# Optimal Components for a Thomson Coil?

I'm interested in making a Thompson's Coil after seeing my University's physics II (EM) professor demonstrate Faraday's Law to us using it. It seems like a fun device to demonstrate various principles of Electromagnetism to lay people (Induced Voltage by Faraday's Law, Resistive Heating, Electromagnetic Levitation (Of rings), Conductivity, etc).

How should I go about building this? I have a functioning knowledge of EM that I learned in class, although we skipped out on inductance unfortunately (The school semester physics layout didn't cover it, god forbid we learn all the school covered-EM topics in the EM physics class right?) At a fundamental level, I get inductance, and am aware of inductive reactance, although I haven't reviewed that chapters in my Giancoli physics text book on it yet. I'm on break for the next month, having just finished a semester at the university. So I have plenty of time to plan and execute the project.

I'm also well set up for woodworking/steelworking equipment for making various components, although I have more of a specialization in woodwork for now. I also have been trying to get into electronics ever since taking the physics II class. I have multi-meters (Including a clamp on one), but don't really have a power supply yet, I also have a 30W Weller soldering iron, and I've wired up a few outlets/switches, as well as all the wiring I did in my Physics II labs.

Functionally, from what I've seen, a Thompson's Coil (Pictured Below)(Source: http://kossover.squarespace.com/journal/2012/5/1/improved-elihu-thomson-coil-jumping-ring.html), really just consists of a coil, with an insulated ferromagnetic iron core, such as this DIY Thompson's coil. I could copy this method outlined in the blog post of the person who made this Thompson's coil (He used the spool, with iron welding rods, which are used in TIG welding, if I'm not mistaken.), but I'm looking to make the best coil that I can, given the time and budget (Around $130-150) constraints, and I'm not sure if this is the design I should go with, hence my post here. Is this an acceptable design? Will it give good results for demonstrations? Will it run on 120V AC? What design choices would make it better? What would be the best inductor material for the core that I can realistically get my hands on? He used insulated (By spray on varnish) iron welding welding rods. I had a recommendation from a retired particle physicist that I previously asked this question to, and he suggested using insulated iron florist wire, which is thin, and he suggested it was a decent inductor. #### Attachments • 3892289-17969826-thumbnail.jpg 38 KB · Views: 545 ## Answers and Replies anorlunda Staff Emeritus This thread caused a lot of concern with the moderators. That is why it was closed for a while. The problem is that our guidelines do not permit discussion of dangerous topics on PF. Although a Thompson Coil per se need not be dangerous, any DIY project that uses electric mains power without suitable safeguards is dangerous. @PhysicistSarah, If the purpose of your experiment is to learn about the interesting physics that a Thompson coil can demonstrate, then you should understand that a smaller scale coil using lower voltages and protected power supplies would demonstrate the same physics and be much safer. For example, it could be done using intrinsically safe SELV power supplies. Are you willing to scale it down for safety sake? Greg Bernhardt and berkeman Sure, the voltage doesn't really matter to me. As long as it can hold a smaller ring. I'm just not sure what material would be best for the core. Someone else I saw used steel welding wire. I'm not sure if that's the best choice for core material that I can realistically get my hands on (Nothing too exotic that would cost$10,000 for instance)

jim hardy
Science Advisor
Gold Member
Dearly Missed
Hi Miss Sara. You seem a handy sort.

I'm a fan of DIY projects.
That said, PF is understandably apprehensive about encouraging beginners to experiment with 120 VAC wiring because of the dangers to curious tiny fingers in the household. We dont want to help anybody hurt themself or others.

The link you posted is an interesting one. It does assume that anyone undertaking the project is already familiar with electrical safety and wiring methods, so it omits those details. In fact, he says:
Note: Some people will want to connect a momentary or push button switch instead of just plugging the coil into the wall or using a power strip. I understand. It is possible that the launcher will overheat and there is live wall current. I didn't do that, but you might have a different standard of safety.
Missed details is what gets people in trouble.
Furthermore they don't teach the "Underwriter's Knot " in most university curricula.
So i'm going to make some common sense safety recommendations, should you undertake to build that Thompson coil.

An aside - in 1962 a kid in my high school electronics class built one. When he pushed the button it'd toss a ring clear to the ceiling.
But his Dad was head of engineering at a local electronics factory so his experiment had really high grade mil-spec wiring.

