Modular-based large DRSSTC

zrg, Wed Jul 30 2014, 11:31PM

Hi everyone.
Our team has been working on this project about a month or two (mainly lazily searching for a water jet cutting service to slice the fiberglass 15mm thick sheet).

This is a regular type DRSSTC, designed to produce 5-6 meters of spark, quite similar to Phillip Slawinski's DRSSTC VI by size. However, having some desire to make it easy to assemble and disassemble for moving the whole thing around, we thought that it would be nice to make it modular: MMC module, bridge module, driver&automation module. All modules are connected with each other by power and/or signal cables and can be easily changed to some other modules (larger MMC, bridge with other bricks, etc) that would fit in this coil.

Total height: about 280-300 cm.
Secondary: 180x31.5 cm, 1.1 mm wire.
Topload: 180x31.5 cm, will be replaced to proper aluminum tubing soon.
Fsec: 35 kHz
Primary: 7/8" copper tubing, air cooled. Stays cool however.
MMC: custom made Russian power film caps, 220nf 20kV each, 5 pcs total. Cost about 82$ per cap.
Power bridge: 2x CM600DU-24NF, 6x CDE 12000@450V caps, 18000uf total.
Power input: 3p 380V, probably something around 20-30 kW.
Driver: Steve-based (well, there aren't much non-Steve based DRSSTC drivers nowadays, are there) custom driver.
OCD set to 4000A.

First post: coil model and power bridge.
1406763049 4762 FT0 V8gq1mia7wc

1406763049 4762 FT0 Yzyqfqdvrce

1406763049 4762 FT0 Img 0275

1406763049 4762 FT0 Img 0280

1406763049 4762 FT0 Img 0284

1406763049 4762 FT0 Img 0890
Re: Modular-based large DRSSTC
zrg, Wed Jul 30 2014, 11:37PM

Second post: primary and secondary.

Other pieces will be shown a bit later, tomorrow or something like that :)
1406763414 4762 FT165062 Img 0311

1406763414 4762 FT165062 Img 0314

1406763414 4762 FT165062 Qtkqm3 Xxo8

1406763414 4762 FT165062 Dh50ncjauii

1406763414 4762 FT165062 Raohpa2zbnw

1406763414 4762 FT165062 Img 1008

1406763414 4762 FT165062 Img 1014

1406763414 4762 FT165062 Img 1020

1406763414 4762 FT165062 Img 1021

1406763414 4762 FT165062 Img 1022
Re: Modular-based large DRSSTC
Mads Barnkob, Thu Jul 31 2014, 10:22AM

This coil looks good, all the waterjetted parts are really nice and makes me wonder wether to look for that option in the future, is it expensive to get such jobs done or do you know someone that can do it for you cheap?

My own large DRSSTC is also made modular into bottom box module with bridge and mmc, upper platform with primary coil, secondary coil and topload, 4 parts that are easily taken apart with a total of 3 screws and locking mechanism.
Re: Modular-based large DRSSTC
Steve Conner, Thu Jul 31 2014, 01:43PM

The modular approach also worked well for me with Odin. When I discovered that I'd made the tank capacitor far too big, it was easy to swap in a smaller one. Later I hope to reuse the original tank cap and swap in a bigger bridge. Obviously it would have been better to get it right from the start, but whatever... smile

Another handy feature was the ability to take the resonator off the top of the electronics box and stand it to one side. It gives some more sparking room in indoor venues with low ceilings.
Re: Modular-based large DRSSTC
Linas, Thu Jul 31 2014, 06:15PM

Steve Conner wrote ...

The modular approach also worked well for me with Odin. When I discovered that I'd made the tank capacitor far too big, it was easy to swap in a smaller one. Later I hope to reuse the original tank cap and swap in a bigger bridge. Obviously it would have been better to get it right from the start, but whatever... smile

Another handy feature was the ability to take the resonator off the top of the electronics box and stand it to one side. It gives some more sparking room in indoor venues with low ceilings.
I have same size coil as you, and right now reworking my MMC, and i can make 0,5 0,65 or 0,8uF tank capacitor. What should i use ? (full bridge from CM600)
Re: Modular-based large DRSSTC
zrg, Thu Jul 31 2014, 06:20PM

MMC, primary support (18 mm thick plywood) and control box.
1406830703 4762 FT165062 Img 0889

1406830703 4762 FT165062 Img 0972

1406830703 4762 FT165062 Img 1001

1406830703 4762 FT165062 Img 1002

1406830703 4762 FT165062 Img 0956

1406830703 4762 FT165062 Img 1004

1406830703 4762 FT165062 Img 0955

1406830703 4762 FT165062 Img 1064

1406830827 4762 FT165062 Img 1060
Re: Modular-based large DRSSTC
Steve Conner, Thu Jul 31 2014, 09:46PM

I see some sort of blurred-out secret circuit board... 8)

On a more serious note, what is the point of putting the key switch and circuit breaker right in the bottom of the coil? You are going to have to walk up to the coil, throw that breaker, turn the key, and hope that the coil doesn't start firing for some reason. And if it starts to malfunction, you have to walk up to it to shut it down.

