Need help to understand the reason for spark-length increase with lower res.-frequency (SSTC theory) / Tesla Coils / Forums

Forums

4hv.org :: Forums :: Tesla Coils

« Previous topic | Next topic »

Need help to understand the reason for spark-length increase with lower res.-frequency (SSTC theory)

Move Thread

LAN_403

nabzim

Wed Nov 21 2018, 03:30AM

Registered Member #60732 Joined: Sun Aug 28 2016, 06:07PM
Location: Villa Park, California
Posts: 8

I know that DRSSTCs need to operate at low frequencies in order to not blow up the IGBTs, but I think I remember reading/hearing somewhere that SSTC's (with MOSFETs) experience a similar effect, in that a lower resonant frequency can reduce switching losses greatly, thereby greatly increasing the coil's performance and increasing the spark length.

I noticed this effect happening with my SSTC as there is a huge increase in spark length when I add a way-bigger topload, so, I made a video to demonstrate it, as well as provide proof of the effect's presence:

(If the embedded video did not work here's a link: https://www.youtube.com/watch?v=AtaoyFd7eeU)
(I measured the resonant frequency with both toploads, numerical values are in the video description, on youtube)

I then decided I would share the video on the Facebook-group page, "TESLA COILS are BOSS", and a fury of confusion ensued as Spark-Gap coilers proposed that I was indeed wrong, saying that the spark-length increase was instead due to the voltage building up to greater levels before breakout occurred. While I know that's not fully the case, because I am running it with a breakout point (because I need to, because it's Solid State) and that's why sparks form even at the lowest levels of the variac.

So that's why I am here now asking for your help... I just need some confirmation, because I am not an Electrical Engineer (yet!), so I am not entirely sure about the whole thing... I could use some confirmation, please, and thank you.

And if anyone wants to tell me how to properly embed a video on the forum, I will fix it.

Sulaiman

Wed Nov 21 2018, 11:30AM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

Becoming an Electrical Engineer will not help much with Tesla Coiling,
ALL of my Electrical Engineer friends and colleagues have little knowledge of T.C.s and could certainly not design one from scratch.

The nature of sparks 'n' arcs varies with frequency,
e.g. for a simple sgtc there is a change usually around 300 kHz

Most importantly, there are several unknown parameters here,
such as wall-socket power consumption.
(at least try to measure the a.c. current consumption)
A larger topload stores more charge at a given voltage,
which gives brighter fatter sparks/arcs,
so the spark length may be shorter but look longer.

I do not have an answer, just more questions.

nabzim

Sat Nov 24 2018, 11:48PM

Registered Member #60732 Joined: Sun Aug 28 2016, 06:07PM
Location: Villa Park, California
Posts: 8

Thank you for your input Sulaiman.
I was just doing some tests now to measure power draw, and I recorded some data that I hope will be useful. However, while conducting these tests, it seems that I was pushing my coil a little too hard. I forgot to stop and feel the mosfets to check they're not overheating and give the circuit a rest, as I usually always do..... It appears that my film capacitors, which form part of the primary's path to ground, have melted! There was an abrupt end to the sparks while at a lower power level, and I could hear a "new sound" coming from below deck...
Luckily it happened near the end of my tests (although I was going to redo a test because some parameters changed between the two tests) So I at least have some data to plot!

I will post my findings as soon as I can.

PS: I do believe Electrical Engineering would aid me greatly in my Tesla-coiling efforts, because for me, the actual electronics involved in solid-state coiling is way way over my head! After about 2 years, I have come to understand many aspects of "pure Tesla coiling" as is demonstrated with my spark-gap coil: https://youtu.be/0jAYsmk9nxE (skip to 2:00 for action) (although, that video is old, I currently only use 2 MOT's and a static gap).
... I just want to be able to design my own things, without solely copying someone else's design...

nabzim

Mon Nov 26 2018, 02:48AM

Registered Member #60732 Joined: Sun Aug 28 2016, 06:07PM
Location: Villa Park, California
Posts: 8

accidently posted the same post twice, don't know how to delete the post, so.. edited it to say this instead!

nabzim

Mon Nov 26 2018, 02:53AM

Registered Member #60732 Joined: Sun Aug 28 2016, 06:07PM
Location: Villa Park, California
Posts: 8

Here is a link to a Google Drive folder, with two .jpg files and a .xlsx file of the data from my two tests: https://drive.google.com/open?id=16FHr6cf31wZuFLFjooDW-3ItaK-Q1JCr
(file attachment did not work, so I must use a link)

First of all, I am now using a new topload, which I built shortly after creating this thread (it's no longer toroidal in shape, but more of a rounded-cylinder, made from two identical aluminum cooking pots. My tesla coil somewhat resembles a Van de Graaf now... hehe..) now it has much more capacitance than previously.

