Tesla Coil

Ellery and I finished our static gap Tesla coil on July 10, 2011.  You can think of a Tesla coil as sort of a lightning generator.  It builds up high voltage electricity until it is strong enough to break down the air and then it is released as what appears to be a lightning bolt.  Tesla coils are inherently dangerous because of the high voltages and currents required to get one to function well.  What’s more is the longer you want the discharges to be, the more power you need the coil to output and so the dangerous it becomes.  We decided to start relatively small for our first coil to lower the chances of one of us seriously injuring ourselves.  Our primary transformer is a 15000 volt 30ma neon sign transformer (NST).  This should give us an output power of 450 watts (15000 Volts * 0.6 amps).  I expect this coil to be able to put out streamers between one and two feet long once everything is tuned properly.  On to the build.

The first thing we had to do was order a few components.  It can be difficult to find workable NST’s for Tesla coils at a good price and I ended up purchasing ours on eBay.  I wanted to source one locally but that could have taken forever.  The reason I wanted to get the transformer from the beginning is because all of the parts of the coil rely on the specificati0ns of the other parts.  It’s very mathematical.  You have to decide on one variable so you can calculate out the rest.  I decided to choose the transformer size first since I knew I wanted it to be on the smaller side.

 

After the NST arrived, we wanted to test it.  So we plugged it in and tried drawing an arc from it.  We used a length of PVC pipe to keep us insulated from the high voltage.

The next thing we wanted to do was create the secondary coil.  We figured this would be the most tedious and annoying part so we might as well get it out of the way first.  Since we had the NST’s specifications we could calculate what size we needed our secondary coil to be.  We used the incredibly valuable resource of deepfriedneon.com to help choose the secondary coil size.  According to their design guide page, a coil of under 500 watts should have a diameter of between 3 and 4 inches.  We chose four inch pipe, with hopes that if we want to up the power later we can do so with this same secondary coil.  Further down on the deepfriedneon design guide, they recommend having a coil length of about 20 inches if you have a diameter of 4 inches.  Deepfriedneon then recommends having between 800 and 1000 turns of magnet wire on the secondary coil.  With this information, we can calculate what size magnet wire we need for the secondary.

 

20 inch length = 508mm length
508 mm / 900 turns = 0.564mm per winding

The magnet wire comes coated in enamel which adds to the diameter of the wire.  To account for this, you add 10% of the diamter.

0.564mm / 1.1 = 0.51mm diameter magnet wire.

 

The closest wire we could find to match this diameter was AWG 24 magnet wire.  We bought an 11lb spool of it online.

 

Now we had to figure out how to wind this thing.  We used two videos posted online by the Geek Group as a guide, but we made modifications where we felt appropriate.  First we cut a length of 4 inch diameter PVC pipe.  We cut it to about 26 inches since our coil has to be 20 inches long.  We wanted to leave room on either side for mounting, etc.

Next we had to build some kind of jig that would hold the PVC pipe and also allow us to slowly rotate it as we fed magnet wire onto it.  We purchased a 10 foot length of 2″ by 8″ board, cut it into three pieces and screwed it together in a kind of “u” shape.  The following pictures help describe this.

We drilled one 1/4″ hole in each of the smaller side pieces of wood.  These holes will need to line up closely when it is fully assembled.  The reason is that we will be putting a threaded rod through the holes to hold the PVC pipe in place and allow rotation.

Now we screwed the three pieces of wood together.  I think we put two screws in each of the side pieces.

Next we drilled one 1/4″ hole in each of two PVC caps.  We wanted the holes to be as close to the center as possible to allow for smooth rotation.  We were originally going to purchase normal PVC end caps but they cost about $8 a piece and we only needed them to wind the coil.  We found these smaller yellow plug type pieces near the ABS pipe at the hardware store and they worked perfectly and only cost about $1.00 a piece.

Next, we fit one cap in each end of the PVC pipe.

Now it was time to put the whole thing together.  We took a 1/4″ threaded rod and pushed it through one of the holes in the side of the jib.  We then made sure to push the rod through the PVC pipe (by way of the holes in the yellow caps) and out the other hole on the other side of the jig.

 

We put one washer and one nut on the outside on each side of the threaded rod.

Now we had to come up with a way to hold the spool of magnet wire in place, but also allow it to rotate freely so we could easily wind the coil.  I found a coat hanger that had a stiff cardboard tubing around the bottom that fit through the center hole in the wire spool quite well.  I bent the coat hanger to a usable shape and clamped it to the cable with a simple clamp.

Before winding the secondary coil, we took the Geek Group’s advice and wrapped two strands of double sided tape around the PVC form in a spiral pattern (sort of like a barber pole).  The idea is that the tape will help stick the wire to the pipe and prevent it from loosening up and undoing your work.  We honestly found that the tape got in our way and we probably would have been better off without it.