Linas, what is the frequency of the system you're planning?
Re: Modular-based large DRSSTC
zrg, Thu Jul 31 2014, 10:03PM

For all DRs we've built so far, none of them ever tried to start without control signal on the optics input. So far that box position seems reasonable enough. However, that circuit breaker is used mainly as a power switch; there is another one in the middle of the power cable, which can be used to power on and off the whole coil if needed.
Re: Modular-based large DRSSTC
Steve Conner, Thu Jul 31 2014, 10:23PM

I thought about this for a while and ended up getting rid of everything except that breaker in the middle of the power cable.

The key switch might be good to prevent unauthorised use of the coil by other people, I guess.
Re: Modular-based large DRSSTC
Linas, Fri Aug 01 2014, 10:29AM

zrg wrote ...


1406830827 4762 FT165062 Img 1060


why driver electronics intentionally have blur added ? what you don't want us to see ? (top right corner)
Re: Modular-based large DRSSTC
zrg, Fri Aug 01 2014, 01:24PM

Linas wrote ...

zrg wrote ...


1406830827 4762 FT165062 Img 1060


why driver electronics intentionally have blur added ? what you don't want us to see ? (top right corner)

Well, you know, Tesla coils are a bit of economical bubble here in Russia nowadays, therefore a lot of people are hunting for a cheap and reliable SSTC/DRSSTC schematics. Our group managed to develop a really stable and well-performing one, and there are approved cases of even blurred and low-quality photos of our schematics being thoroughly examined by some interested in that guys cheesey . Sounds paranoid, but I'd prefer not to publish anything related to that driver. And this photo of the box was the only one available on my memory card, so I decided just to blur it finely. Just in case wink

Whole coil assembly.

1406899376 4762 FT165062 Img 1045

1406899376 4762 FT165062 Img 1050

1406899376 4762 FT165062 Lhatakrfopg

4a

1406899376 4762 FT165062 Img 1026

1406899376 4762 FT165062 Img 1216

1406899412 4762 FT165062 Img 1032

1406899412 4762 FT165062 Img 1041
Re: Modular-based large DRSSTC
Linas, Fri Aug 01 2014, 02:15PM

zrg wrote ...


Well, you know, Tesla coils are a bit of economical bubble here in Russia nowadays, therefore a lot of people are hunting for a cheap and reliable SSTC/DRSSTC schematics. Our group managed to develop a really stable and well-performing one, and there are approved cases of even blurred and low-quality photos of our schematics being thoroughly examined by some interested in that guys cheesey . Sounds paranoid, but I'd prefer not to publish anything related to that driver. And this photo of the box was the only one available on my memory card, so I decided just to blur it finely. Just in case wink


Any good new features, like phase lead, or you some have managed to get out of closed loop and still make good feedback ?
Since in University we developed technique that at start up scans for frequency, where is highest power level, and use that for 50% PWM signal. Now if we use LC circuit, should we care about current feedback , if we just can measure at each startup Fo, and use HRPWM to generate switching frequency, and at this point, iGBT will always work in ZCS ?

I have idea simply to use 100W 50V transformer to scan coil frequency range, and after that switch high voltage for normal operation (and short CT with relay for extra protection), what does members think about that ? Since we don't have any IGBT fail at this time, but coil was used only few times
Re: Modular-based large DRSSTC
Steve Conner, Fri Aug 01 2014, 05:08PM

The resonant frequency falls over the course of the burst as the streamer grows. It also changes suddenly when a ground strike connects. You have to track these changes in real time, so measuring the frequency once at startup isn't good enough.
Re: Modular-based large DRSSTC
zrg, Sat Aug 09 2014, 12:03AM

Got the coil outside and running. Tomorrow's first big public perfomance. Keep an eye on updates.
1407542579 4762 FT165062 Img 7036

1407542579 4762 FT165062 Img 7037

1407542579 4762 FT165062 J8wtsw5nnei

1407542579 4762 FT165062 Img 7755

1407542579 4762 FT165062 Img 7828

1407542579 4762 FT165062 Img 7824
Re: Modular-based large DRSSTC
loneoceans, Sat Aug 09 2014, 12:24AM

That looks great! I always wanted to make a spirally toroid like the one you made the first time I saw some field diagrams of tokamaks ( Link2 ) but never had the resources (or need really) to build one. I'm glad you pulled it off so nicely smile Looking forward to your public performance!
Re: Modular-based large DRSSTC
Patrick, Sat Aug 09 2014, 03:52AM

very cool torus !

