l'm chatting with friends about QCWDRSSTC

[曹靖]

This is the viewpoint put forward by a friend of mine when chatting with me regarding QCWDRSSTC. It is possible that the translation software may not be very accurate in terms of translation, and I am not sure if it will affect reading

[davekni]

Glanced at a bit of the discussion, which I gather is about theoretical limits of QCW arc length.  One we've already discussed is straightness.  The few posts I've seen of QCW coils running outside away from structure show arcs curve down much more than indoors.  Hitting ground is certainly a limit to arc length.
Another is electrostatic repulsion of the arc to itself amplifying wiggles.  My initial low-frequency QCW thread has some discussion and images around this:
https://highvoltageforum.net/index.php?topic=1268.msg10111#msg10111
Resistance along arc path is likely to be limiting at some length too.

[曹靖]

Resistance along arc path is likely to be limiting at some length too.

My friend proposed a new approach to solve the mismatch caused by the arc being too long

[davekni]

Decades ago before FETs were very good, saturating-core inductors were sometimes used as switches on the secondaries of switching power supplies.  They were called "magamps" or "magnetic amplifiers", though they were really switches, not amplifiers.  Cores had very high permeability until saturation and a sharp transition to low permeability at saturation flux density.
For TC use, core would need to be in upper part of secondary away from primary in order to avoid raising frequency of both.  That makes insulation difficult if any control coil is to be included.

Most QCW coils are limited by available bridge power and/or stored energy in bulk caps, not by arc capacitance.  Easiest way to handle arc capacitance is with high coupling.  My latest QCW has a coupling of 0.71.  Previous low-frequency QCW experiment had coupling of 0.9.  Even with more common 0.4-0.5 coupling of QCW coil, energy transfer doesn't decrease enough to matter due to arc capacitance.

For conventional DRSSTCs, arc capacitance is a key limiting factor, especially for larger coils.  High coupling tends to cause racing arcs (excess local voltage gradient along secondary near primary).  I spent way too much time and money trying to use TRIAC arrays to switch in more MMC capacitance to track arc capacitance.
https://highvoltageforum.net/index.php?topic=2230.msg16382#msg16382
I still keep thinking about other options for dynamic tuning of my DRSSTC (not QCW).  Saturating ferrite cores keeps coming to mind.  However, I haven't come up yet with any workable ideas.  Triggered spark gaps are another possibility, but with lots of issues too.

[曹靖]

Most QCW coils are limited by available bridge power and/or stored energy in bulk caps, not by arc capacitance.  Easiest way to handle arc capacitance is with high coupling.  My latest QCW has a coupling of 0.71.  Previous low-frequency QCW experiment had coupling of 0.9.  Even with more common 0.4-0.5 coupling of QCW coil, energy transfer doesn't decrease enough to matter due to arc capacitance.

But my friend believes that sword arcs with a length of more than five meters have already reached a very serious degree of detuning of secondary frequencies

[davekni]

But my friend believes that sword arcs with a length of more than five meters have already reached a very serious degree of detuning of secondary frequencies

Wow, do you have a link showing a QCW arc of 5 meters?  Longest QCW arc claim I've seen is 2.5 meters.

[曹靖]

Wow, do you have a link showing a QCW arc of 5 meters?  Longest QCW arc claim I've seen is 2.5 meters.

I'm sorry, he just did simulation and didn't actually build it. I saw a 2.6 meter QCWDRSSTC on a forum in China, and the following is the image provided by the translation software. The original author only provided arc images and a brief introduction without adding specific parameters and circuits, which is very regrettable

[Uspring]

But my friend believes that sword arcs with a length of more than five meters have already reached a very serious degree of detuning of secondary frequencies

Moving away from resonance is a problem particular to lowly coupled coils. With a high coupling, the coils act more like a standard transformers than as tanks and the voltage depends on the turns ratio of secondary to primary and the coupling. Resonant amplification of voltage plays a minor role. Also keep in mind, that long arcs draw very much power from the secondary tank and so the Q of it is quite low. Not operating at resonance still involves a penalty, but the high coupling compensates for that.

