High Voltage Corona Nuts




Introduction: High Voltage Corona Nuts

About: Dave Reens here. Supposedly I have degrees in math, engineering, and physics from MIT, but really I just like tinkering. Special thanks to my wife, my son, my cat, and last but not least my faith in Jesus fo...

WARNING! High Voltage is dangerous. All work with high voltage should be done under the auspices of an organization with proper safety and training policies implemented. My system is a bit of an extreme case, since it runs at 25kV and 30A. This means it has a peak power of 1000 horsepower, so getting a shock would be like being hit by a sports car. In order to do this work, I am CPR/AED certified by my employer, have gone through high-voltage safety training, and have a defibrilator within 10 feet of my apparatus.

In High Voltage systems, sharp edges must be avoided, since they amplify electric fields and lead to corona discharge or sparkover. Corona discharge is a process where gas is ionized by the high electric fields, and then flows through the air leaking current and corroding parts. Pointy solder joints, crimps, or screw terminals are not recommended in systems above a few kilovolts or so. Commonly it is recommended to join circuit elements with excessive solder balls to avoid any pointy places. Check out these NASA workmanship standards for example.

Solder balls can be annoying, especially in a circuit that needs continual maintenance, for example in a physics experiment such as the one I'm currently working on for my PhD, where the high voltage system is constantly being altered to serve slightly different functions.

One way around this is to use corona nuts. A Corona nut is an intentionally large, rounded electrode, either a sphere or a toroid, used for connecting high voltage components. They are available commercially, for example from Ross Engineering. For my experiment, I decided to make my own to be more cost effective and to have greater customization.

The main idea is to connect components by pushing their leads into a conductive sphere, which has setscrews embedded that can be used to lock the leads in place. In the attached images, you can see many of these spheres in place in my high voltage system, connecting wires, resistors, fast push-pull switches, and capacitors. The links here go to the actual manufacturers of the components we use in our system.

While I'm on the subject, I'll also mention that my system operates between +/-12.5kV, with square wave pulses ranging from 1-200kHz and peak currents close to 30A. Corona nuts like these are useful for circuits in air for a wide range of voltages, although sphere size and distance from other conductors will need to scale up for higher voltages. In my 25kV system, I use 1" spheres separated by at least 3" from other conducting surfaces.

Teacher Notes

Teachers! Did you use this instructable in your classroom?
Add a Teacher Note to share how you incorporated it into your lesson.

Step 1: Purchase Metal Threaded Ball Knobs

Begin with metal threaded ball knobs purchased from McMaster.com for another supplier. Choose ball knobs with a common thread such as 1/4-20 or with a thread that matches an existing thread you want to connect to. For example, high-voltage feedthroughs from MDC vacuum corps have 1/4-28 thread. My High voltage system connects to these vacuum feedthroughs which finally attach to our high voltage load in vacuum. If you want to save some money, you can instead begin with plain spheres and add the thread from scratch.

In some cases, a metal threaded ball knob may already be enough to connect HV components, for example a wire may be connected to a threaded stub by pinching it at the bottom of the metal threaded ball knob.

Aluminum or brass are good choices because they're easy to machine and polish. Harder metals can be better as electrodes in vacuum (See Latham's High Voltage Vacuum Insulation) because it is harder for electrons to tunnel out of them, but that is not the dominant process for spark-over or corona in electrodes operated in air like these corona nuts. So don't waste your time trying to machine these out of stainless steel.

Step 2: Extend Thread Through Ball

The threaded ball knob will only have thread partway through the ball. The design principle for these Corona nuts is that you can push wire into the sphere and tighten them with a setscrew, but the best way to do this without damaging the wire is to have setscrews coming from both sides, so as to compress the wire rather than pinching it.

Select the appropriate aluminum tap hole for your thread. For 1/4-20, this is a #7 drill. Use an alignment tool to match the existing thread with the lathe axis, mounted in a lathe collet which matches the diameter of the sphere. Use a collet stop so the sphere doesn't get pushed in, since the collet won't have a good hold on the sphere as it is designed for cylinders.

A shortcut to avoid the alignment step is to push the #7 drill-bit into the existing thread, mount the drill-bit on the lathe, and push the sphere into the collet. Unorthodox, but if you're making many of these, avoiding tool changes will save a lot of time.

Take out the sphere and widen the hole where it meets the back of the sphere. This way, when you extend the thread, it won't stick out of the sphere and get flattened when you polish the sphere later.

Now take a tap and begin on the already tapped side. You can't start on the new side, or the threads won't lineup. Extend the tap through the hole you have drilled. Make sure to use lubricant, and to turn the tap backwards a half turn for every full turn forward to break off burrs.

