160mm PLL sstc

[oneKone]

I've decided to build a PLL sstc while saving for my drsstc build.

For a while now I've wanted a big-ish CW sstc, so I've decided to pull the 160mm secondary off the shelf I used for my second sstc. For the driver I'll be using this: https://www.mikrocontroller.net/attachment/284427/U3RBsZAI.png

One question I have is how to tune it! On pins 11/12 I have 10k resistors with 50k pots so I can easily set the lower and upper frequency limits.... But I think that's where I get confused!

I've read that I set the upper and lower limits by adjusting the vco pot. So by turning the vco pot all the way in one direction set pin 11, then turn the vco in the other direction and set pin 12, then finally fine adjust the frequency with the vco to get a perfect waveform (or as close as possible) on the output of the bridge. So basically if I had a secondary with a fres of 170khz the upper limit would be 190khz and the lower 150khz and the vco would be tuned to 170khz.

Hopefully I haven't made a total mess of the paragraphs and it can be understood! The problem with things like this is I read too much into it without getting an understanding and just end up getting more confused!

[alan sailer]

The antenna on the schematic implies that the system tunes itself using feedback. The output from the secondary
creates an electromagnetic field which is picked up by the antenna and fed into the 4046.

I'm not familiar with the 4046 chip but it is a phase locked loop controller so it's function is to tune your output to
match the input. I'd hazard a guess that the upper and lower pots set the basic oscillation frequency close to the
natural frequency of your tesla coil. This would get the darn thing started. As it starts up the coil will go towards
it's natural frequency sending a signal through the antenna to the chip so that it tunes it's oscillation frequency
closer to the coil. After some cycles the chip frequency matches the coil frequency and bingo, your loop is phase locked.

The benefit of this scheme is if you detune the coil by bringing your hand/CCFL etc close to the coil it changes the
natural frequency of the coil. Normally this would reduce your output but with the feedback the frequency changes
automatically for maximum output. Within limits...

Cheers.

[oneKone]

@alan cheers for the reply. You were correct! Part of my problem was that I had the antenna too short so it didn't have enough feedback to lock on.

One thing I'm thinking with this coil is to use a cm75bu-12h brick. The setup itself will only be drawing 2kw, but the brick design makes for a very clean full bridge setup. Hopefully someone has some thoughts on this?

[Mads Barnkob]

I guess you can find some information and knowledge from the original PLL SSTC design by Steve Conner: http://www.scopeboy.com/tesla/dwsstc/wk2.html

Looking forward to see your results, 2kW of CW flame is somewhere around 25-30 cm of white hot flame dishcarges!

[oneKone]

I guess you can find some information and knowledge from the original PLL SSTC design by Steve Conner: http://www.scopeboy.com/tesla/dwsstc/wk2.html

Looking forward to see your results, 2kW of CW flame is somewhere around 25-30 cm of white hot flame dishcarges!

cheers again mads.

i had originally looked at steve conners site. At the time that was the mains source of my confusion! At the time i just overwhelmed myself with peoples findings on the pll circuit. Also..... 25-30cm!!! thats huge! would that be half wave rectified? or fullwave and heavily filtered?

[Mads Barnkob]

Full-wave rectified with 1500 uF on the DC bus, its the SSTC2: http://kaizerpowerelectronics.dk/tesla-coils/kaizer-sstc-ii/ it is not a PLL coil, but the power level should give the same result.

From 1:40 you can see it running in CW mode

[oneKone]

Cheers for the video mads.

I've been having some dramas driving the brick at 180khz. I have a rise/fall time of about 0.7us/ 700ns, To me that kinda sucks! my current drive method is as i linked in my first post except i'm using tip41/42 instead of bd139/140. A downside of this is that i can only drive the gates with 12v and if i make a step-up gdt the waveform turns sinusoidal. I think the best way to fix this is to make an output stage with npn/pnp mosfets and a gate drive ic (the same method as almost every UD drsstc driver)

I'm thinking that this brick might just not be able to switch fast enough? One thing that i could just do (that will make me more comfortable) is just to have a parallel fullbridge of irfp460 mosfets, OR should i just stick it out with the brick and not give up so easy?

Unfortunately i don't have any pictures of my setup yet but hopefully i'll get to it today! i do however have a short video.... like really short! the power consumption is about 1400 watts with about 240mm discharge to air.

 

[oneKone]

Woah....... What a long ass time!  After a pretty damn long hibernation I'm back!

So this project...... This sat in limbo for some time. At the time I had some problems driving the brick which eventually lead to its failure, but since then I've learnt a few things. I don't have many photos of the construction but it's a pretty simple coil.

Drive circuit remains a simple pll, the gate drive I hacked from my flyback driver - which I basically stole from the ud. Drivers. The bridge is a dual h-bridge of 460 MOSFETs, at this time I can't remember the number of primary turns.

Pics!

/>
I'd like to thank Dean aka Tesla explorer, he was instrumental in putting my ass into gear! Especially for the dual h-bridge. Check out his channel:
https://www.youtube.com/channel/UCn9M2WGRdbJxhX-Z80IJ41Q