Re: Modular-based large DRSSTC
Linas, Sat Aug 09 2014, 07:41AM

zrg wrote ...

Got the coil outside and running. Tomorrow's first big public perfomance. Keep an eye on updates.
So, can you give any feature set of your DRSSTC controller ?

For toroid, i have very cool idea what i am going to try out. Where is solid extruded polystyrene foam, that is very strong, i will use it to create shell that is impact resistance ( or will need resurfacing after big impact)
i will buy 5cm extruded foam sheets, and with CNC router create layers, after gluing and lot of aluminum tape, it should be perfect
Re: Modular-based large DRSSTC
Steve Conner, Sat Aug 09 2014, 11:44AM

Cool! Very stylish toroid too! smile
Re: Modular-based large DRSSTC
zrg, Sat Aug 09 2014, 08:54PM

First day over, second day incoming.
Too tired to post anything interesting, so here's single photo.
1407617641 4762 FT165062 Bsfp8pxxpxg
Re: Modular-based large DRSSTC
zrg, Mon Aug 11 2014, 01:51PM

Second day over.
The coil performed nice, maximum registered spark length was around 6 meters. However we didn't have any chance to run it at night, so I'm unable to provide any really spectacular photos or video.
Primary was detuned about 10-15% from secondary resonance, which gave impressive sparks on 200-300 Hz and at the same time failed to play high MIDI notes. Looks like big coils need primary slider motor, like the one on Thumper DRSSTC smile

OCD was on 3600A, maybe 3800. The hottest parts are wires between primary and MMC or bridge (they're 50 mm^2, and that is surely not enough).
Actual power consumption unknown, but I think it must be around 25-30 kW.




1407765096 4762 FT165062 Img 8071

1407765096 4762 FT165062 Img 8076

1407765096 4762 FT165062 Img 8077
Re: Modular-based large DRSSTC
Steve Conner, Mon Aug 11 2014, 02:57PM

This looks great! smile I'm not surprised the 50mm2 cable gets hot though. Skin effect means that most of the copper in it does nothing. You would get better results with, say, 20 pieces of 2.5mm2 PVC insulated wire bundled together.

Why can't it play high MIDI notes, what happens when you try?
Re: Modular-based large DRSSTC
Finn Hammer, Mon Aug 11 2014, 04:40PM

zrg wrote ...

Second day over.

snip

Looks like big coils need primary slider motor, like the one on Thumper DRSSTC smile

ZRG,

The slider motor, and associated RAT technology, is an asset during the initial setup and tuning of the coil, because the sweet spot can be situated on a part of the primary coil only 5cm long. It may therefore be hard to find using "suck it and see". In this case, the sweet spot I refer to, is the tuning point where the primary current displays distinct notches.
I would love to see it developed to the point where it can track different operating conditions.
Thumper, also, has an affinity to playing the bass notes, a tendency that I think is caused by insufficient buss capacitance.

Russia very much appears to be the center of creativity these days, with both a DRSSTC Magnifier, and now a very seriously kick ass spiral toroid. The ability to be disassembled into a very compact unit during storage makes this toroid very attractive, apart from its aestetically very pleasing appearance.

Hats off to the Russkies!

Would you please expand on ways to determine the curvature and length of the individual segments of this toroid?

Cheers, Finn Hammer
Re: Modular-based large DRSSTC
zrg, Mon Aug 11 2014, 05:25PM

Finn Hammer wrote ...


Would you please expand on ways to determine the curvature and length of the individual segments of this toroid?


3D modelling in 3ds max and Pro/E (aka Creo Elements), then exported to CNC files and regular blueprints. It can automatically calculate all necessary data directly from the model.

By the way, the whole coil appears to be really steady and firm. Four of our team had to stand on the primary to erect the topload on the secondary top, and the coil didn't notice that at all cheesey

1407777713 4762 FT165062 V9qhlkbmb6c


1407777713 4762 FT165062 Hpiuprlbbb8

1407777713 4762 FT165062 Gcdnkrf2oq0

1407777713 4762 FT165062 94ms2ruevog

1407777713 4762 FT165062 Tuz2yeaaple

Mazuoki0

1407777713 4762 FT165062 L2wkatyyyto

1407777713 4762 FT165062 Img 7770

3ypdck
Re: Modular-based large DRSSTC
Hydron, Mon Aug 11 2014, 11:55PM

Steve Conner wrote ...

Why can't it play high MIDI notes, what happens when you try?
I assume his midi controller limits the on-times of high notes to keep the duty cycle down. With a 10-15% detuning for peak streamer length it might not be able to breakout quick enough to form a decent streamer and pull the secondary into tune with the shorter on-times of the higher notes. I have notice similar behaviour myself when using significant detuning.
Re: Modular-based large DRSSTC
Intra, Tue Aug 12 2014, 12:47PM

Yeeeeaa!! Great job, guys! So sweet long sparks.. So pleasing to the eye..
Re: Modular-based large DRSSTC
zrg, Tue Aug 12 2014, 09:16PM

Thanks for the feedback.
We're now planning to upgrade the bridge to lamitated bus and four single FZ600 bricks, since paired CM600's contact pads get overheated. Jet cutters promised to have it done by Friday.
1407878165 4762 FT165062 Ga3zkmegffm
Re: Modular-based large DRSSTC
Steve Conner, Thu Aug 14 2014, 09:35AM

Surely to replace 2x CM600 in parallel you need a FZ1200?
Re: Modular-based large DRSSTC
Hydron, Thu Aug 14 2014, 12:31PM

Steve Conner wrote ...

Surely to replace 2x CM600 in parallel you need a FZ1200?
They aren't in parallel, just two of the huge half bridge CM600 bricks to make the full bridge (see pics earlier).
I'm assuming the output contact for each half bridge is getting too hot from the current density - I've noticed that this can be a limiting factor on big bricks, even single IGBTs (see Link2 for an example limited to 500A in a 600A+ class IGBT).
Re: Modular-based large DRSSTC
Steve Conner, Thu Aug 14 2014, 03:02PM

Ah, I didn't notice it was the CM600DU halfbridge module.
Re: Modular-based large DRSSTC
Intra, Fri Aug 15 2014, 08:53AM

zrg wrote ...

Thanks for the feedback.
We're now planning to upgrade the bridge to lamitated bus and four single FZ600 bricks, since paired CM600's contact pads get overheated. Jet cutters promised to have it done by Friday.
1407878165 4762 FT165062 Ga3zkmegffm


Hm.. Look like exactly what I say in Link2 (Dangerous Link! Cyrillic Font!)
Re: Modular-based large DRSSTC
Intra, Fri Aug 15 2014, 08:57AM

It's not so far, that people will use 4 single IGBT in parallel with another 4 or more, only for reduce overheating of IGBT contact pads.
Re: Modular-based large DRSSTC
Goodchild, Fri Aug 15 2014, 02:01PM

I really like your guy’s toroid design! Have you noticed any corona on the ends of the tube segments? I ask because with our ring style toroids any imperfections or bumps on the outside of the ring caused breakout, typically unwanted!

I wonder if the way you have them positioned helps to shield the ends inside the e-field such that they don’t break out.

None the less awesome coil guys and my hats off to you.
Re: Modular-based large DRSSTC
zrg, Fri Aug 15 2014, 04:08PM

Goodchild wrote ...


I wonder if the way you have them positioned helps to shield the ends inside the e-field such that they don’t break out.


This. All the endings are at least 5-10 cm inside the outer contour of the whole thing. Together with 60-90 cm terminal they are well shielded.
Re: Modular-based large DRSSTC
Goodchild, Fri Aug 15 2014, 05:02PM

zrg wrote ...

Goodchild wrote ...


I wonder if the way you have them positioned helps to shield the ends inside the e-field such that they don’t break out.


This. All the endings are at least 5-10 cm inside the outer contour of the whole thing. Together with 60-90 cm terminal they are well shielded.

ah this makes sense! very cool design. I may have to “barrow” your design one day on some of my coils, if nothing else simply because it looks awesome. I also imagine it’s easier to manufacture than a full ring toroid. My experience with the ring toroids are that they can be finicky to assemble and have issues with breakout near the connections points/joints.
Re: Modular-based large DRSSTC
HV Enthusiast, Tue Aug 26 2014, 12:32PM

Wow! Thats a VERY beautiful and artistic looking toroid. May I ask what pushed your to design your toroid in that spiral fashion? Was it to create something very artistic looking, or was it out of necessity to to make fabrication simpler etc...

Looks great and I love how you do all your modeling. Very professional and very good engineering skills to have!
Re: Modular-based large DRSSTC
Chris Cristini, Sun Oct 19 2014, 04:42AM

Love the 3D modeling great work that is great engineering!!