Testing the Synchronous Spark Gap Circuit

Tonight I have just done a little bit of testing. I wanted to make sure all was OK with the spark gap circuit so first I made sure everything was physically OK. I checked that the disc holding the four tungsten spark gap electrodes was fastened securely to the mandrel. Also that the tungsten electrodes were secure and the spark gaps were set at the correct distance (0.5mm + or - 0.2mm).

All was fine so I proceeded. I plugged the main power supply module into the mains and connected the phase shift module to it. Then connected the phase shift module to the SSG circuit iec socket on the side of the main tesla base module. The main supply to the neon transformer was left disconnected as, at this stage, I only want to test the SSG circuit.
Turned the phase shift module to its minimum setting, a final check then I threw ON/OFF switch.
The SSG motor hummed into life and spun up to full speed very quickly. I sat and watched for a while. Nothing seemed to be smoking or melting and the motor was running smooth. There was virtually no vibration, the electrode disc seemed to be balanced very well. I decided to give the phase module a try and  rotated the variac knob a few degrees. Obviously there was no viewable change to the rotation of the SSG but I could hear the motor lose and regain its lock on the frequency each time I changed the position of the phase shift variac (well I think that's what I was hearing). On further increasing the phase module the SSG motor would completely lose lock and stop but regain and restart when the setting was reduced. By this time perspex case around the phase shift module was starting to cloud up with what looked like condensation. A good time to end this first test run.
After unplugging and further inspection I concluded that the fogging was just solvent being driven out of the varnish on the coil windings in the phase shift module.
I wanted a visible confirmation that I was actually achieving some degree of phase shift. First, I thought if I illuminated the SSG with a fluorescent strip light I might be able to "see" some phase shift as the phase shift module is adjusted due to the (50HZ) flashing of the strip light. Unfortunately no access to a strip light prevented that little test. After explaining what I was trying to achieve Damien came up with an idea. He could create a 50HZ strobe with a simple little program on an Arduino microcontroller.

Damien explains:

A 50Hz wave has a period of 0.02 seconds.  This is composed of half-on and half-off output.  Repeat this procedure and you get a fairly accurate 50Hz square wave.  Hook this output to an LED and you have a simulated strip light running at 50Hz.  While it isn't in phase with the mains (Unless you're really lucky when you turn on the Arduino!) and it's not massively accurate because of the inherent overheads introduced by the delays in the microcontroller and timers, it's a good way to get a strobing effect and get an idea whether the phase-shifter is working or not.  Back to dad...

Here's a pic of the little unit Damien put together. We switched off the lights and fired up the SSG circuit.

Without adjusting the phase module the Arduino strobe was shone on the spinning electrode disc. The strobe effect worked well, the electrodes appeared to be rotating anticlockwise very slowly. This was due to the slight difference in frequency between mains and the Arduino strobe. If the frequencies were exactly the same the electrodes would appear static. I adjusted the phase shift module and this resulted in a change to the slow rotation of the electrodes. This change only occurred during the adjustment of the phase shift module. I think this pretty basic test is a good indication that there is a change in phase. Below is a video of the process, it's just about possible to see the moving electrodes as the phase module is adjusted, it was more obvious to the naked eye. I think the video camera frequency also combines with the other frequencies to complicate matters.

If you like this blog you can show your support by one or all of these. 1. +1 my blog and email it to a friend. 2. Follow me.... It's good to know someones interested. 3. Leave a comment.... All are appreciated.

Primary Slider Clamp

Todays horrible weather gave me a good excuse to do some more work on the Tesla. My next task is to make some kind of easily moveable clamp for the outer connection to the primary coil. It needs to be easily moveable so you can adjust its position (tap point) on the primary coil. The position of the clamp on the primary coil determines the energised length of the primary coil and hence its resonance. The tap point is moved so you can adjust the resonance of the primary to match that of the secondary coil. Well there's the basic tech stuff, here's how I made the clamp.

I started off with a chunk of copper that I picked up off ebay for about £8. It was about 3" long, 1" wide and 3/8" thick. The clamp would have to be of an unusual design because of the perspex disc that covers the whole of the primary coil. I will have to clamp onto the primary coil from below and, depending on the eventual tap point, may have restricted access due to the perspex disc underneath the primary.

After taking a few measurements I made a quick drawing using iDraw on my MacBook just to work out if this design was feasible. I decided this design should work even if the tap point is in the more inaccessible places on the primary coil. The length and shape of the righthand end of clamp may change as the design develops.

