be higher.

And the firing voltage won't be dependable, especially if the bulb is in the high side.

John

Sorry, typo. Should read

dt = C * dv / I

But you knew that already:)

Yes, I realized that just after posting. Searching google for the extinguishing voltage gives values all over the map. But I think the bulb will light at about 90V and go out around 60V. So the equation should be

dt = C * dv / dt = 0.047e-6 * (90 - 60) / 1e-6 = 1.41 seconds

or thereabouts. The small voltage drop should have little effect on a 12KV electrostatic filter.

I mentioned it should go in the ground side to prevent corona. The firing rate will vary with different bulbs, and probably other factors such as temperature. The cap will also have a tolerance. And you monitor the firing visually, so it's not going to be anywhere near the accuracy of a dvm.

But it's just a monitor, basically to alert you to problems. The current drain of an electrostatic filter can vary by many orders of magnitude, depending on leakage on the insulators from dirt or contamination, ambient humidity, how much corona there is from sharp points, etc. So it's more than good enough for the application.

My biggest problem is mounting the filter in the window and trying to keep the rain out. Raindrops hit the window ledge and make a fine mist. This gets sucked into the filter and shorts the high voltage to ground. I was trying to measure the voltage by holding a grounded lead near the electrode and estimating how much gap it would jump. But the loud bang made me nervous every time it arced over. The NE-2 seems like a much better idea. Simple, reliable, and much quieter:)

Mike Monett

A DVM in series would have to be set in current mode, which may give poor sensitivity. Use of a fixed, soldered in resistor at the "ground" or low side allows one to use a current meter in parallel with that resistor, and ther would be no problem with "dangerous voltages" at any time (meter connected or not). Anything else like a neon bulb is also an *OBVIOUS* solution. The patent should never have been awarded!

just put the ne2 in series with the unit's psu and monitor with a photodiode, we used that technique to monitor currents down to 10 nA or lower in a electrostatics lab, when we needed a floating fiber optic current sensor. Or put a 100K ohm 1/4th watt resistor in series with the device and put a 10 nanofarad ceramic cap across it to protect the meter from the turnon surge. We also made floating meter assemblies in plastic boxes with the RC circuit.

Steve Roberts

Better yet, use an LED; there is light from nA to mA...