Tesla Coil Topologies: Which to Choose?

[Rectifier]

Hi! Nice forum you have here.

I am an engineer about to build my first (SSTC) Tesla coil transformer, and have been confronted with the myriad of different topologies available. I would be very thankful for any help in orienting myself and choosing the right one. My main questions are at the bottom of this post

I'm looking for a version which is very small/compact and efficient, outputting ~20-30kV in about ~40W power - but not the most complicated to begin with.

My best understanding at the moment is that there are many different topologies, but that one can capture the most common ones with these categories (I think! You may correct me if I'm wrong):

• Single-transistor (Those going by the names "slayer exciter", "single mosfet", and sometimes "HFSSTC", etc.)
• Half-bridge single-resonant
• Full-bridge single-resonant
• DRSSTC (driven by half or full-bridge)
• Class E (Those based on "Class E" RF amplifier driving topologies)
• (And I'm sure there are plenty more, but these seem to be the most popular ones.)

Now as I understand it, DRSSTC's are not recommended for a first build due to their complexity, so I will rule them out for now.

Also, based on the website by Richie Burnett (and others), the "Class E" topology is very sensitive to load impedance matching, and can thus be relatively complicated to tune such that it gets efficient, and the MOSFET doesn't overheat. It seems that this topology should therefore be excluded for simplicity reasons. It looks simple in its component count, but I gather this might be a bit "deceptive" for the newbie. Would be interested in some feedback on this point, if I'm understanding things right.

Some questions I have about the most suitable topologies:
1. Are Half-bridge, single-resonant, SSTC's somehow inherently more powerful or efficient than single-MOSFET topologies? What would be the benefits of "upgrading" to a half-bridge?
2. What would be the advantages of a full-bridge (single resonant) driver over a half-bridge or single MOSFET version?
3. Youtube is full of small single-transistor "HFSSTC" tesla coils said to be operating at several MHz. These look very interesting to me. Are half or full-bridge designs capable of these high frequencies, or are they necessarily limited to somewhat lower frequencies (and thus need larger coil sizes etc.)?
4. Most single-MOSFET designs I've seen have fairly short sparks, which indicate a relatively low output voltage compared to other designs. Is this necessarily so with these topologies? Can one get higher voltages from them, perhaps by pulsing them?

I have more questions about this interesting field but these will be enough for now

/Regards

[Weston]

I saw you were asking about this on IRC, but I was AFK at the time.

Typically performance improves as complexity improves, as is true with many things in life.

Half Bridge / Full bridge tesla coils place less stress on the transistor for a given power rating and as such are typical more efficient. A full bridge tesla coil compared to a half bridge doubles the voltage swing across the primary. For a given supply voltage and MOSFET rating it will allow you get higher power than a half bridge, otherwise its basically the same. Once you have a working half bridge coil its pretty easy to upgrade it to a full bridge coil.

Those HFSSTC coils on youtube are basically just self oscillating class E coils and have the same tuning constraints. Class E uses a tuned network across the MOSFET to drive the voltage to zero before it turns on, greatly reducing switching loss. Something similar is possible with a half bridge or full bridge, but its a lot more tricky to get right. With a half bridge / full bridge tesla coil you are going to be limited to ~500KHz, and preferably a lower frequency, before things start getting complicated.

Most non DRSSTC coils are going to have relatively short spark lengths. High frequency tesla coils are also going to have a shorter and hotter spark. Typically the spark length is limited by the current and voltage stress on the MOSFETs or over heating. A lot of the single MOSFET designs based on the traditional "slayer" circuit are not well optimized so they stress the MOSFET a lot more. With an interrupter you can run the coil at a higher peak power for a short duration of time, the DRSSTC is the ultimate example of this, mant kW of peak power for only a few hundred microseconds.

Its easy to get discouraged if nothing works,  the single transistor self oscillating designs are typically the easiest to get some spark out of and are a really easy way to start. Beyond that the phase locked loop (PLL) half bridge designs are pretty good.

I am not that fond of the open loop frequency control and the focus on snubber networks, but I would strongly suggest watching Diode Gone Wild's youtube video series where he builds a SSTC, its pretty informative about power electronics in general and some tesla coil theory

/>

[johnf]

Ditto to what Weston has said
Slayer and other single device oscillators seem to be sacrifice options to the gods of high voltage
ie when not everything is right poof there goes another bit of sillycon.

half or full bridge are the best options --these are very probable with oscilloscopes etc to see what is going on when things are not right as long as you are not going full out ie things turned down a bit --lower buss voltage --lower duty cycle.
There are plenty of users here that have posted with half and full bridge builds that havee not sent too many Fets to the sacrifice table

A good circuit diagram and a layout that follows that in actual circuit build is a very good start

[Mads Barnkob]

Hi and welcome to HVF! It is indeed a lovely place we have here

Which direction of engineering did you study/do you work in?

It took me a long while to jump from SSTC to DRSSTC, but I really should just have tried earlier than I did. It is not as complex, if you look away from understanding how the universal drivers work internally and only focus on understanding what dual resonance means and how it works in a Tesla coil. I tried to put some words on that here: http://kaizerpowerelectronics.dk/tesla-coils/drsstc-design-guide/drsstc-faq-frequently-asked-questions/

If you do not have a oscilloscope already, get a cheap sub 100MHz DSO (Rigol, Owon, Awtek etc) or even good old analog 2ch 20MHz can be used. But it is mandatory to find errors in these inverters and drivers. Do not get tempted to buy the cheap DSO138 handhelds and such on ebay, wish, aliexpress, they are close to useless for anything above 20-50 kHz.

I will any day recommend Steve Wards SSTC-5 or known as the MINI-SSTC as a beginners choice. He developed the driver over a course of 5 coils and it reached a perfect state of simplicity, robustness and spark output capabilities. If you want to control the output voltage, as much as it is possible. You can do so from on-times, BPS or input voltage. Even if you do not have a variac to control the input voltage, a SSTC could be controlled at low power with a simple SCR based light dimmer.

Steve Wards SSTC5: http://stevehv.4hv.org/SSTC5.htm
My version of his SSTC5: http://kaizerpowerelectronics.dk/tesla-coils/kaizer-sstc-i/ + http://kaizerpowerelectronics.dk/tesla-coils/kaizer-sstc-ii/ + http://kaizerpowerelectronics.dk/tesla-coils/kaizer-sstc-iii/

Gao / Loneocean have also written some great articles on his SSTC builds: http://loneoceans.com/labs/sstc2/

I put some SSTC specific topics into this guide: http://kaizerpowerelectronics.dk/tesla-coils/sstc-design-guide/ and many of the topics in the DRSSTC guide goes for the SSTC as well: http://kaizerpowerelectronics.dk/tesla-coils/drsstc-design-guide/

I am looking forward to see your project