Now it was finally time to start winding the secondary coil.  We wrapped a couple turns of magnet wire around the end of the pipe and taped it down well with masking tape.  We wanted a few extra turns around the ends so we would have something to work with later.

Now, we just had one person slowly turning the PVC pipe (Avoiding contact with the double sided tape) while the other person guided the wire onto the form.  It is important to keep tension on the wire and prevent any overlapping of coils or gaps between coils.  This process took about an hour to complete.

 

 

Once we reached the 20 inch mark we wrapped a few extra turns of wire up in the mounting section of the pipe and taped it down good with masking tape.  The final thing to do to the secondary coil was to varnish it.  Varnish isn’t necessary but supposedly it helps insulate the wire and prevents running arcs which can destroy the coil.  It also protects the wire insulation in case you accidentally bump or ding it.

I put on six coats in total.  I left at least 4 hours between coats for the varnish to dry.  I noticed that when I applied the varnish bubbles would appear in the wet coat, but they slowly dissipated on their own.  I read that the amount you want to put on is as much as possible but not so it drips to the bottom as it dries.  I put in a few coats that were too thick.  I can tell because it left these large varnish bumps on the bottom of the pipe.  I think the best way to prevent this would be to motorize the jib so the pipe is constantly turning as it dries.  I don’t think the bumps will effect the coil performance at all, but it doesn’t really look nice.

 

That’s it!  We were finally done with the secondary coil.  The next thing we tackled was to build a bucket capacitor.  We pretty much followed the guide posted by the Geek Group to the letter on this one.  The first step was to get a 5 gallon bucket and fill it about 1/2 of the way with the hottest water we could get.  I just used hot tap water.  Then we mixed in two full containers of table salt.  The idea is to mix in as much salt as possible into the water.

 

Next, we poured some of the water into a pitcher for easy pouring and we used that to fill up 12 empty Corona beer bottled.  We filled each bottle up with the salt water to the point where the neck starts.  We got them as close to the same levels as we could.

Next we placed all of the filled Corona bottles in the bucket and removed water from the bucket until the water level in the bucket was about the same as the level in the bottles.

Next we poured in the motor coil.  We used the cheapest motor oil we could find at Walmart.  This acts as an insulator and prevents the outside water from reaching the water on the inside of the bottles.  It also is for corona suppression.  We filled up all 12 bottles, and then poured the rest in the rest of the water.  We ended up using exactly 6 quarts.

 

Now we had to connect all of the bottles together electrically.  We cut about 13 or 14 lengths of 10 AWG wire and stripped the ends back.

Next we bent all of the wires into “u” shapes and connected all of the bottles to each other.  We tried to make sure none of the wires touched the bottoms of any of the bottles.  All of the wires have to penetrate through the oil layer and get into the salt water.  It is important that none of the bottles are electrically connected to the salt water that is just in the bucket.

Next we had to make two wires to connect to the outside terminals.  We cut two longer lengths of wire and crimped on a connector to one end of each wire.  I’m not sure what the connector is called, but here is a photo.

 

 

Next we drilled a small hole into the side of the bucket (Above the oil level) and one in the bucket cover.  We threaded in a small 1/4″ bolt through the crimped connector and hole in the bucket and fastened it from the outside with a washer and nut.  We placed some silicone caulk under the washers to help form a more water tight seal in case the bucket ever tips over or splashes around.

 

 

 

As you can see from the photos, the wire on the side of the bucket went into the bucket salt water.  The wire that connects to the top terminal goes into one of the bottles.

We mad sure the cover was on nice and tight.  Then we cut another length of wire and crimped a connector onto both ends. This wire will get bolted from one terminal to the other.  This ensures that no charge can build up in this capacitor while it is idle.  It’s a safety precaution.

Finally we added a warning label with a black Sharpie to make sure that anyone who sees this thing realizes it is dangerous.

After the capacitor was constructed, we decided to test it.  We hooked the NST up to the capacitor and tried to draw an arc.  We figured if the capacitor was working we should be able to draw an arc and the arc would probably be stronger than it was with just the NST.

That was it for the capacitor!

Next we started working on the base platform.  We purchased two 1/4″ pieces of plywood.  Both are 2 feet by 2 feet.  One of them was going to be the primary coil platform.  This platform will hold the primary coil and also have a mount for the secondary coil.  The first thing we did was to measure a spot as close to the center of the board as we could get.  Then we laid out a toilet flange as close to the center as possible and marked three spots to drill mounting holes.  The toilet flange will eventually be used to hold up the secondary coil.

 

After the holes were drilled out, we mounted the toilet flange onto the board.

