Maker Pro
Maker Pro

Improvised Alternatives to a High Voltage Probe for Oscilloscope

I built a 20cm square plate capacitor using aluminum foil and an office laminating machine. It is being charged with a 20KVAC square wave from a modified domestic ion generator, which is a low current source.

I would like to connect an oscilloscope across the plates to observe the waveform of the applied signal while experimentally adding various values of parallel resistance to influence the charge/discharge time.

Is there any way this can be done without purchasing an expensive high voltage probe, or trying to build one? Maybe a type of improvised sensor that does not require physical contact.

All I need is a visual, or relative, approximation of what is happening, not actual measurements. What are my best options?
 

hevans1944

Hop - AC8NS
You haven't provided enough information for a meaningful response.

How can the capacitor you built be "charged with a 20kV AC square wave from a modified domestic ion generator?" Capacitors are "charged" with DC not AC. What model and manufacturer was the domestic ion generator before you "modified" it? What modifications were made? Be specific in your answers and provide a schematic clearly illustrating your modifications.

There are good reasons why high-voltage oscilloscope probes are expensive, and why their high-frequency response is limited. These probes must be carefully matched to the oscilloscope you are using and proper high-frequency compensation adjustments made, a procedure you are probably not qualified to perform. More information can be found at the links provided by this Google search result. You can find similar results using other search engines. Try searching with this string as the target: "high-voltage compensated oscilloscope probe theory."

There are non-contact methods for measuring voltage, but while simple in principle they are complicated in practice. An electric-field mill can be used to modulate the field surrounding the capacitor, thereby allowing a small voltage, proportional to the voltage "seen" by the field mill, to be produced. Modulation is typically performed with a motor-driven "chopper" and the high-frequency response is limited by the modulation frequency to a few hertz. I used one of these devices to measure and control the acceleration potential, up to 1.7 MV, of a tandem particle accelerator.

If all you need "is a visual, or relative, approximation of what is happening, not actual measurements" then a leaf electroscope might be an inexpensive approach. You can build one of these yourself using just an insulating cork, a copper wire, a glass jar fitted with the cork, and thin strips of aluminum foil. Insert the wire through the cork and provide an "L" shaped bend inside the jar to support a strip of foil. Use a drop of superglue to secure the center of the length of foil strip to the wire. Bend the two foil "leafs" down so they are parallel and touching each other. Voila! You have built a leaf electroscope. More information can be found here.

Early adopters used gold leaf, probably because it can be hammered into very thin sheets and is "heavy," but I found that some types of capacitors could be disassembled to salvage very thin pieces of aluminum foil used for the capacitor plates. That may no longer be true today, because it is easy to metalize thin plastic film to make capacitors. Worth investigating though. You can also purchase gold leaf and other thin foils from art supply stores.
 
Last edited:
AC square waves are in effect pulsed DC. The frequencies involved are less than 100Hz, not HF. I want to set the cap up to charge and discharge at this rate by applying parallel bleed resistors. In other words, not simply charge up to 20KV and stay there. I can obtain some indication of this activity by placing the scope probe against the outer dielectric. I was hoping for an improved version of this technique or something like it. The leaf method would reveal averaged charge but not any frequency component.
 

hevans1944

Hop - AC8NS
AC square waves are in effect pulsed DC. ...
No, AC square waves ARE NOT in effect pulsed DC. AC means Alternating Current. DC means Direct Current. Pulsing has nothing to do with it. Perhaps you should consider taking an introductory course in the study of electricity.

I was hoping for an improved version...
Hoping is what I used to do after placing a tooth under my pillow, before going to sleep, and hoping the Tooth Fairy would be generous. Sometimes the tooth would be gone the next morning, but the Tooth Fairy forgot to leave me anything in exchange for it.