Step 3: Add Connecting Holes

Drill two or three holes all the way through the sphere, intersecting the thread at the center. Choose a drill size so that the wire gauge can fit in the holes, or any other elements you wish to connect, for example high-voltage resistors. I used #47, close to 0.080". Use the lathe so that the holes will be centered on the sphere. Use a centering drill to make sure the holes aren't thwarted by the curved surface.

Afterward, run a tap back through to clean up any internal burrs. At this point you can also put in the setscrews, peaking through the holes you've just added to make sure the setscrews are just barely impinging on the connecting holes.

Step 4: Polishing

Use a Scotch-Brite wheel to lightly brush the area around the holes you have drilled and the taps. Alternatively, put the sphere in a tumbler for an hour, but you may lose any shiny finish the spheres originally came with.

Clean the sphere well so no oil residue remains on the setscrews to reduce conductivity of the joint.

You may wish to add a thin layer of super corona dope to the sphere. This is an additional corona suppressing agent, commercially available. Don't overdo it or the setscrews might get stuck.

Be the First to Share


    • Backyard Contest

      Backyard Contest
    • Silly Hats Speed Challenge

      Silly Hats Speed Challenge
    • First Time Author Contest

      First Time Author Contest

    7 Discussions

    Saesun Kim
    Saesun Kim

    10 months ago on Step 4

    Thank you for sharing. I am having some trouble working with corona discharge. In order to improve it, I am thinking of some kind of coating. Do you think the corona dope is vacuum compatible? Thank you!


    Reply 10 months ago

    Actually if vacuum isn't critical you could also consider potting in epoxy or oil, but this will effect capacitance unless you shell out for low dielectric oil and you'll have to make sure to use a low viscosity oil / epoxy to avoid bubbles, which can go into corona and heat up and explode (personal experince there). Anyway I could go on all day, so you should tell me about what you're actually doing :-)

    Saesun Kim
    Saesun Kim

    Reply 10 months ago

    I really appreciate your reply! I did not expect to get the response, but I am really excited to talk about my work! My name is Saesun Kim at the University of Oklahoma. I am working on quantum optics research with Dr. Marino.

    I am trying to insert the Rubidium inside of the vacuum chamber, and heat it up to ~120C. Which will be close to 10^-2~10^-3torr.

    Initially, before I insert the Rubidium vapor, vacuum is near 10^-8. At this point I do not have glowing, but there has breakdown voltage at ~25 KV (my spacing is <2mm apart). It starts glowing around <10KV when I insert the rubidium and heat it up. I want to reduce the effect from the corona, so I can apply high electric field to rubidium atom.

    I think vacuum is only critical because I want to prevent the interaction of the rubidium. I was thinking about the dielectric coating of the plate, and changing the geometry of the plate.

    If you have any question, please let me know. I am willing to share the geometry and design with you. If you do not mind to take a look, please leave me a email address to share. Again, Thank you so much for your response! :D


    Reply 9 months ago

    Hi Saesun,
    Thanks so much for telling me about this. That is an exciting and challenging project! I have had trouble applying 200 kV/cm in a 1e-8 torr vacuum before, so I imagine that getting to 125 kV/cm in your low vacuum with Rb impurity will be very tough.

    Can you get away with admixing N2 or would the collisions interfere with what you want to study? I think you may need to do this in a beam- let a small jet through a hole and then have it fly through plates somewhere so that you can avoid having the Rb too close to the plates. Obviously that’s a lot more experiment complexity and reduced signal.

    Sorry to leave you hanging there. Just started a postdoc and totally forgot to reply.

    Saesun Kim
    Saesun Kim

    Reply 9 months ago

    I appreciate your reply. Congratulation for the new postdoc!
    Since we need to reach a certain density and experimental complexity, but atomic beam will be hard to implement, but Mixing with N2 or buffer gas is a good idea to try. Currently, I am just trying to modify the geometry, but I will think about how to insert the other gas into the chamber. Thank you.
    I wish you have the best luck in the new research group.


    Reply 10 months ago

    Hi Saesun, I wouldn't put corona dope in a vacuum. Then again- how good of a vacuum? That could also be part of your challenge, arcing gets worse when you start pumping vacuum before it gets better:


    If your vacuum isn't so great, corona dope could help. If its a good vacuum, then first of all it isn't corona. Corona is when gas is ionized but there's no breakdown:


    With vacuum, there's no reason the ions wouldn't go to the cathode, so its likely field emission or other forms of "pre-breakdown" current, see this abstract for a definition of terms of sorts for the field:


    I would recommend pressurizing with nitrogen if vacuum isn't critical to what you're doing, pressurizing with SF6 if that isn't enough and you have money, or polishing whatever is arcing. Even if your vacuum doesn't need to be good, corona dope will develop "patch charges" which can be quite significant and will periodically break down across the vacuum.


    4 years ago

    Thanks for sharing!