 

Now that the flange was bolted on, we had to tackle the primary coil.  We used the program Tesla Map to get these calculations.  After putting in all of the parameters we knew, we determined that we needed to make the primary coil using 1/4″ flexible copper tubing.  We weren’t sure exactly how many turns we would need because we don’t know the exact value of our bucket cap.  We figured if we used a full 15 foot length of pipe, that would give us about 13 turns in the primary coil, which should probably be plenty.  We can then use an alligator clip to tune the coil down if needed.  Tesla Map told us that our turns should be 1/4″ apart all the way around.  We had to come up with a way of mounting the coil.

We decided to use plastic cutting board material.  We figured it would be a better insulator than wood and be less flammable.  I cut four 1 inch scripts of cutting board with the band saw (Well I cut five but we threw one out).  Each strip is about 8 inches long.

Because I did not cut them perfectly straight, I sanded down one side of each strip to make it flat for easier mounting.

The next part is tricky.  I had to cut grooves in the cutting board pieces so that the copper pipe would spiral around the secondary coil, staying about 1/4″ apart all the way around.  Tesla map gave us the starting diameter (I think it was about 6 inches), so I marked three inches from the center on all sides.  We knew the slots on each piece of cutting board had to be 1/2 inch apart because that would give us a 1/4″ gap for each turn.  Since we had four standoffs, we had to calculate how much each standoff would differ from the others.  This turned out to be 1/8 of an inch.  So if we cut 13 slots in one standoff, we had to basically shift them all down 1/8 of an inch for the next standoff, and so on. This will force the tubing into a spiral pattern.  I also still had to come up with a way to cut grooves.  I decided to drill 13 1/4 inch holes in each standoff, and then cut out the extra plastic with the band saw.  The photo below illustrates this.  The photo also shows how each piece has the holes offset just a little bit.

Once we had all of the slots cut, we had to mount the standoffs to the primary coil platform.  I didn’t want them to be too permanent, just in case we wanted to replace the coil later, but it did have to be sturdy. I decided to tie them down to the platform with zip ties.  I drilled two holes in each standoff, and then four holes in the platform for each standoff.  I then threaded a zip tie through each one and tied the standoffs down good and tight.

 

Next, we had to wind the primary coil.  The copper tubing is a lot for bendable than I originally thought.  I just started on the inside and popped the tubing into the slots as I slowly went around.  I tried to bend the tubing to keep each turn about 1/4″ away form the last turn.  It’s not perfect but I think it’s about as good as we will be able to get with this method.

There were a few parts of the primary coil that kept popping up from the standoffs.  To fix this, I just used a few more zip ties to tie down those parts of the coil.  Rather than drilling more holes, I just ran these zip ties through the old ones.

 

That was it for the primary coil!  The next thing to work on was the primary spark gap.  There are a few different designs to work from.  I decided to use a design where you place several small chunks of 1/5″ copper pipe next to each other about 1/8″ apart.  You then use alligator clips to select which pipes are included in the gap.  If you have 4 pieces of pipe, that equates to about a 1/2″ gap.  If you skip two pieces of pipe, then it is only a 1/4″ gap.  This makes it easy to tune and is reportedly more reliable than one large gap.  The first thing I did was to cut several small sections of 1/2″ copper pipe.  Each chunk is about 3 inches long.  Using a copper pipe cutter was much easier than a hacksaw.

After the pieces were cut, I measured about 1/2″ from either end and drilled a small hole.  The hole had to be big enough for the small bolts that would bolt it down.

Next I cut a piece of 4″ PVC pipe about 7 inches long.  The length probably isn’t really that critical but I didn’t want it too big so it would fit nicely on the mounting platform.

Now we had to drill holes in the PVC pipe to mount the copper pipe in the inside.  I marked down about 1″ for the first hole.  Then I held up a section of copper pipe and used it as a guide to mark where the second hole should be.  Once I had one section of pipe mounted, I placed another section on the inside and basically just eyeballed it and estimated where I would need to drill for  about a 1/8″ gap.  I then followed the same procedure for measuring the holes on the outside of the pipe and drilling.  We ended up mounting 5 sections of pipe, but it may turn out that we need more.

 

You can see the bolts I used are a bit too short.  I barely managed to screw on the nuts so I might have to replace them with slightly longer bolts.  I also still want to mount a 12V computer fan to the top of this thing to low air through it while it is in operation.  This will help to remove ionized air caused by the spark gap to keep the gap firing more consistently.  To finish the spark gap, we just screwed it directly to the bottom platform with small wood screws.  We then took a 5″ 12V computer fan and mounted it a few inches away from the PVC pipe opening.  This fan blows air through the gap.  It helps keep it cool for longer run times as quenching the ionized air out of the gap so it will fire more consistently.  The fan is powered by a separate 12VDC power supply.

