The small coil reworked

This page describes the MODEL1B I have built. Although, almost everything has been built new, it is callel "model 1b" because its power level and physical dimensions are the same. It's just a reworked version of the first coil model.
This page is not fully completed yet and more information will follow.
The technical specs
Circuit diagram
Construction details, plans and photos
Primary Capacitor
Primary Coil
Secondary Coil
Main sparkgap
Safety sparkgaps
Power control unit
Filter unit
Operating coil

The technical specs

The following information is about the technical specifications of my coil
Power consumption 400VA
Transformer 8kV/50mA (Neon)
Primary windings 11
Primary capacity 17.8 nF
Secondary diameter 75mm
Secondary windings 1520, 0,25mm copper wire
Secondary winding height 380mm (5:1 ratio)
Resonance frequency about 220kHz
Toroid diameter
Toroid inner diameter
RF Ground 100cm copper tubing, 22mm diameter

The coil uses the same 8kV/50mA neon sign transformer (NST) as model 1 uses. The primary coil is also a conical type coil, built with 8mm copper tubing.

Circuit diagram

The coil circuit is fed by the power control unit. Tr1 is the neon sign transformer, which delivers 2x4kV center tapped. The elements C1a, C1b, L1a, L1b, R1a and R1b build the filter unit. S1a and S1b are protection sparkgaps for the transformer. The main tank circuit consists of Cm, L2 and spargap S2. Sparkgap S3 is a protection sparkgap for the main capacitor. L3 is the secondary coil connected to RF ground and with the toroid on top.

Construction details, plans and photos

This chapter shows information and pictures from the construction of my tesla coil. It is sorted by the different components.
Primary Capacitor
Primary Coil
Secondary Coil
Main sparkgap
Safety sparkgaps
Power control unit
Filter unit

Primary Capacitor

The primary capacitor has been replaced. I use 2 of my caps for the new 2kVA coil in series now, the 34.6nF and 35.9nF units, which makes 17.6nF. For more information about these caps click here.
I also did a test with my Cornell Dubilier 940C caps in an MMC array, which failed. I was aware of the problems with these caps, but just thought to give them a try in this low powered coil. See discussion on PUPMAN.
The following pictures show the results of the failure. The array was built with 3 strings of 10 47nF/3kVDC. Each cap has a bleeder resistor of 20MOhm, build from 2 series resistors.
You can see on the photos, that the ends of the caps are pressed to the outside. The current even in this low power setup seems to be too high for the connection of the wires to the capacitor foil. I'm not sure, why 2 bleeder resistors got on fire. This must be a different failure.
The rest of my Cornell Dubilier caps will be used for filter units. I have done this in model 2, and they seem to be good for that purpose.

The new MMC

I have built a new MMC for the model 1b coil. It is made of Philips type KP/MMKP376 caps. Each cap has a value of 22nF and a voltage rating of 1600VDC. The caps are arranged in strings of 10 caps each. The strings are built identically to the MMC construction of my BONSAI coil.

The strings are mounted on 2 aluminium rods with quadratic cross section. Single strings may be added or removed to change overall capacity.
The strings facing each other are mounted in a way, that all wires facing each other have the same potential. The same applies to strings next to each other. This prevents arcing over. See a more detailed description in the construction page.

Primary Coil

The primary coil assembly is again made of chipboard, but now I used chipboard with white surface so I did not have to paint myself. There are 10 turns of 8mm copper tubing mounted in a conical form with an angle of 20 degrees.
An important change has been made for the supports of the copper tubing. It is now made of LDPE to avoid the problems with arcovers I had with the first coil. It is the same technique used for model2.
Detailed photos of the construction are not yet available.

Secondary coil

The secondary coil is again wound on a PVC tube (walls about 1mm thin) with 75mm diameter. I used 0,25mm copper wire now and wound about 1500 windings. Photos will be available at a later time.

Main sparkgap

I have rebuilt the static type sparkgap (Richard Quick design) for better performance. It is now made from a larger diameter PVC tube and also larger diameter copper tubing. The fan also is now a much larger one. You can see the sparkgap on the following photo.