The copper block was marked up and sawn into the rough shapes required. The parts were cut slightly larger than required as they would be milled down to the required dimensions.

The off-cut removed was marked up to form the second part and roughly sawn out again oversize to allow for milling.

I fitted a 10mm milling bit and did the first lengthways run to produce the correct thickness required (6mm).

Here's a vid of a bit of the milling process. Milling copper is pretty easy as it's quite soft, I had to use the milling vice because the small height of the pieces I was working on. It's good practice to clamp work as close to the milling bed as possible so any slop in the bed is reduced to a minium. I couldn't do this with these pieces so I kept cut depths small.

After milling the two pieces were clamped together and drilled at the intersect to form the clamp jaws. The primary coil is 1/4" tubing so I used a 1/4" drill bit to form the jaws. To get the jaws to clamp tight on the tube I will remove some of the material at the vertical intersect just below the drilled hole.

A test fit of the piece works fine. It's a tight fit between the coil and the perspex disc underneath so I will mill another 0.5mm off the bottom of the lower assembly.

This picture really shows the need for this particular design of clamp especially if the tap point lies inside the external diameter of the primary coils support disc. Next session I will reduce the clamp thickness, join the two pieces together with some kind of slide mechanism and add the electrical connection.

If you like this blog you can show your support by one or all of these. 1. +1 my blog and email it to a friend. 2. Follow me.... It's good to know someones interested. 3. Leave a comment.... All are appreciated.

Primary Coil Clamp Pin

I spent sometime yesterday trying to figure out a good way to make the required electrical connection to the innermost end of the primary coil.

As you can see in the picture the connection needs to clamp onto the end of the 1/4" copper tubing and then pass through the 12mm perspex disc underneath the primary coil. In this pic the disc is still covered with protective plastic coating.

This is a side on pic taken level with the primary coil. You can see the anti-strike perspex disc above the primary coil and the 12mm thick disc below. After long consideration I decided that I needed a fitting that would clamp on to the end of the primary and have a protrusion that would pass through a clearance hole in the 12mm disc below. This "protrusion" would need to be threaded to allow a copper lug to be fitted on the underside. I had bought a drawing app for my ipad called iDraw and thought I would throw a quick design together. I've used this app before for designing the toroid mounting pin, it's not the most technical drawing app but it matches my abilities.

I started with a small length of 16mm diameter round copper bar. I turned down 25mm of it to a diameter of 8mm and then the next 25mm to 14mm diameter.

I cut the turned section off with a hacksaw then cleaned up the cut end in the lathe finishing the 14mm diameter section to a length of 22.375mm.

Here's the finished blank. Next I drilled the hole for the 1/4" copper tube (6.75mm drill bit used) and the holes for the clamping M3 grub screws. Decided to deviate from the plan and have 2 clamping grub screws at 180 degree to each other.

Using an M8 X 1.25 die I cut threads on the end of the 8mm diameter section (about 15mm of threads) and then cut the M3 threads for the grub screws.

Here's the finished article fitted with an M8 stainless nut and copper electrical lug.

If you like this blog you can show your support by one or all of these.
1. +1 my blog and email it to a friend.
2. Follow me.... It's good to know someones interested.
3. Leave a comment.... All are appreciated.

230V to 110V for Synchronous Motor

Did a little more wiring today. It consisted of connecting one of the female IEC socket in the front panel to the step down transformer on the second level of the tesla coil. I used a length of standard 3A 2 core round black flex. The synchronous motor is rated at 7.5W at 110V so by the equation

                  Power (Watts) = Current (Amps) X Voltage (Volts)

            so   Current (Amps) = Power (Watts)
                                                ------------------
                                                Voltage (Volts)

                                             = 7.5
                                                ----
                                                110

                                             = 0.068 Amps

So, as you can see it draws very little current and even the lightweight 3A cable is a bit of an overkill.

The top IEC socket will supply 240V to the step down transformer which in turn will supply 110V to the synchronous motor. The transformer was sourced from Newlec and came in panel form (uncased) but mounted on a pressed steel bracket. I removed the bracket so I could mount the transformer on a perspex base. This meant that I had to remove the connection strip. This task was undertaken several months back and I thought it would be wise to check that the connections were correct and the transformer is doing what's expected, also a good reason to play with my Xmas present.

The transformer has inputs for 230V and 240V with an output of 110V. I decided to supply a 50Hz AC voltage across the 230V inputs and measure the output of the terminals marked as 110V. I used the function generator built into my oscilloscope to supply this signal to the transformer at a very safe amplitude of about 3 volts. The same signal was feed into channel 2 on the oscilloscope.

