My 160 (?)mm sstc

[oneKone]

Woah, that's pretty nice. I'm looking forward for more updates! Going to work on my coil after my major exams Busy studying lately .

For smaller coils, any thoughts on the type of circuitry you are going to use? Was thinking to make some miniature tabletop sized tesla coils. PLL circuit seems interesting especially the one from E-Schmok

Cheers, I think for the circuitry I might use the classic ucc3732x combo (what this coil is running from) E-Schmok's coil is definitely on my list! Come to think of it that may be the next coil I build..

Good luck with you're exams.

[oneKone]

Just a quick vid of a short run. I forgot the voltage but I think it's around 430v input, still using the current limiting resistor....... That I still haven't calculated yet. Man I'm lazy!

/>

[Mads Barnkob]

Smaller coils are not necessarily safer, you risk taking chances because its looking small and innocent. It is always safer when you have to assume everything is dangerous and measures has to be taken to ensure its dead, unplug, bleeding resistors, ground crowbar circuits, earthing wires etc

I first started to worry when I had 50 cm sparks from a table top coil, then it was near to component drawers etc.

You are improving your coil with every update, that is great, what kind of interrupter are you using in that short video? It sounds like its a rather high frequency.

[oneKone]

Smaller coils are not necessarily safer, you risk taking chances because its looking small and innocent. It is always safer when you have to assume everything is dangerous and measures has to be taken to ensure its dead, unplug, bleeding resistors, ground crowbar circuits, earthing wires etc

I first started to worry when I had 50 cm sparks from a table top coil, then it was near to component drawers etc.

You are improving your coil with every update, that is great, what kind of interrupter are you using in that short video? It sounds like its a rather high frequency.

Cheers for the compliments. As for a smaller coil I was thinking something around what E-Smock made or something like the tiny Tess from oneTesla.

I'm actually happy with the progress I've been making, considering its still half bridge.... For some reason the same igbt always dies in fullbridge, I don't know if it's because I'm driving that gate with only 10v.

I'm using steve wards drsstc interrupter in burst mode, just basically adjusted the frequency until the biggest sparks were given

[oneKone]

I am at a complete loss...... I fail to understand how incorrect phasing between the primary and secondary can kill igbts? Power input was at 15v with current limit set at 0.5A. I've killed (all Chinese) 4 60n60 and yep....... 2 ixgn60n60. It makes no sense at all to me.....

Edit:
I've been driving them at about 18v at the gate. Just had one fail is I increased the gate voltage. I thought driving them at this gate voltage would be safe?

second edit:
i have now found the reason in why i killed the igbts. i had the variable mag set wrong on my oscilloscope...... so what seemed like 18v was actually closer to 40v, so no wonder why they died!   

[Mads Barnkob]

I am at a complete loss...... I fail to understand how incorrect phasing between the primary and secondary can kill igbts? Power input was at 15v with current limit set at 0.5A. I've killed (all Chinese) 4 60n60 and yep....... 2 ixgn60n60. It makes no sense at all to me.....

Edit:
I've been driving them at about 18v at the gate. Just had one fail is I increased the gate voltage. I thought driving them at this gate voltage would be safe?

second edit:
i have now found the reason in why i killed the igbts. i had the variable mag set wrong on my oscilloscope...... so what seemed like 18v was actually closer to 40v, so no wonder why they died!   

That is a funny thing about IGBTs, they do not suffer gate breakdown until gate voltage is above 80VDC, but the higher gate voltage makes it able to conduct so much more current from the internal workings of the IGBT layers, like explained in details here:

An IGBT in the forward-blocking state can be transferred to the forward conducting state by removing the gate-emitter shorting and applying a positive voltage of sufficient level to invert the Si below gate in the P base region.

This forms a conducting channel which connects  the  N+  emitter  to  the  N--drift  region.
 
Through  this channel, electrons are  transported  from  the  N+  emitter  to  the  N--drift.  This  flow  of  electrons  into  the  N--drift lowers  the  potential  of  the  N--drift  region  whereby  the  P+  collector/  N--drift  becomes  forward-biased.  Under  this  forward-biased  condition,  a  high density  of  minority  carrier  holes  is  injected  into  the  N--drift  from  the  P+  collector. 

When  the  injected  carrier  concentration is very much larger the background concentration, a condition defined as a plasma  of  holes  builds  up  in  the  N--drift  region.  This  plasma  of  holes  attracts  electrons  from   the   emitter   contact   to   maintain   local   charge   neutrality. 

In   this   manner,   approximately equal excess concentrations of holes and electrons are gathered in the N--drift  region.  This  excess  electron  and  hole  concentrations  drastically  enhance  the  conductivity  of  N--drift  region.  This  mechanism  in  rise  in  conductivity  is  referred  to  as  the conductivity modulation of the N--drift region.