These two tests were conducted first with just the new topload, then once with the big aluminum pan added on top of the new topload.

*After I began testing with the big pan added on top, I had to move the breakout point up and outward towards the edge of the pan, because there were a few strikes to the primary coil. That is why I wanted to redo the first test, to avoid changing any parameters other than topload capacitance, but one of my half-bridge capacitors failed (melted and the endcap popped out) at the end of the second test.
[Surprisingly, after everything had cooled down, the coil still seemed to work fine, and produced streamers that appeared to be the same size as before, however, I will not test it any further until I replace that capacitor, to avoid further damage to the bridge that may occur when the dielectric inevitably melts again.]

The secondary's resonant frequency during the first test, was 281 kHz. During the second, 222 kHz.

Hopefully all of this information can help with determining what makes the spark length increase.
My original thought was that it is some effect related to the semiconductors and their switching losses, because I thought I had read somewhere, that lowering the frequency will bring the switching times closer and closer to near-perfect ZCS. (Maybe because the effects of feedback phase delay becomes less?)
To me, the effect on spark-length increase seems too significantly large, to be solely due to the old "more charge builds up on the toroid" explanation... Although in reality, I have no idea. I could be totally wrong. But, I would appreciate more feedback from various Tesla-coilers on the matter.
Thank you.

Uspring

Fri Nov 30 2018, 10:53AM

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

For similar input power levels, e.g. at 300W, spark lengths don't seem to differ much. You can supply more power to the coil at the lower frequency, though, making your arcs longer. At lower frequency, the primary inductance allows for more current, so the input current and also the input power increases.

Additionally there is an effect of the secondary current on the primary current. With a big top load secondary current increases. That might also increase primary current draw.

So the longer arcs in the low frequency case might be related to how your coil draws power and not to an arcs requirement for power with respect to its frequency.

Sulaiman

Fri Nov 30 2018, 11:22AM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

From what I have seen and read, and a little experience,
John Freau derived an empirical formula for spark length using rsg,
spark lenght in inches = 1.7 x SQRT(power)
I built a couple of ssg, none reached the target above unless operated at low bps.
But it seems a good target.

I gave up on sstc after blowing too many fets,
but from what I see the Freau equation may not be applicable for drsstc etc.

teravolt

Tue Dec 04 2018, 03:31AM

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

I have also noticed this adding a top load. the tesla has distributive capacitance witch is in parallel with windings reduces the frequency from a quarter wave length frequency. adding a top load adds capacitance in series and voltage increases with a series resonance. current also increase through the secondary and through coupling the primary. The way I make my coils is to make them a higher frequency than desired knowing a top load will put them at the desired frequency. I set up a drsstc circuit and with a 4"coil and set the primary with +-50v and was able to add top torides till I got to about 30kv, inch long sparks. at some point the length of the spark stopped increasing with top load. all the while I made Shure the primary was also in resonance without frequency splitting. my point is that voltage gain in my case is about 5k to 6k. the only thing I can come up with is the Q of primary times the secondary Q in some way is responsible for the tesla being able to create tremendous voltage from wall voltage.

nabzim

Wed Dec 12 2018, 07:56AM

Registered Member #60732 Joined: Sun Aug 28 2016, 06:07PM
Location: Villa Park, California
Posts: 8

Uspring wrote ...

At lower frequency, the primary inductance allows for more current, so the input current and also the input power increases.

This seems like the most plausible explanation to me (so far).
Really; I had forgotten that without an SSTC having a resonant primary, there isn't current flowing in the primary when the transistors are off. Unlike a DRSSTC, where you'd better hope that there's no current flowing before you turn your transistors off. Otherwise, your silicon gets a huge smack in the face, and you lose some of that energy which could have been transferred to the secondary instead. (Now it makes sense to me why it has nothing to do with switching loss, or ZCS)

So it's simply inductive reactance... BAM, mystery solved! Occam's razor, slices again...

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