But QCWs can work successfully with a somewhat lower coupling, when utilizing upper pole operation. You can tune the primary somewhat below the secondary and start the coil on the upper pole. This requires extra electronics to force start it there, since primary zero current switching will usually start at the lower pole if the coil is tuned like this.
What will happen in this mode of operation is, that the operating frequency, i.e. the upper pole, will be much closer to the secondary resonance frequency and follow it downwards as the arc capacitance increases. The important quantity determing the power transfer between primary and secondary is the difference between operating frequency and secondary resonance frequency. This difference is kept low by this method of running the coil.

[davekni]

I'm sorry, he just did simulation and didn't actually build it. I saw a 2.6 meter QCWDRSSTC on a forum in China, and the following is the image provided by the translation software.

2.6m is great.  Nice images.  More reasonable for an actual coil than 5m.
BTW, my ceiling is only 1.52m above my QCW breakout point.  A few of my slightly-curved arcs did not quite hit my ceiling and extended horizontally for ~2m.  They were probably close to 2.6m total, but I don't have any measurements.

[曹靖]

Moving away from resonance is a problem particular to lowly coupled coils. With a high coupling, the coils act more like a standard transformers than as tanks and the voltage depends on the turns ratio of secondary to primary and the coupling. Resonant amplification of voltage plays a minor role. Also keep in mind, that long arcs draw very much power from the secondary tank and so the Q of it is quite low. Not operating at resonance still involves a penalty, but the high coupling compensates for that.

I also lean towards your point of view. My 2.4-meter QCWDRSSTC has a much higher secondary frequency than the primary, and it performs poorly at inputs below 150V. However, when the voltage exceeds 200V, the arc growth is very significant during the process of reaching 420V. I also believe that the energy storage of DC bus capacitors is more important than secondary detuning for longer arcs, as it largely determines the final length of the arc. As you described, QCWDRSSTC is not Loose coupling like the traditional DRSSTC, but more like a common transformer. The secondary voltage largely depends on the turn ratio rather than resonance.

[曹靖]

2.6m is great.  Nice images.  More reasonable for an actual coil than 5m.
BTW, my ceiling is only 1.52m above my QCW breakout point.  A few of my slightly-curved arcs did not quite hit my ceiling and extended horizontally for ~2m.  They were probably close to 2.6m total, but I don't have any measurements.

I am still looking forward to my friend making a 5-meter sword shaped arc. His simulation and description are very reasonable and I look forward to his practice

[black.yang]

I'm sorry, he just did simulation and didn't actually build it. I saw a 2.6 meter QCWDRSSTC on a forum in China, and the following is the image provided by the translation software.

2.6m is great.  Nice images.  More reasonable for an actual coil than 5m.
BTW, my ceiling is only 1.52m above my QCW breakout point.  A few of my slightly-curved arcs did not quite hit my ceiling and extended horizontally for ~2m.  They were probably close to 2.6m total, but I don't have any measurements.

Hi dave, I'm glad my friend put our discussion here.
After reading the above discussion, there are some differences between my ideas and the screenshots displayed. The idea behind all of this is that Caojin and I are fantasizing about a QCW with an arc length of 5M and discussing the technical solutions needed to complete it.
High coupling can turn QCW into a traditional transformer, and the ratio of primary and secondary coils can become an important factor affecting the voltage ratio. But when the arc needs to become longer, higher voltage gain Q is also needed to achieve the energy requirements brought about by the continuous growth of the arc. At this point, make the frequency of the primary and secondary coils consistent. It seems beneficial.

I am considering using some method (like adjustable inductor)to achieve tuning, if necessary.
This method also has the advantage of locking the QCW at the "upper pole" or "lower pole", and achieving soft switching at the same time. Compared to the "leader pulse" achieved through the intervention of the driving signal, it has certain advantages.

[Uspring]

Steve Ward has built a 160 kW QCWDRSSTC: https://www.flickr.com/photos/kickermagnet/15705193211/
It creates 11 feet arcs, runs at around 350 kHz with 20 ms burst length. Steve mentioned, that arc length and branching didn't live up to his expectations. I think, that probably the branches contribute to the lower than expected total length. He experimented a bit with a higher frequency and found, that this seemed to reduce branching.
Another experience he made, is, that under the severe arc loading the upper pole ceases to exist. His explanation was along the lines of this post: https://highvoltageforum.net/index.php?topic=113.msg15012#msg15012 . He observed a decline in primary current followed by a jump down in frequency to the remaining lower pole. He was able to avoid this by increasing the coupling.