26 or 18 awg for 315*1650mm secondary (and Balancing MMC and Primary on chosen Secondary in DR) / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

26 or 18 awg for 315*1650mm secondary (and Balancing MMC and Primary on chosen Secondary in DR)

1 2 3 next

Move Thread

LAN_403

Intra

Tue Jun 05 2018, 04:45PM

Registered Member #2694 Joined: Mon Feb 22 2010, 11:52PM
Location: Russia, Volgograd (Stalingrad).
Posts: 97

Hi guys.

I'm getting ready to winding a 315*1650mm secondary for CM600HA-24H bridge and 1600*400mm toroid.

Which wire will be best for this?

I have a 26awg wire 10lb, but I found that reactance at resonance is about 108kOhm with 3874 turns of wire.

Is 26awg good for this or 18awg will be better?

With 1600 turns of 18awg reactance at resonance is about 44kOhm.

Is reactance at resonance important?
What difference does it make on coil between 26awg and 18awg?

On Mads Barnkob's page (

) I found that something about 50kOhm should be good and even 60kOhm already isn't good.

Also, I got a confusing results with ScanTesla on that. With 26awg ScanTesla shows significant lower peak&rms currents on same bps, with same streamer length comparing with 18awg.
Also shows almost the same coil power and voltages on both circuits.

Is frequency 47-50kHz is suitable for CM600HA-24H of 19kHz will be better?

Many thanks on any advice.

Mads Barnkob

Wed Jun 06 2018, 06:14AM

Registered Member #1403 Joined: Tue Mar 18 2008, 06:05PM
Location: Denmark, Odense C
Posts: 1968

A 108 kOhm coil would require much longer on-times for the same spark length output compared to a 44 kOhm coil. It is not the primary to secondary turn ratio that is important for building up potential in a DRSSTC.

Regarding the CM600 Fres(max), you are lucky that I used a CM600 in my example:

it will do fine at 50 kHz and I would at all cost avoid having a multiply kW coil working in the audible range as 19 kHz would be on the edge of.

Intra

Wed Jun 06 2018, 10:46AM

Registered Member #2694 Joined: Mon Feb 22 2010, 11:52PM
Location: Russia, Volgograd (Stalingrad).
Posts: 97

Mads Barnkob wrote ...

A 108 kOhm coil would require much longer on-times for the same spark length output compared to a 44 kOhm coil. It is not the primary to secondary turn ratio that is important for building up potential in a DRSSTC.

Why on-time length is important?

How exactly potential in DRSSTC builds up with primary to secondary turn ratio?

Many thanks for advice.

Mads Barnkob

Thu Jun 07 2018, 07:40AM

Registered Member #1403 Joined: Tue Mar 18 2008, 06:05PM
Location: Denmark, Odense C
Posts: 1968

Its a matter of average power transfer. A low impedance coil can absorb higher peak energies, while the high impedance can not, so that will have to be driven at lower primary peak current, but for longer time, in order to achieve the same spark length.

Resonant voltage rise is more dominant than transformer ratio.

teravolt

Thu Jun 07 2018, 06:52PM

Registered Member #195 Joined: Fri Feb 17 2006, 08:27PM
Location: Berkeley, ca.
Posts: 1111

26 awg will give you a lower frequency witch will be easier on your IGBT and more inductance for more voltage. if you use 18 you can use a bigger top toride to reduce f res enough. With 18 and a bigger toride your ground strikes will be more crashing however your power demands may be larger. what does your tesla calculator say about f resonance

Intra

Thu Jun 07 2018, 09:44PM

Registered Member #2694 Joined: Mon Feb 22 2010, 11:52PM
Location: Russia, Volgograd (Stalingrad).
Posts: 97

teravolt wrote ...

With 18 and a bigger toride your ground strikes will be more crashing

what means "crashing" here? Is that an adding a capacity from the ground to secondary when ground strikes and droping fres? if so then that will happens even with EMI filter terminal?

teravolt wrote ...

what does your tesla calculator say about f resonance

did not quite understand the question. 19khz with 26awg, 47khz with 18awg.

Intra

Fri Jun 08 2018, 12:54AM

Registered Member #2694 Joined: Mon Feb 22 2010, 11:52PM
Location: Russia, Volgograd (Stalingrad).
Posts: 97

Mads Barnkob wrote ...

Its a matter of average power transfer. A low impedance coil can absorb higher peak energies, while the high impedance can not, so that will have to be driven at lower primary peak current, but for longer time, in order to achieve the same spark length.

Resonant voltage rise is more dominant than transformer ratio.

I understand low impedance coil can absorb higher peak current. I posted a scantesla screenshots with almost exact voltage and besides extra 100kkV for the almost same streamer length has the secondary with 26awg. (okay, it's 19kHz. close to hearing range, my bad. the matter I want to understand is beyound the scope of hearing range and so.)

Next.. Okay, high impedance coil can't absorb HIGHER peak energies, but the reason I'm here, I'm not sure the coil need it.

If I see it correctly, if we have a low impedance coil, then coil can get up high power with a short time. If so, then, question is, why the short time needed? When if for shorting heat loss, then okay.

But if it is the only reason then why on-time length is important and does it important at all or not?