Science never depends on hope to achieve an objective, and quite soon I no longer believed in the existence of a beneficial Tooth Fairy, and no longer hoped for things I wanted. Instead, I studied and worked for them. You were given some suggestions of areas you could explore, but these you rejected without offering anything in their place. So, I will offer no more suggestions but hope you will find "an improved version" to aid you in your "research."

All I need is a visual, or relative, approximation of what is happening, not actual measurements. ...
Perhaps you need to study some history first. Lord Baron William Thomson Kelvin (long-bearded, white-haired, physicist) once began an academic lecture with these words:

"In physical science a first essential step in the direction of learning any subject is to find principles of numerical reckoning and practicable methods for measuring some quality connected with it. I often say that when you can measure what you are speaking about, and express it in numbers, you know something about it; but when you cannot measure it, when you cannot express it in numbers, your knowledge is of a meager and unsatisfactory kind; it may be the beginning of knowledge, but you have scarcely in your thoughts advanced to the stage of science, whatever the matter may be.
 
You implied the plate cap would not be charged by the HV source I described in my OP. I know from firsthand experience (as in "ouch") this is not the case.

Thank you for taking the trouble to answer anyway. I was simply seeking expert practical advice on a specific problem. I understood that to be the purpose of this forum.
 

hevans1944

Hop - AC8NS
You implied the plate cap would not be charged by the HV source I described in my OP.
I implied no such thing. Since you "modified" an ion generator, it is a pretty sure thing that your original device was originally intended to make positive ions, by removing one or more electrons from atoms or molecules in a conductive gas environment. So, yes, expect your DIY capacitor to charge up and down, but always having the same polarity of output.

You can easily check the polarity of your charged capacitor by connecting a string of NE2 neon lamps in series with a large-valued current-limiting resistor. Connect one end of the resistor to your circuit common and connect the other end of the NE2 neon-lamp string to the high voltage terminal of your cap.Turn the ion generator on and observe that one and only one of the two electrodes inside the NE2 has a glow discharge surrounding it. Can you explain why this is? Is it the positive or negative capacitor voltage that is causing just one, and only one, of the electrodes to glow?

There are many ways to extract high voltage DC from pulsed AC sources, but somewhere along the way rectification MUST take place to create positive ions. Cocroft-Walton voltage multipliers seem to be popular because they are uncomplicated and relatively inexpensive. Another method uses field emission of electrons from needle-like sharp-pointed electrodes that will leave behind a positive charge as electrons are sprayed into the air. Not a particularly efficient way to produce positive ions though, because it also excels in manufacturing ozone (O3) from ordinary air.

Again, it would help us to help you if you answered my original questions: who made your ionizer, where are the schematics for it before you "modified" it and where are the schematics after you "modified" it. But most important, and I forgot to ask this specifically, What the fck are you trying to DO?

Oh, and what about creating negative ions with a similar high-voltage electrical discharge? Sorree, Charliee... the physics just doesn't favor that happening.
 
Hmm ... I wonder why they call them negative ion generators?

Anyway, my "modification" is just switching the HVDC output so it pulses at low frequency. Yes, this is observable via a neon bulb. Interesting point you made about only one electrode glowing.

The original question was about measurement technique, not a particular project. I only wanted to see the cap charge and discharge on the scope, and how this cycle is affected by parallel resistance.

I gather from the response so far that I probably need to save up for a proper probe. But I may try to rig something up based upon electrostatic coupling. Thank you for your input.
 

hevans1944

Hop - AC8NS
All is not lost yet! The main reason HV attenuating probes cost so much is that they are built to higher expectations of accuracy and frequency response. However, getting a voltage divider rigged up to allow you see slowly changing charge/discharge wave forms is not difficult. All you need is a high-ohm value resistor connected between your oscilloscope low-voltage signal input and the high voltage you want to monitor. The resistance must be large enough that it will not appreciably "load" down your HV DC output, which would interfere with the experiment you are attempting to make by paralleling different values of resistance with the output.