So you were properly forcing it way outside its SOA (Safe Operating Area)

[oneKone]

so.... after a break i decided to see what spark length i could get from a half bridge of 60n60s. with 520v dc input i got (just) 400mm sparks to an ungrounded ikea desk lamp....... thats from a 300mm long secondary, and of course i didn't have the camera rolling! what was weird to me is that the igbts failed when i turned down the duty cycle on the interrupter to walk over and turn down the variac..... if someone could elaborate more on that it would be appreciated. the next step for sure is to build a full bridge now that i have the gate voltage problem outa the way and be a bit more sensible on the input voltage!

also this weekend I've managed to score two variacs! but the most impressive part to me was the price....  $80 aud for a 5 amp and 3 amp core. in perth they rarely come up, let alone at a price like this.   

[oneKone]

With a bit of messing around with the primary coil I've managed to get 350mm spark to an ungrounded IKEA lamp @ 320v dc input on a half bridge configuration. I'm waiting on another batch of Chinese igbts to try a fullbridge setup.

[futurist]

IGBT/MOSFETs are sensitive to overvoltage, I wouldn't go above 400VDC for 600V rated parts just to have a bit of overhead

what was weird to me is that the igbts failed when i turned down the duty cycle on the interrupter to walk over and turn down the variac..... if someone could elaborate more on that it would be

Did you lower the BPS too? I noticed on my DRSSTC when I lower BPS below the certain value the primary current raises. It may have something with the ionization of air, at lower BPS air could have enough time to deionize.

[oneKone]

IGBT/MOSFETs are sensitive to overvoltage, I wouldn't go above 400VDC for 600V rated parts just to have a bit of overhead

what was weird to me is that the igbts failed when i turned down the duty cycle on the interrupter to walk over and turn down the variac..... if someone could elaborate more on that it would be

Did you lower the BPS too? I noticed on my DRSSTC when I lower BPS below the certain value the primary current raises. It may have something with the ionization of air, at lower BPS air could have enough time to deionize.

this could be my lack of knowledge showing.... but in a half bridge configuration at 500v input wouldn't each igbt be switching 250v? or have i confused this.

the bps were the same (sorry i don't have a number), but to turn down the variac my arm had to be near the antenna for the feedback. i'm not sure if this accounts for anything or not. 

[Mads Barnkob]

350-400 mm sparks from a 300 mm secondary coil is great results for a SSTC, it is hard to get them much longer than this without really stressing the switches and failure rate rises.

That is a fine deal getting two nice and small variacs at that price. a 5A will do fine in almost every aspect, its just a matter of low enough BPS and you could feed a large coil from it

IGBT/MOSFETs are sensitive to overvoltage, I wouldn't go above 400VDC for 600V rated parts just to have a bit of overhead

what was weird to me is that the igbts failed when i turned down the duty cycle on the interrupter to walk over and turn down the variac..... if someone could elaborate more on that it would be

Did you lower the BPS too? I noticed on my DRSSTC when I lower BPS below the certain value the primary current raises. It may have something with the ionization of air, at lower BPS air could have enough time to deionize.

this could be my lack of knowledge showing.... but in a half bridge configuration at 500v input wouldn't each igbt be switching 250v? or have i confused this.

the bps were the same (sorry i don't have a number), but to turn down the variac my arm had to be near the antenna for the feedback. i'm not sure if this accounts for anything or not.

In in half-bridge (load connected to negative rail) each switch sees the full DC bus voltage + switching spikes/transients, since the current is only passing through one switch to the load. If you are using a capacitive voltage divider for the load, instead of negative rail, the voltage is of course halved.

In a full-bridge you have a upper and lower leg switch with the load in the middle, here the voltage is across two switches and serial connected impedance sharing of voltage applies, but never assume that it is a equal share.

This is why almost all Tesla coil builders only run their bridges at 2/3 of the voltage rating, 400VDC on 600VDC devices and 800VDC on 1200VDC devices.

[oneKone]

350-400 mm sparks from a 300 mm secondary coil is great results for a SSTC, it is hard to get them much longer than this without really stressing the switches and failure rate rises.

That is a fine deal getting two nice and small variacs at that price. a 5A will do fine in almost every aspect, its just a matter of low enough BPS and you could feed a large coil from it

IGBT/MOSFETs are sensitive to overvoltage, I wouldn't go above 400VDC for 600V rated parts just to have a bit of overhead

what was weird to me is that the igbts failed when i turned down the duty cycle on the interrupter to walk over and turn down the variac..... if someone could elaborate more on that it would be

Did you lower the BPS too? I noticed on my DRSSTC when I lower BPS below the certain value the primary current raises. It may have something with the ionization of air, at lower BPS air could have enough time to deionize.

this could be my lack of knowledge showing.... but in a half bridge configuration at 500v input wouldn't each igbt be switching 250v? or have i confused this.

the bps were the same (sorry i don't have a number), but to turn down the variac my arm had to be near the antenna for the feedback. i'm not sure if this accounts for anything or not.