First - how will you safely power this thing ?
I'd suggest you make yourself an extension cord, 3 conductor of course, with a GFCI receptacle at the end.
I use a simple outlet box akin to this

mounted with screws through the back to a 1X4 plank about a foot long. Mount it near one end.
Mount the GFCI receptacle in it

and wire to a cord & plug of convenient length.
Remember
black wire is "hot" and goes under the brass colored screw, same side as shorter slot
white wire is "neutral" and goes under the silver colored screw, same side as longer slot,
both at top of receptacle marked "Line" .
Green wire goes under the green screw on the receptacle's metal frame.
Now you have an outlet that provides personnel protection. But not overcurrent protection.
I'd accomplish overcurrent protection by placing an inline fuse holder in the cord you make for the main coil.
One of these

here

Place it in the black wire, or if using lampcord as in your link place it in the wire going to narrow prong on the plug.
Be sure to solder and heatshrink all your splices. Of course neatness is a must, check there's no loose strands of wire and no sharp points sticking through the insulation.
Now, put perhaps a 10 amp 250 volt fuse in your holder so you won't trip any breakers or make big sparks.
That's a pretty safe source for housepower.

Mount another outlet box near other end of the plank for your pushbutton switch. It might be a handy place for your fuseholder, too.
I've not yet found a suitable momentary switch. Microwave oven door interlock is a possibility, every junkpile has a dead one.

i find #14 wire about 40 bucks for a 500 foot spool on Home Depot's site.

For a ring to throw i think i'd try one of these and use 1½ inch sink drain pipe to encircle my magnetic core.

Welding rod can get expensive.
Ceiling tie wire or rebar tie wire is cheap. Smaller gage will have less eddy currents.

Straightening small 16 gage will be easier than larger 12.
I hook one end around the trailer hitch on my car, other end around something immobile . I have a steel bar embedded in a large rock for just that purpose but another car would work as well. Most cars have a tow hook under the bumper.
Pull just enough to feel it stretch a teeny bit - as soon as it starts to yield it'll become quite straight. Relieve tension and cut into lengths.
That soft iron wire cuts easily with a wire cutter or lineman's plier..
Of course stand clear while stretching ! If it snaps it'll fly back unpredictably.

That's my "practical" advice.
If you're not experienced with electrical wiring get some help. Surely your school has an electrical lab in EE department, or an electrical guy in your Physics department lab who'd help out an enthusiastic student.

Lastly, i always caution everybody "Tiny fingers are curious. Keep them safe." Any in your household?

old jim

#### Attachments

• upload_2018-5-27_18-34-26.png
22.1 KB · Views: 522
• upload_2018-5-27_18-38-1.png
13.6 KB · Views: 496
• upload_2018-5-27_18-45-56.png
16.9 KB · Views: 487
• upload_2018-5-27_18-49-39.png
37.4 KB · Views: 456
• upload_2018-5-27_19-14-52.png
32.8 KB · Views: 489
• upload_2018-5-27_19-18-38.png
70.2 KB · Views: 490
Last edited:
jrmichler and Greg Bernhardt
jim hardy
Science Advisor
Gold Member
Dearly Missed
All you need is current through your primary..

If your EE lab has a surplus transformer, about 1kva stepdown to 48 volts , that would relieve a lot of safety issues.
An electric welder set for 30 amps would probably do.the job.

Ask around.

old jim

#### Attachments

• upload_2018-5-27_20-1-40.png
23.1 KB · Views: 447
jrmichler, anorlunda and Tom.G
jim hardy
Science Advisor
Gold Member
Dearly Missed
I always use fuses in my projects, and have such a transformer already. It seems like there should be a nice sticky about electrical safety, detailing general safety practices. It should probably be linked in the read this before posting to defuse a lot of these issues before a new thread is started.

I really liked the rebar tie idea for the inductor material. It's a very low gauge which is great for eddy currents as long as they are properly insulated. They also appear to be annealed already, and don't have to be thrown in an over for a few hours.

I tend to be a fan of the "nice/clean" kind of DIY projects. I plan them out meticulously before starting and have everything secured and properly protected, as opposed to switches not grounded, no boxes on switches, things soldered way too close together, the electronics not bolted/held down, cords not tied together, mains terminals not in a box to prevent accidental electrocution.