Circuit assembly

This picture shows the new circuit board made from polypropylene. On the right you see the main sparkgap, on the left the filter unit. In the back, one of the new caps can be seen.
These 2 pictures show a more detailed view of the sparkgap.

Safety sparkgaps

I have built new safety sparkgaps. In model 1, the gaps were just made with screws with hat nuts applied for a round surface. Adjustment was a little difficult, as you had to turn and fix the screws. The new safety gaps can be seen here.

Safety gaps

This picture shows the safety sparkgaps. They are made from brass with steel balls for use with halogen light systems. The base is made from epoxy. Adjustment is very easy.

Power control unit

The power control unit has not been changed. Click here for a description.

Filter unit

The filter unit is needed to prevent high voltage of high frequency to strike back to the power supply. As you can see in the circuit diagram above, the filter consists of 2 capacitors, 2 resistors, 2 chokes and 2 sparkgaps.

Filter unit

The filter unit has not been changed. But is has been removed from the extra PVC plate and is now mounted directly on the PP circuit board.
After first tests, it seems that the isolation of the resistor is not good enough. There is no arc over, but when the sparkgap does not fire, there is a slight blue glowing, which comes from the filter unit.
I will investigate this.

Filter unit capacitor

The capacitor of the filter unit has not been changed. Click here for a description.

Operating coil

I have started to do tests with 2 toroids, the toroid from model 1 and the new larger toroid for model 1b. Secondary resonance is 250kHz with the smaller toroid and 220kHz with the new one. Both toroids stacked result in 209kHz resonance.
When using the MMC with 6, 7 or 8 strings and different primary tap points, the following primary resonance frequencies can be tuned (only full turn tap points measured):

Strings Capacitance Tap at 11 Tap at 10 Tap at 9 Tap at 8
6 13.3nF 236kHz 263kHz 292kHz 329kHz
7 15.5nF 218kHz 243kHz 271kHz 306kHz
8 17.8nF 205kHz 227kHz 254kHz 286kHz

With the given values the tap points can be defined for use with different cap/toroid combinations. The following table shows the tap points. (The windings are rough numbers, I have marked all points on my primary exactly)

Toroid C=13.3nF C=15.5nF C=17.8nF
small toroid E (10.5 Wdgs) D (9.65 Wdgs) B (9.15 Wdgs)
big toroid n/a C (11Wdgs) A (10.4 Wdgs)
both toroids n/a n/a F

Running the tests

Up to now, only the tap point E using the 13.3nF cap has been tested. I have made a video of this setup. The performance is much better than with model 1. The sparkgap used in this setup is still the small gap from model 1, as I had some problems with the new RQ gap described above.

Download the 24 seconds video (1.2MB MPG, no sound) here:

It is interesting to see, that the sparks are now breaking out only at one place at a time, and that they are moving slower on the toroid than with my first coil or with bonsai.
I will try to do the other tests as soon as possible.
The tests using the 17.8nF version will also be compared to the LDPE oil filled cap version to check for differences in the MMC performance.
The problem I had with ignition of the coil when adding more gaps during tunigng seems to be caused by the sparkgap. I tested with the old gap from model 1 (which also is used with BONSAI). With this gap, everything works fine. I will stay with the new sparkgap with only 4 or maybe 5 gaps and compare the results to the small gap which uses 6 gaps.

Interesting effect

When finishing my tests, I found that the secondary gives shocks, even if there is no more connection to any other parts of the coil and you hold the secondary unmounted in your hands. This lasts for several minutes. If you touch the surface of the windings (sealed with epoxy), you get these electric shocks. With my PU sealed coil I did not find these effects. This has been found by several coilers, and there was also a discussion on PUPMAN. A possible reason is an electret effect.
I found an interesting thing, that a missing RF ground connection did not have effect on coil performance. Of course, I always use the RF ground, as this will certainly have an effect on RFI. Also, this is only a small coil, and with larger coils it might look different.

© 2001 by Herbert Mehlhose