Can I increase on-time until RMS current reaches Imax (continious) of igbt datasheet or there is an additional restrictions?

May I have what I see on ScanTesla screenshot - ~750kkV Vsec with ~127Arms and ~847Apeak on 26AWG with 440bps instead of ~650kkV Vsec with ~190Arms and ~2127Apeak on 18AWG with 440bps, even with 170uS on 26AWG instead of 63uSec on 18AWG?

Because if I can this acheive according ScanTesla screenshot, then why Q-factor is important? Up to what point it is essential?

ScanTesla here is just say me: Hey, put a pulse width on top, high impedance will not allow the current to grow.

So, as my question is here, then, I'm not sure this is work and I still can't understand, what is better, 315*1650mm with 26AWG or 315*1650mm with 18AWG?
(let the 19kHz secondary give a chance to exist just in theory. If that direction turns up as a better, then I just recalculate it for 25-30kHz for avoid being close up to hearing range.)

I will remind, 26AWG is 13kW line for 259 inch streamer and 18AWG is just 11.5kW line for 259 inch streamer. Difference is only 1.5kW in favor of 18AWG, but it isn't important for my setup so let's tries to neglect this 1.5kW losses just for me.

Mads Barnkob wrote ...

so that will have to be driven at lower primary peak current, but for longer time, in order to achieve the same spark length.

That is aslo on my screenshots. 170uS for a 26AWG and 63uS for 18AWG for almost the same streamer length of 259.15 inch average.

Mads Barnkob wrote ...

Resonant voltage rise is more dominant than transformer ratio.

If so then 750kkV Vsec with 26AWG turns out better than 650kkV Vsec with 18AWG for 440bps 259inch streamer. Is I understand you right?

Many thanks on advice.

Uspring

Fri Jun 08 2018, 02:48PM

Registered Member #3988 Joined: Thu Jul 07 2011, 03:25PM
Location:
Posts: 711

I believe that trying to achieve a fast power rampup might not be the preferable way to achieve big sparks in a DRSSTC. Sparks live on power, so it is desirable to input as much energy as the bridge allows to the primary tank. As a consequence it is better to have the arc loading limit the primary current to an acceptable level instead of an OCD stopping the burst. After an OCD cutoff much of the energy in the primary is returned to the bus caps via the IGBTs diodes, while it is better to have that energy in the secondary and the arc. Also an OCD event will stop further energy input to the primary.

Arc loading shouldn't limit the primary current too much, since that means lower power input, so there is an optimal matching between the bridge and the coil. The effect of the arc on the primary current depends on tuning, coupling, secondary Q (Q being mostly determined by the arc load, wire resistance contributes much less) and the primary tank parameters. There is an optimal secondary Q for strongest arc loading effects, which is around Q=1/k. Since the arc load varies much depending on size and frequency the value of the best secondary impedance is uncertain, but values of around 50kOhms seem to work best. You do want strong arc loading effects, since this implies an efficient power transfer from primary to secondary.

The max current in the primary due to arc loading is dependent on the choice of the primary tank impedance. Typical primary Q due to loading maybe around 5 or 10, so you have to choose the primary L and C accordingly to match the bridge capabilities.

Intra

Fri Jun 08 2018, 08:24PM

Registered Member #2694 Joined: Mon Feb 22 2010, 11:52PM
Location: Russia, Volgograd (Stalingrad).
Posts: 97

Uspring wrote ...

I believe that trying to achieve a fast power rampup might not be the preferable way to achieve big sparks in a DRSSTC. Sparks live on power, so it is desirable to input as much energy as the bridge allows to the primary tank. As a consequence it is better to have the arc loading limit the primary current to an acceptable level instead of an OCD stopping the burst. After an OCD cutoff much of the energy in the primary is returned to the bus caps via the IGBTs diodes, while it is better to have that energy in the secondary and the arc. Also an OCD event will stop further energy input to the primary.

Arc loading shouldn't limit the primary current too much, since that means lower power input, so there is an optimal matching between the bridge and the coil. The effect of the arc on the primary current depends on tuning, coupling, secondary Q (Q being mostly determined by the arc load, wire resistance contributes much less) and the primary tank parameters. There is an optimal secondary Q for strongest arc loading effects, which is around Q=1/k. Since the arc load varies much depending on size and frequency the value of the best secondary impedance is uncertain, but values of around 50kOhms seem to work best. You do want strong arc loading effects, since this implies an efficient power transfer from primary to secondary.

The max current in the primary due to arc loading is dependent on the choice of the primary tank impedance. Typical primary Q due to loading maybe around 5 or 10, so you have to choose the primary L and C accordingly to match the bridge capabilities.

Thank you for your answer. I found it very helpful.
Some things I still do not understand.
JavaTC show secondary Q in hundreds. 100, 200, 300. If I divide 1 by k I got units. 6, 5, etc. Do you mean "k" is coupling?

Many thanks on advice.

Uspring

Sat Jun 09 2018, 08:38AM

Registered Member #3988 Joined: Thu Jul 07 2011, 03:25PM
Location:
Posts: 711

JavaTC's secondary Q is calculated from the wire resistance plus skin effect. It does not take into account arc loading. That will reduce Q much more than the wire resistance alone once the arc has broken out.
Yes, k is the coupling.

1 2 3 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob