Secondary Coil Design Decisions

[sparkydownunder]

Hi Guys

New poster here, just starting out on the DRSSTC journey.
I've been in the research phase for about 2 months now and am finally ready to be locking down some final designs.
Im shooting for around a 70khz resonance and (from looking at Mads' DRSSTC1 and build guides) id be looking at around 2200 turns at 60cm or 2100 turns at 55cm.
Does anyone have any preference with a longer coil with slightly thicker wire or a shorter, denser coil?
If the coil is well below the inductance I need expanding the topload will help bring down the freqency. Is there any issues with an oversized topload?   

Cheers

SDU     

[Mads Barnkob]

Welcome to HVF

A medium sized coil, like the first I built too, is a good start, its not too big and too small and certainly not too expensive either

There is not much difference from 2200 to 2100 turns, but it is seen as a common mistake to make the overall coil length too short, also along with base and topload, so that spark length actually gets hindered from a too low sitting break out point.

I prefer the 50k impedance / longer secondary coils as described in my guide, with even dimensions between coil and topload: http://kaizerpowerelectronics.dk/tesla-coils/drsstc-design-guide/

A large topload can store a bigger charge, so it will represent a bigger load to the inverter feeding the secondary circuit. A big topload discharging a large spark will also kick the secondary circuit far more out of tune or result in whiplash like effects with a voltage gradient travelling very fast on the secondary winding, perhabs resulting in flashovers on the secondary coil.

A too wide topload can also result in loss of electromagnetic field control and you can get problems with sparks going downwards for the primary circuit.

That being said, you have to actually go to extremes to get this far, larger topload can give longer sparks as they store more charge, but they also demand more of the inverter.

Building a DRSSTC is one big impedance matching of 3 circuits, inverter, secondary and spark and none of them are really steady and constantly dynamic, its easy ain't it?

[sparkydownunder]

Hi Mads

Thanks so much for the response. Helpful as always.
The final thing I wanted to sort out before winding was the issue of spacing between the turns. Your DRSSTC1 guide shows a winding length of 605mm but with 2200 turns of 0.25mm wire. Is this extra 55mm between turns or are you including offsets in coil height?
In your guide you mentioned about a 10% increase in length to accommodate varnish etc (I assume this is the where the extra length came from in DRSSTC1). Have you found this rule of thumb to hold true for coils of all sizes?
Im going to be winding this on a industrial machine (I work in a coil shop on the side) so calculated spacing etc is possible. Think its worth the extra effort?
One of the insulation engineers suggested laying down a thin layer of varnish impregnated felt on the PVC before winding to eliminate air bubbles below the winding, I will report back with how successful this is
 
Thanks again

SDU

[futurist]

Extra winding length is due to wire varnish, but don't worry, winding isn't that critical and turns don't need to be perfectly aligned and without gaps

Winding wire on varnish would improve mechanical rigidity of the coil, however it makes winding more difficult
Commercial alkyd varnish I use dries quickly and without someone to help you it quickly turns out into a mess, at the end I gave up and wound the coil the standard way

[Uspring]

From the point of view of impedance matching, there is an optimal choice of secondary Q. The Q originates mostly from loading by the arc and is given by arc resistance / secondary impedance. The arc resistance changes considerably during arc growth and decreases for longer arcs. Secondary Q ranges typically from a few 100's down to 2-3 for strong arc loading.

So it would be better to have a large secondary impedance during the initial arc growth time and a smaller one later, when the arc is long. That's not really feasible, so one has to make a compromise. A value of 50kohms is a good choice for most coils.

There is a slight dependence of the secondary impedance on the power of the coil. The arc resistance is roughly inversely proportional to its length, i.e.

1 / Rarc ~ Arc length

That would suggest a lower secondary impedance for a more powerful coil to obtain the optimal Q. From theoretical consideration I believe the arc resistance to also depend on the frequency, with lower frequencies having a higher Rarc. Since more powerful coils usually run at lower frequencies, that implies a higher Rarc. So both effects are opposite to each other and might cancel. The frequency effect is weak, though, so it's probably is preferable to have a somewhat lower secondary impedance for larger coils.