Solid state tesla coils - general notes

<< back

1. Which topology?

Which switch mode power supply (SMPS) topology should one start with? IMHO although the schematic of a full-bridge looks a bit complicated compared to push-pull and half-bridge designs, sticking straight to a full-bridge topology or its smaller version, the half-bridge, is absolutely worth the initial extra effort.

The half-bridge topology:

(Simplified schematic, gate drive transformer and drive circuit as well as the required extra diodes around the mosfets, and capacitor equalizing resistors are not shown)

The full-bridge topology:

(Simplified schematic, gate drive transformer and drive circuit as well as the required extra diodes around the mosfets are not shown)

In the end, despite the more complex look of these topologies, there's really much less that can go wrong when compared to the various troubles that come up in other topologies like push-pull, forward converter variants, flyback, etc

The main "challenges" in the SSTC design process are to make a good RF PCB layout, and to properly design the gate drive transformer.  Both can take some time to get finished... The rest of the circuit is very easy to build.

Here's a "short" comparison of the usual switch mode power supply topologies used for SSTC driving.

  Half-bridge and full-bridge Push-pull Forward converter
minimum mosfet voltage rating supply voltage plus a safety margin

230VAC*1,414 + 50V
=> 400V mosfets
double the supply voltage plus a large safety margin

2*230VAC*1,414+100V
=> 800V mosfets
double the supply voltage plus a large safety margin

2*230VAC*1,414+100V
=> 800V mosfets
other mosfet properties medium voltage 400V fets, so a <0.2 Ohm channel resistance is typical => high current, low loss high voltage 800V fets, >2.0 Ohm channel resistance is typical => low current, high loss high voltage 800V fets, >2.0 Ohm channel resistance is typical => low current, high loss
mosfet body diode: must be disabled, otherwise mosfets explode

if supply <50VDC:
one reverse schottky in parallel is ok

if supply >50VDC:
more complicated - one series low voltage high current schottky, one parallel high voltage ultrafast recovery diode (<250ns)
can be ignored can be ignored
realistic power levels many kW
large impressive SSTC
some 100W
small SSTC
some 100W
small demo SSTC
what limits the power level base-feed transformer core power handling capability (saturation, induced currents causing core heating)

mosfet current ratings (paralleling more than two fets, the right way, is tricky)

mosfet switching and conduction losses
primary leakage inductance, huge voltage spikes (up to kV range) at increasing power levels, makes use of snubber circuits imperative (=>high heating losses and low efficiency, and high circuit complexity)

>=800V fets are expensive and can't handle much current
(same as for push-pull)
interruptor can be used for higher peak powers at same average power yes, no problem, as long as mosfet peak current ratings are not exceeded yes, but with all the ringing and voltage spikes the peak power push-pull can handle is also the maximum average power, i.e. there's no real TC streamer gain from using an interruptor (same as push-pull)
direct TC primary drive yes, easy yes, needs two identical windings yes, but inefficient
base feed transformer design needs only one primary

transformer design non-critical
needs two identical and well coupled primaries, critical design - requires skills! ;o) critical design, only one primary

only the first quadrant of the ferrite cores' B-H curve is used, i.e. "transformer core running at only half of what it could handle".
base feed transformer volt-seconds (Vs) imbalance:

(=swirls of smoke from the transformer and silicon shrapnel embedded in the wall)
full-bridge: minimal danger of saturation, Vs imbalance mainly due to slight differences in mosfet channel on-resistances

half-bridge: if the primary has a series coupling capacitor, then Vs imbalance is no big problem

major problems with Vs imbalance as fully identical pri windings are almost impossible to make. The driver circuit absolutely must have pulse-by-pulse current limiting.  
tuneable down to DC / 0Hz yes, by using a primary series coupling capacitor no (short-circuit at 0 Hz) no (short-circuit at freq towards 0 Hz)
problems grief with gate drive transformers or floating channel mosfet driver ICs or optocoupler-tweaking grief with mosfets constantly dying on overvoltage, gate drive noise (same as for push-pull)

Why only these four topologies? The main reason is that all of them can be manually (or automatically) tuned in output frequency, without this SMPS/SSTC running completely out of control.

Of course there exist other topologies too. Actually there's a HUGE variety of topologies and their small variations, so it is maybe tens? hundreds? even thousands? Jeez, who invented them all...? ;) These other topologies are either very complex, or require careful supervision by a dedicated fast controller IC so that nothing blows up, or they run only at a fixed frequency, or for some other reason won't work so well in power (>100W?) SSTC use.  Then again, if you've found a promising SMPS topology that might easily work in SSTCs too, feel free to correct me (jwagner@cc.hut.fi), or mayber better, suggest it on the Tesla coil mailing list www.pupman.com


Update: in '03 Terry Fritz devised a new semi-SSTC type of silicon controlled TC, the OLTC (off-line Tesla coil). It reminds a bit of an off-line buck boost converter with a freely oscillating resonant load. The OLTC runs without any HV transformers, and uses a single IGBT switch (or parelleled IGBT's) to replace the spark gap. Steve Conner [scopeboy] has more infos at OLTC FAQ 1.0 and working OLTC designs.



next ("Recommended reading") >>