Next we build the safety gap.  The safety gap is made of two brass L-brackets, two 1/4″ bolts, 4 1/4″ nuts, 2 1/4″ end nuts and a block of wood.  The end nuts are these rounded end cap things that screw onto the end of the bolt.  The rounded edge is ideal because it will allow for a more consistent spark.  The spark gap is adjustable bu moving the nuts on each bolt on either side of the L brackets.  To tune the safety gap, we hooked it up to the transformer with nothing else.  We adjusted the gap size so that when we turned on the transformer it was barely able to spark across the gap.  Any bigger, and there would be no spark.  We screwed this to the bottom platform of the coil.

Next we had to built the top load.  We originally had a larger top load but we traded it for a smaller one to ensure that everything was working properly.  We built the top load from 4″ diameter aluminum ducting, aluminum tape, a pie tin and aluminum foil.  We also used a nylon bolt, nut and one of the yellow plastic end plugs that we originally used when we made the secondary coil.  Based on some calculations from TeslaMap, we determined that our toroid should be about 12″ in diameter.  We stretched out the aluminum tubing and cut it down so we could curl it into a donut shape of about 12″ in diameter.  We used the aluminum tape to tape the ends together to hold the donut shape.  We wrapped the entire thing in aluminum foil and smoothed it out as best as we could.  We figured this would be smoother than the tubing since the tubing had lots of ridges.  We cut the edges off of one of the pie tins and drilled a hole in the center.  We threaded a nylon bolt through the yellow end cap and pie tin and used the nylon nut to hold it together.  It is important to use nylon for this because a metal bolt would be conductive and create a breakout point for arcs.  We then used aluminum tape to hold torroid to the pie tin.  We also ended up using a roofing nail as a breakout point later on.  This just got taped to the toroid with aluminum tape.  I will have to add some photos here to make it more clear.

Next up is the strike rail.  This is basically one turn of copper tubing that is only attached to RF ground (a ground rod).  It goes above the primary coil.  The purpose is to attract arcs so they arc to ground instead of the primary coil.  Arcing the the primary coil can damage the Tesla coil and also be dangerous.  We mounted four blocks of wood at the corners of the top platform to hold up the strike rail.  We then drilled a hole through each block of wood and used zip ties to hold the ground rail in place.  I believe it is important to not connect the ends of the strike rail together so it doesn’t have an inductance of its own.

Next we had to mount everything together.  The bottom platform is where the transformer, spark gaps, and fan are mounted.  It is just a 2 foot by 2 foot piece of 1/4″ plywood.  Eat each corner we mounted one 8 inch length of 2″x2″ wood as standoffs for the top platform.  We also attached one 2″ piece of 2″x2″ wood on the underside of each corner to act as feet.  The top platform was then screwed to the 8″ standoffs.  That was it for mounting.

The final step was to wire everything together.  The circuit is very simple.  The transformer has two leads coming out of it.  Each of these are about a foot long and each one mounts to one side of the safety gap.  This way, if nothing else was hooked up and the power was turned on, the safety gap would constantly have a spark.  Next, each side of the safety gap is attached to each side of the adjustable primary spark gap.  It attaches on this side with alligator clips.

On the other side of one of the alligator clips is a wire that goes to the primary coil.  To attach that, we drilled a hole in the top platform and fed the wire through the hole.  We stripped off about 2 inches of wire and just wrapped the wire around the end of the primary coil.  This was the end that is closest to the secondary coil.  This could use a more secure connection, but it works.  The other alligator clip has a second wire that goes to one side of the bucket capacitor.  The other side of the bucket capacitor has a wire that has another alligator clip.  This alligator clip connects to the other side of the primary coil.  This clip can move around on the primary coil for adjusting the tank circuit frequency.

Next we had to attach the top wire from the secondary coil to the top load.  We burnt off the enamel at the end of the magnet wire at taped the wire to the under side of the toroid with aluminum tape.  The top load in the below photograph is our older, larger top load.  That is why it looks different than in the description of how we built our top load.

The bottom wire from the secondary coil has to attach to RF ground.  We drilled a small hole at the base of the secondary coil and another small hole at the base of one of the strike rail standoffs.  We routed the magnet wire from the bottom of the secondary coil along the underside of the top platform and then back up one of the strike rail standoffs using those holes.  We burnt the enamel off the end of the magnet wire and wrapped it around the strike rail.

For an RF ground, we took the 1/4″ threaded rod from earlier and pounded it into the ground outside of my house with a hammer.  We then attached a wire that goes from the ground rod to the strike rail.  This provides an adequate ground that is not attached to the house wiring.

I believe that was everything in a nutshell.  After it was all wired up we had to tune it by adjusting the spark gap size and the primary coil size.  We found that 2 gaps and 9.5 primary turns is about as good as it gets.  We get about 11″ long sparks this way. Here are some videos of the coil in action!

The sound is a bit fuzzy in the videos. I figure it’s from the Tesla coil causing interference with the camera’s audio circuit. It’s actually very loud in person.