The oscilloscope probe was plugged into channel 1 and then connected across the 110V output terminals. With the same volts per division set on both channels and the scope set to dual I could view the input and output waves at the same time. I did take a pic to show you the result but my camera only got an image of a few dots on the oscilloscope screen due to the high speed of the trace. So I took a quick video with my Sony Bloggie.

As you can see, the terminal connection are correct and the transformer is stepping down the voltage by  a factor of about 2 so 230V would be output at about 110V.

I can now finish the wiring to the transformer with confidence. Very pleasing to see actual inputs and outputs on the oscilloscope screen also interesting to see how the 2 signals stay in phase.

Fitting the Primary Coil strike rail

With all the strike rail mounts finished and the strike rail formed into the approximate correct shape it was time to fit it all to the anti-strike disc. It was possible to loosely fit all the strike rail mounts and then thread the strike rail through each one.

I have used 15mm X M4 nylon bolts to secure the mounts may substitute these for stainless steel, still not decided.

Made a quick video just to give you a bit better view of the complete strike rail.

The rail is not a complete loop, it has a gap which is located inside one of the mounts. It needs this break to prevent induced eddy currents when the coil is running.

Here's a festive shot of the top of the tesla coil. Well it is Christmas!

Finishing the strike rail mounts

Finished for the xmas break today and thought I would get the strike rail mounts finished off. I had already drilled a 3.3mm vertical hole in each one which would be tapped to accept a 4mm bolt to fasten to the anti-strike disc. Before tapping I needed to drill a horizontal hole through each mount through which the strike rail will pass. The strike rail is 4mm in diameter and as the rail is a large loop the holes in the mounts need to be slightly over 4mm to allow for the curve. I opted for 4.4mm and set up the milling machine for drilling.

 I marked up the vice with masking tape so I could position each mount in the exact same place so I could lock the bed once the first mount is lined up for drilling.

All I needed to do is line up each mount with the masking tape as above and my holes will be in the same position on each one.

It was nice to be able to see the progress of the drill through the top of the mount. When approaching breaking through I greatly reduced pressure on the drill to prevent chipping the edges of the exit hole. This technique worked well, on inspection the exit holes were very neat. With the first mount drilled I checked that the 4.4mm hole would accept the curved 4mm strike rail. Bingo, 4.4mm is perfect.
  After another 15 mounts the drilling was done. Now to tap the mount holes.

It's about now that I'm regretting the choice of using 16 mounts for the strike rail, 8 would have been plenty, keep telling myself it will look really good.

9 done, only another 7 to do. Didn't take quite as long as I thought as there's only about 10mm threads to cut in each one.

Here are the 16 mounts done and de-masked. They look fantastic.

Marking out and drilling mounts for the primary

Today I spent a few hours progressing with the primary coil. I needed to mark out the positioning of the 8 primary support combs. These will be attached to the underside of the large anti-strike disc by 3 nylon bolts per each comb. I had already masked up the anti-strike disc to protect it and to allow marking out, I marked up 8 "spokes" onto the disc by taping the disc down to our glass dining table and taping a 360° protractor into the centre cut-out.

The position of the 8 spokes could then be easily marked out using a long metal rule placed over the relevant degree markings and lined up with the dead centre.

Also mounted to the underside of the anti-strike disc will be 16 evenly space strike rail holders around the perimeter so I marked on another 8 spokes at 45° intervals between the primary coil combs. With all the spokes marked up I could decide on the nylon bolt positions and mark them out for later drilling. 

Next job was to mark out the primary support combs. I don't want to "notch out" the combs as is the usual way to support the copper pipe of the primary. This method means you have to either glue or zip tie the tubing down, neither of which appealed to me. I plan to drill "full" holes into the combs and thread them onto the copper coil.

If you look at the coil plan above (or into my eyes, not around my eyes) you can see that the positions for the holes in each comb will be different. For my coil there is 8mm space between each rotation of the coil and the diameter of the copper tubing is 6.5mm so the holes in the combs need to be 14.5mm apart (centre to centre). This measurement is the same for each comb. As you move around to the next comb the spiral moves out by 14.5mm divided by 8 (1.8125) so there is an offset of 1.8125mm as you move around from comb to comb. You can avoid this complication by just moving the combs outwards by the same value when fixing them in place but I wanted all the combs to be the same distance from the secondary coil.