I've seen a lot of really poorly done electronics projects and it isn't safe and looks kind of ugly as well, like it was really thrown together. Not a really big fan of that type of work ethic. I should add that my linking of that particular article was only to serve as a template and was not an endorsement of his electronics habits.

Wow, I'm jealous of that kid back then. Having a dad that was an engineer and had an "in" at a local electronics factory. Must have been an amazing childhood.

I do have a bit of minor amusement at your motto field though. I think it will be interesting as AI advances and starts being entrusted with more and more things. Unfortunately, accountability goes straight out the window with AI systems. That poorly tested Uber AI system that had a fatality on the road comes to mind. Or the Youtube algorithm. The future seems like a mix of both utopia and dystopia with regards to that.

jim hardy
Science Advisor
Gold Member
Dearly Missed
Thanks Sara. We (or at least I) had no idea whether you were practical and 'electrical safety aware' . Thank you for the reassurance.
And what a great idea about a sticky for the DIY thread.

My motto came about from experience with computers in my nuke power plant which made me a bit of a luddite. I believe they should be used only to monitor and report on a good solid analog control system. They're capable of all too subtle failures.
What amazes me about the automotive CAN-bus is not so much that the one in my sixteen year old Ford Escort is still working after so many years, but that it worked for the first five minutes back in 2002.
That newer cars let the computer control steering, throttle and brakes horrifies me..
I'm the guy who, when making airline reservations, checks the "Equipment" column and finds a route with no Airbus segment.. That's my eccentricity , i just won't ride those overautomated beasts. As my airline pilot neighbors say "If it ain't Boeing I ain't going" .

But i digress. Sorry !

Keep us posted on your project?

old jim

Thanks Sara. We (or at least I) had no idea whether you were practical and 'electrical safety aware' . Thank you for the reassurance.
And what a great idea about a sticky for the DIY thread.

My motto came about from experience with computers in my nuke power plant which made me a bit of a luddite. I believe they should be used only to monitor and report on a good solid analog control system. They're capable of all too subtle failures.
What amazes me about the automotive CAN-bus is not so much that the one in my sixteen year old Ford Escort is still working after so many years, but that it worked for the first five minutes back in 2002.
That newer cars let the computer control steering, throttle and brakes horrifies me..
I'm the guy who, when making airline reservations, checks the "Equipment" column and finds a route with no Airbus segment.. That's my eccentricity , i just won't ride those overautomated beasts. As my airline pilot neighbors say "If it ain't Boeing I ain't going" .

But i digress. Sorry !

Keep us posted on your project?

old jim

Sounds great. I really don't like the double/triple booking schemes that the airlines do, but I understand why they do it. I really love some of the advances made by AI, but some of them are are really nerve wracking. https://aeon.co/essays/judge-jury-and-executioner-the-unaccountable-algorithm

I was somewhat curious on your take on the wire gauge though. Do you think I should use magnet wire or just stick with the 12AWG. And if you do suggest magnet wire, what wire gauge do you think would be appropriate and how many turns for this application?

Willco on posting pictures when done

jim hardy
Science Advisor
Gold Member
Dearly Missed
I was somewhat curious on your take on the wire gauge though. Do you think I should use magnet wire or just stick with the 12AWG. And if you do suggest magnet wire, what wire gauge do you think would be appropriate and how many turns for this application?

Wire type would be whichever is practical for you to acquire. The beauty of 12 or 14 gage house wire is it's so readily available at any lumberyard or electrical supply house. Of course you'd get more turns of magnet wire into the same volume because of the thin insulation.
I'd stick with #12 or 14 because of its low resistance, which you'll want because of your substantial primary current.

https://www.amazon.com/dp/B003JCB1EK/?tag=pfamazon01-20.
i think that spool will be about 800 feet but it costs a sizeable chunk of your budget..

compare to https://www.homedepot.com/p/Southwire-500-ft-14-White-Solid-CU-THHN-Wire-11580858/203401597

....................................................................................

I don't know how to answer "How many turns" . The more the better? B=μ0NI/length .

Thompson found in 1911 that at about 1500 amp-turns per inch his AC magnetic field caused people to see blue lights. So stay below that.

https://books.google.com/books?id=B...DoAQg7MAY#v=onepage&q=thompson's coil&f=false

old jim

#### Attachments

• upload_2018-5-30_0-46-7.png
30.7 KB · Views: 429