A typical HV probe provides a 100:1 attenuation ratio, so 10 kV is measured as 1 V. Most o'scopes have about 10 meg-ohms input resistance, so to get an attenuation ratio of 100:1 requires (roughly) 100 times that resistance in series with the input, or about 1000 meg-ohms in series with 10 meg-ohms. Such a high resistance is unlikely to excessively "load" your ion source generator. It is also unlikely you will find such a large value resistance easily available.

So, you can try to make one using what is easily available: 10 meg-ohm resistors. Wire ten in series and you have 100 meg-ohms.,, more or less, depending on resistor tolerance. Do this ten times and you have your 1000 meg-ohm (approximately) resistor. If you want to try this, I would suggest purchasing some clear plastic tubing you can slip over the series-connected resistors, ten or twenty at a time. This will help to minimize "leakage" paths around the 10 meg-ohm resistors that will reduce their effective series resistance.

You might also try experimenting with "shielding" your resistor strings to avoid noise pickup from the surrounding environment. The internal woven shield from coaxial cable, such as the kind used with television sets, works well after you remove the inner insulated conductor by clamping one end in a vice and pulling on the cable jacket. This only works well for short lengths... about two or three feet... before it becomes too difficult to remove the outer jacket with the shield inside. Experiment to find out what works best for you. The shielding will add capacitance in parallel with the attenuating resistance, lowering its frequency response somewhat.
 

hevans1944

Hop - AC8NS
Hmm ... I wonder why they call them negative ion generators?
They sell better that way because it has been shown that positive ions are not good for your health or sense of well-being. Just because you can emit electrons into the atmosphere with a corona needle generator doesn't mean you are creating a significant quantity of negative ions. However, corona discharges from needles are an easy way to create ozone... Mmmmm. Ozone smells even better than Robert Duvall's "napalm in the morning." BTW, Ozone is NOT an ion. Ozone is an unstable molecular form of oxygen.
 
Mouser sells 100M 10KV resistors for about $4.50 each which should simplify construction a bit, assuming the input's HV distributed over a shorter distance does not encourage arcing. If so, perhaps I could seal them in a PVC tube filled with motor oil. Copper might be a better alternative to reduce noise.

Now that you have alerted me, I notice there are similar designs online. Some use added caps, and parallel RC at the scope's input, which I understand would not be needed for my low frequency application.


On the subject of ion generators, are you saying there is no characteristic difference in charge between ions emitted from the positive and negative needles? I have seen some high end ionizers that advertise to produce more negative ions, but I am not sure how this would be achieved.
 

Harald Kapp

Moderator
Moderator
perhaps I could seal them in a PVC tube filled with motor oil.
I suggest silicone - less messy.
are you saying there is no characteristic difference in charge between ions emitted from the positive and negative needles? I have seen some high end ionizers that advertise to produce more negative ions,
Charge cannot be created. Therefore the amount of charge on the positive electrode is the same as on the negative electrode.
However, charge is not equal to "amount of ions". How many ions are produced on each electrode depends on the charge per ion. If, for example, one were to split H2O into positively charged H ions and negatively charged O ions, one would have a charge of 2e (e = elementary charge) per molecule on each electrode, but in terms of number of ions one would see 2×H- ions versus 1×O2+ ion.
 

hevans1944

Hop - AC8NS
On the subject of ion generators, are you saying there is no characteristic difference in charge between ions emitted from the positive and negative needles? I have seen some high end ionizers that advertise to produce more negative ions, but I am not sure how this would be achieved.
I have NO practical experience with atmospheric ion generators, allegedly generating either positive OR negative ions or both. You are on your own as to what you want to believe, or how you think ions can be generated and presumably diffuse through air like smoke.

Your original question was how to (sort of, but not really) measure a high-voltage generated by a low-current source. See your quote below:
All I need is a visual, or relative, approximation of what is happening, not actual measurements.

I will assume you now have sufficient information to answer that question yourself. Good luck with your experiments.
 
Top