• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Building a DRSSTC Pt. 10 - Building the Secondary

    Blog entry posted in 'Building a Dual-Resonant Solid State Tesla Coil', September 04, 2014.

    Hi everyone, welcome back to my build of a Dual-Resonant Solid State Tesla Coil!

    Now that we've covered the basic theory behind how a Tesla coil works, and the components that make up a DRSSTC, it's time to get on with the actual build!

    It's important to remember that a vast majority of the planning and design must be completed and solidified before you begin the build. As I mentioned before, the design is always going to change depending on various factors, so you should make sure all your parts are designed to work together perfectly before you actually start building.

    I have found that an excellent place to start is deciding what frequency your coil will run at. The best way to calculate this is to start by designing your secondary coil. My particular secondary has gone through several revisions already, the first of which reused a secondary from an old spark gap Tesla coil that I built about 10 years ago. It was a 4" PVC tube (4.5" outside diameter) with a 20" coil of #26 AWG magnet wire wrapped around it. However, even with the largest topload I had--4.5" minor diameter x 15.75" major diameter toroid--the resonant frequency was around 210kHz according to JavaTC. The coil itself (with a smaller topload) looked like this:


    While this may have worked, the frequency was a bit high for my liking. I really wanted to keep the frequency close to 100kHz, which is about what the IGBTs I was planning on using can handle. In order to obtain this frequency, however, I really had to overhaul my design. I decided to construct a completely new secondary coil that would get me closer to the base frequency of 100kHz. When constructing resonant tank circuits consisting of an inductor and a capacitor, there are two ways you can lower frequency: Increase the capacitance or increase the inductance (or both). I decided to try both. As mentioned, I made a large topload with a 4.5" minor diameter and a 15.75" major diameter. This physically larger toroid provides a higher capacitance with respect to ground, thus lowering the frequency. Unfortunately, simply putting this larger topload on my old secondary wasn't enough, so I decided to increase the inductance of the secondary as well. Once again I started with a 4.5" outer-diameter PVC pipe. This time, however, I bought a few thousand feet of #34 AWG magnet wire.

    88139 88140

    Thinner wire has a higher resistance, but it allows more windings on the coil which means a higher inductance. Ten and a quarter hours and over 1800 feet of wire later, I finished winding my new secondary. It ended up being 12 inches long with approximately 1500 turns, which gave me an inductance of around 87mH. To give you an idea, my old 20" coil of #26 wire only had an inductance of around 27mH!


    The next step was to coat it in polyurethane. There are several reasons why we do this: One, it protects the fine secondary wire from being damaged by bumps and dings, Two, it helps prevent arcing from the middle of the secondary (instead of from the topload) and from arcs racing across the surface, and Three, it gives us a nice, smooth, glossy surface that just looks gorgeous when done right.

    Doing it right, however, is the tricky part. In order to get a smooth surface from the polyurethane, you really must have the coil in a jig that spins it at a fairly low speed. Setting the secondary horizontally and applying the polyurethane works, but you are almost sure to get air bubbles and drips if it is left stationary. This results in a very bumpy, horrible-looking secondary coil. It is often recommended that you apply up to 16 coats of polyurethane with a foam brush or credit card to a spinning secondary and leave it spinning for a few hours (or overnight, depending on if your poly is fast-drying or not) between coats. If there are air bubbles when a coat is dry, you can gently sand them down using 400-grit (or finer) sandpaper. Just MAKE SURE not to sand into your wire!

    A wood lathe is probably the best option for turning your secondary as you apply the poly and let it dry, since they tend to have reasonably low speeds. Unfortunately I didn't have one, so I was forced to be a bit more creative. I had an old 120VAC vacuum cleaner motor that I clamped into my bench vise. The motor shaft was threaded, so I drilled and tapped the inside of a bolt so that it would screw onto the shaft. This gave me a slightly better surface to attach the secondary. I then wrapped a few layers of duct tape around the bolt until it would fit snugly into a long, straight piece of 1/2" copper pipe. I then cut holes in two cottage-cheese container lids (they were a perfect size to fit over the end of the PVC) just big enough to pass the copper tube through. Now I had a motor (the vacuum cleaner motor), and an axle (the copper pipe) that would hold the secondary coil and spin it. This presented another issue, however: Vacuum cleaner motors generally spin at a very high speed. For this reason I connected mine to my Variac so that I could adjust the voltage being supplied to it. Please note this is not a very good way to do this. I would strongly recommend setting up a pulley system with a large ratio of the secondary coil diameter to the motor shaft. This is a much better way of doing it.

    Anyway, I found that running my motor at around 50VAC would spin the secondary at a good speed, but due to the unexpected load it would get very warm after only a couple of minutes. This meant I would not be able to keep it spinning as it dried. Unfortunately, for this reason, my secondary ended up with some air bubbles and drips. However, if I can set up a new jig for turning the coil, I will probably sand it lightly and apply a fresh coat while the secondary is spinning, which should clean it up nicely.


    Using JavaTC, which I talked about in a previous post, I determined that my new secondary with the 4.5" x 15.75" toroid would give me a resonant frequency of around 122kHz--Much closer to the 100kHz that I was looking for!

    The following is the JavaTC input and output for my Tesla coil:

    88148 88149

    The last photo gives you a general idea what the final product will look like. I'm going to skip ahead slightly and show you how the actual coil and the JavaTC output compare:


    Just look at that! Pretty darn close!

    By the way, I'm not going to do a separate entry on making the topload, so I'm just going to explain how I made it now. Its construction is very simple. I first took some 4" aluminum dryer duct and bent it into a donut shape. I then used some aluminum foil tape (available at most hardware stores) to tape the ends, and then I taped around the entire outside of the duct. This gave me a nice, smooth (ish) surface free of holes or sharp ridges that were in the dryer duct. The final step was to fill the hole in the center. For this I used a 9" aluminum pie plate and taped it in (both from the top and bottom) with the same foil tape I used earlier. It's that simple!

    Now that the secondary and topload is done, and we have our data (resonant frequency, especially), the next post will cover the construction of the primary resonant tank circuit!

    Thank you very much for reading, and if you have any comments, questions, or feedback, feel free to send me a PM or, even better, leave a comment on the blog!

    killivolt, September 04, 2014
    I was thinking to myself that sure looks like a Dryer Vent hose. Very nice. It's much more complex than I thought in the beginning of the blog; I thought you just calculated things out and spin some wire around some PVC? It'll be some time before I attempt anything like this........ Nice work.
    DerStrom8, September 04, 2014
    Thanks KV. Spark gap coils tend to be much more forgiving when it comes to the math--in a lot of cases you can just throw a bunch of parts together and get at least something, then you can tweak it from there to get a good output. However, with solid state coils, especially DRs, if you don't do the math you will likely destroy components. It doesn't take much for parts to go pop when running a DRSSTC, so it's very important that you do a lot of math and design first. This is a completely new thing for me--I have never built a solid state Tesla coil before, so I'm learning and writing as I go :) Matt

EE World Online Articles