In in half-bridge (load connected to negative rail) each switch sees the full DC bus voltage + switching spikes/transients, since the current is only passing through one switch to the load. If you are using a capacitive voltage divider for the load, instead of negative rail, the voltage is of course halved.

In a full-bridge you have a upper and lower leg switch with the load in the middle, here the voltage is across two switches and serial connected impedance sharing of voltage applies, but never assume that it is a equal share.

This is why almost all Tesla coil builders only run their bridges at 2/3 of the voltage rating, 400VDC on 600VDC devices and 800VDC on 1200VDC devices.

Cheers for the info, everytime I visit this forum I learn something!

I changed the primary wire awg and was able to get 400mm sparks to free air on 400v, it has given me something to consider when I make the fullbridge PCB.... Anyways a picture! The coil still runs at a relatively low bps.

[oneKone]

So last night I remembered about madz talking about an internal primary, so I thought let's try it! I wound 5 turns internally and gave it a shot....... It preformed basically the same as an external primary.

Sorry for the crappy picture.


After this it got me thinking....... Twin primaries! So in total I now have (pretty average windings) 5 internal and 5 external. I tried series and parallel, the series really didn't make much sense to try but I did anyways without good results. The parallel windings did though! With 400v DC input I got a spark length of 460mm (although this seemed like a freak spark) I haven't measured multiple sparks yet but it definitely performs better than a single primary.


This weekend I'm planning to make the base and some more topload capacitance. I'll also wind both secondaries the same to hopefully increase efficiency.

[Mads Barnkob]

Interesting with internal/external dual primary coil, this really raises the coupling between primary and secondary, without having the primary coil higher up on the coil.

This might actually be a quite clever way of enhancing the performance of SSTCs, but eventually you will have flashover issues between secondary coil and primary coils, that is the end result of high coupling and pushing a higher and higher voltage out of the secondary coil

Your coil could certainly use a larger topload, minor toroid diameter the same as secondary form diameter and major toroid diameter the same as secondary coil length, that is a fairly certain rule to not oversize it.

[aie212]

sstc is cool ..
I learned a lot here.
how to make good coil, topload, etc ..

[oneKone]

I got to work this weekend and finished the sstc.... Almost! I'm going to have to shield the driver as it sits below the secondary and sounds like it's getting interference. I ended up going a full bridge of Chinese 60n60s. It has the dual primary internal/external with 5 turns. At 360v DC input it produces over 500mm sparks to free air.

/>

[oneKone]

I don't suppose anyone runs CT feedback on there sstc? With my coil I have to have the antenna about 300mm away from the coil. I'm not sure if this is a shielding problem yet, as the controller sits basically under the secondary.

Essentially what I'm asking is CT feedback a reliable alternative to an antenna or should I just live with (basically stop bitching!) Setting up an antenna for when I demo it?

Cheers

[Mads Barnkob]

I don't suppose anyone runs CT feedback on there sstc? With my coil I have to have the antenna about 300mm away from the coil. I'm not sure if this is a shielding problem yet, as the controller sits basically under the secondary.

Essentially what I'm asking is CT feedback a reliable alternative to an antenna or should I just live with (basically stop bitching!) Setting up an antenna for when I demo it?

Cheers

The only alternative is a secondary base current CT, but there is not that many advantages except that you no longer have the antenna to set up, but it can also be unstable and if you have a bad grounding of the secondary coil you can also risk flash over to the CT.

Since a SSTC is a single resonant Tesla coil, we need some sort of feedback from the secondary circuit to drive the primary circuit from, it is not like in a DRSSTC where we can use primary current feedback.

[oneKone]

I don't suppose anyone runs CT feedback on there sstc? With my coil I have to have the antenna about 300mm away from the coil. I'm not sure if this is a shielding problem yet, as the controller sits basically under the secondary.

Essentially what I'm asking is CT feedback a reliable alternative to an antenna or should I just live with (basically stop bitching!) Setting up an antenna for when I demo it?

Cheers

The only alternative is a secondary base current CT, but there is not that many advantages except that you no longer have the antenna to set up, but it can also be unstable and if you have a bad grounding of the secondary coil you can also risk flash over to the CT.

Since a SSTC is a single resonant Tesla coil, we need some sort of feedback from the secondary circuit to drive the primary circuit from, it is not like in a DRSSTC where we can use primary current feedback.

Cheers.
I should have specified it was a secondary CT I was talking about. But by the sounds of it an antenna is the easiest option while being reliable... I'll just make a small base for it.

[Hydron]

It's probably easy to test both. A CT is simple to wind, and you'll already have the clamping diodes for the antenna, could potentially just connect the CT in where the antenna would otherwise go. I have not built a non-DR SSTC though, so I may not know what I'm talking about!