All the combs were masked up in the usual manner and then through carefully measuring my full scale primary plan I determined the position for the hole for the innermost coil on comb number 1. It was the easy to mark the position for the 1st hole in comb number 2 by adding 1.8125mm. And so on up to comb number 8. I didn't bother marking any subsequent holes onto the combs as I decided to fire up the digital vernier scales on my milling machine and locate them by just adding 14.5mm each time.

Not the best video but its really hard trying to film while drilling. Think I might invest ia a small tripod for my little Sony Bloggie.

I set up the milling machine so each perspex comb could be clamped in the the same place with an 10mm piece of pine stripwood underneath to prevent the drillbit chipping the back of the perspex as it breaks through. I positioned the bed so the drill bit was directly over the datum mark for the first hole. The hole was drilled and then the bed was moved 14.5mm to the left and the next hole drilled and so on.

I jotted down all the datum points as I drilled the first comb so I could use the info for the rest of the combs. And about 140 holes later......

Here's the drilled combs underneath the anti-strike disc. I have drilled all the necessary pilot holes in the strike disc, this will make it very easy to mark up the 3 bolt holes onto each comb by simply sticking a pencil through the pilot holes in the anti-strike disc.

Starting on the Primary Coil

Spent the last few weeks decorating the lounge, those of you who have followed my blog may have spotted the rather flowery decor in some of the pics, for this I apologize but can now confirm that the decor is a little more up to date. Now onto the primary. As per the original plans from Tesslastuff.com I acquired a fifty foot length of 1/4' copper tubing. This can be sourced from many places, I picked a well priced listing on ebay from a seller with a healthy feedback score.

I have spent a long time deciding on the final design of the primary coil, the plans from Tesslastuff.com give you the basic idea and all the correct measurements but, as usual, I wanted to enhance the design with my over engineering. After reading several blogs and coil build diaries there seemed to be two ways to help prevent primary coil strikes (streamers hitting the primary coil- not good). Firstly to fit a ground rail around the primary and secondly to completely cover the primary with a non-conductor. So I decided to do both.

Above is my solution to covering the primary coil. It's a 5mm thick perspex disc with an external diameter of 900mm and an internal diameter of 170mm. The cut-out allows the secondary coil to pass through and as this disc is prominent I want it to be aesthetically pleasing so paid extra for the inside and outside edges to be polished. Here's another pic to give you an idea of the disc size next to the tesla base (room still with old decor).

I would also need some way of supporting the copper tube spiral of the primary. Most people used strips of plastic with evenly spaced cut-outs with accepted the coils of the primary. The coils were secured with glue or by zip ties. I didn't like the idea of glue cos of it's general messiness which doesn't go well when combined with lovely clear perspex. The other option of zip ties just didn't look that good. The best looking install I saw was to use nothing to hold the coils in place. The coil passes through holes drilled in the supports, very neat but a lot harder to achieve. I ordered the material needed for the supports.

These were 8 pieces of 250mm X 30mm X 10mm clear perspex with the top and bottom edges polished.

Primary thoughts

As the secondary coil is pretty much finished I have already been thinking about construction of the primary coil. Both coils will sit on a perspex 50cm 12mm thick disc which is supported above the second layer of the main module on 8 30mm diameter 20cm long perspex pillars.

The second level contains the MMC module, the spark gap module and a step down transformer module for the spark gap motor. All modules are mounted on ceramic standoffs.

I have yet to finish off wiring and the support disc is not yet fitted. As you can see I have coated it in masking tape ready for marking up the holes for drilling.

The pillars are drilled out to 5mm and then threaded to take M6 bolts. If you want some advice for drilling and tapping perspex takes a look at my Drilling and Tapping Perspex page.

The primary and secondary coil will sit on the top perspex disc (covered in masking tape above). At this point I have decided to upgrade the design from the original plans (yet again). I thought it would be a good idea to mount the primary coil underneath a perspex disc which should help prevent strikes to the primary coil. Also I heard if you do this any strikes that hit the perspex create an interesting effect as they spread out over the perspex. I will install a more traditional strike rail around the edge of this protection disc. The disc diameter was decided at 90cm and it would need a 170mm diameter central cut-out to allow the 150mm diameter secondary coil to pass through.

And here it is. Ordered from Trent Plastics, it turned up at work this morning. Took me about 10 minutes to remove the 3 layers of cardboard packaging.

The thickness is 5mm and I paid extra to have the inside and outside edges polished. It's a big piece of plastic and does flex a little, but there will be 8 radial combs from the centre outwards to support the primary coil and these will add extra stiffness.