About thyratrons

[PartialDischarge]

I'm considering buying and putting to work a thyratron, probably hydrogen, with the purpose on creating short steep current pulses. For something like 100s of amperes and 50-100us in duration I understand nothing beats the turn on speed of a thyratron.

Well, I stumbled upon one model that is described as both a "mercury vapor rectifying tube" and as a "HV mercury vapor thyratron". So my question is are this the same? I'm missing for example the maximum dI/dt that the tube allows, typical in thyratron datasheets.


Here are the datasheets:

https://frank.pocnet.net/sheets/030/d/DCG7-100.pdf
https://www.relltubes.com/filebase/en/src/Datasheets/DCG7_100%20and%20DCG7_100B.pdf

[ArthurDent]

Many years ago I worked at a company where they did testing to see how watthour meters (like you see on the electrical service for houses) stood up to induced high current pulses similar to lightning strikes on power lines. The equipment used a large thyratron to dump a bank of capacitors charged up to 20,000 VDC through a very large conductor running through the meter socket. The resulting pulse was 20 microseconds long and a peak current of 20,000 amps. The pooled mercury in the bottom of the tube was vaporized when the tube was triggered and then the tube would conduct basically becoming a direct short. 
 
If you visualize thyratrons as the tube version of an SCR it may make more sense. They are not linear but triggered diodes or switches.

[PartialDischarge]

The equipment used a large thyratron to dump a bank of capacitors charged up to 20,000 VDC through a very large conductor running through the meter socket. The resulting pulse was 20 microseconds long and a peak current of 20,000 amps.

Those figures are amazing, a hockey puck thyristor can match that amount of current and of voltage if put in series, but they are much slower in current rise time, and will break if the dI/dt exceeds for example 500A/us, where a thyratron can handle thousands of A/us.

[drussell]

Well, I stumbled upon one model that is described as both a "mercury vapor rectifying tube" and as a "HV mercury vapor thyratron". So my question is are this the same? I'm missing for example the maximum dI/dt that the tube allows, typical in thyratron datasheets.

Well, a thyratron is a rectifier, but a "controlled" one, just like an SCR is a controlled version of a silicon rectifier.

[N2IXK]

Hydrogen (or deuterium) thyratrons have much faster de-ionization times than mercury vapor or other gas types. They were pretty much the standard for high repetition rate pulsers in radar applications.

[T3sl4co1l]

Do you really need the speed (~nanoseconds for the better hydrogen thyratrons)?  That's a disproportionately long pulse.

A pulse that long, can practically be generated with a xenon or mercury type, or much better, with SCRs that don't need heaters and have lower voltage drop.

Also consider using a saturable reactor to sharpen the leading edge.  This can get SCRs into the sub-microsecond range.  Or IGBTs down to tens of ns, or still faster devices into the sub-ns range.  Example: http://www.slac.stanford.edu/cgi-wrap/getdoc/slac-pub-5432.pdf

Tim

[PartialDischarge]

Do you really need the speed (~nanoseconds for the better hydrogen thyratrons)?  That's a disproportionately long pulse.

A pulse that long, can practically be generated with a xenon or mercury type, or much better, with SCRs that don't need heaters and have lower voltage drop.

Not interested much in the length of the pulse, but in the dI/dt of a high current pulse. And this is non-repetitive for testing purposes only, so better.
Thyratrons I'm learning about them and researching for now.
SCRs have high I2t ratings which is good, but leaving aside that don't turn off at will, have around a measly 3 to 10 Amperes per 10ns critical rise time, at least the >3 kV models I've researching.
IGBTs are faster and they are my first choice for now, I have a few FF1200R12KE3  and FF1000R17 and I will definitively play with them to see where I can take them. At least the specs are promising

[T3sl4co1l]

Right, high voltage parts are slow as molasses.  They have to be, it's just physics.

There are a lot of old machines, still around to this day, that used stacks of 200V SCRs (operating at, say, ~650V DC link, they'd need at least four in series for each switch in the inverter), because they're faster than the higher voltage parts.  One example I've seen, a Radyne Ltd. power supply (ugly green paint -- classic 70s industrial!) driving an induction heating load at 3rd harmonic for ~40kHz output.  Used small (37mm?) "hockey puck" devices to deliver something like 15kW.

So, you may find it's advantageous to go down rather than up.

Likewise for IGBTs, Stanford and others (or maybe not, I forget who all exactly) have some papers about extracting maximum peak power, how far the SOA can be pushed and what help can be drawn from overdriven gate voltage.

Offhand, I've seen quite peppy IGBTs for the plasma driver market.  They're the lowest voltage that's even practical in an IGBT, usually 330V (since any lower and a regular MOSFET is better).  A TO-3P rated for 500A peak is not bad at all.  These may be losing availability as PDPs fade back into obscurity.

Win Hill (of Harvard and AoE fame) is currently experimenting with SiC MOSFETs for high voltage ~(single digit)ns edges.  A few dozen in parallel will get you there.

Tim

[PartialDischarge]

Offhand, I've seen quite peppy IGBTs for the plasma driver market.  They're the lowest voltage that's even practical in an IGBT, usually 330V (since any lower and a regular MOSFET is better).  A TO-3P rated for 500A peak is not bad at all.  These may be losing availability as PDPs fade back into obscurity.

Win Hill (of Harvard and AoE fame) is currently experimenting with SiC MOSFETs for high voltage ~(single digit)ns edges.  A few dozen in parallel will get you there.

Tim

Ill look into those igbts. For lower currents this solution from TI seems pretty useful due to the internal driver and included protections, although of course it can be put in parallel

http://www.ti.com/lit/ds/symlink/lmg3411r070.pdf

[PartialDischarge]

I forgot to add an interesting reference about how Tektronix resorted to thyratrons to test their A6303 current probe. It is a very interesting read and one more reason why I personally admire this probes


(From http://w140.com/tekwiki/wiki/AM503)


The P6302 and P6303 probes utilized a Hall device that was manufactured in the clean room of the Accessories Manufacturing group. It was deposited onto a bar of ferrite that was later assembled into a U shape with other ferrite and potted in a mu-metal can along with the transformer bobbins. The Hall device used for the DC measurements used a vacuum deposition process with indium antimonide. The cores were lapped and polished to a few Fresnel lines flatness to minimize the gap on the sliding ferrite. The L/R time constant affects the point where the Hall device and coils’ bandwidths crossover.
An interesting aspect of the design was trying to find a way to test and calibrate the peak current pulse on the larger P6303. Luckily we had the tube lab. We developed an argon filled thyratron that could discharge a 4 kV charge line into a 4 Ω load. The load resistor was designed and built by Tektronix. A large rectangular ceramic plate was coated with a metal film resistor. It had a voltage divider tap to allow for a safer measurement point. It was laser trimmed for accuracy. The current probe would measure the current to ground through this resistor, so the voltage was near zero for the user. The 4 kV supply was charged into a 4 Ω transmission line so a clean high current pulse would be generated. This concept was taken from Tek’s 109 pulse generator.
A trigger circuit was designed to fire the grid. It was based on the xenon flash circuit of the C5 camera flash, also in the Accessories group. The first prototype I made arced across the laser trim lines in the metal film resistor as they were cut perpendicular to the current flow. This caused high voltage gradients across the film and thus the arcing across the cuts. We changed the laser trim to be parallel to the current path along the outside edges. That’s what was used in manufacturing for calibrating the risetime of the P6303.
Also interesting, the P6303 required special potting epoxy developed by 3M. The epoxy used in other current probes put excessive stress on the larger ferrite that caused the inductance to drop to zero due to the magnetostriction property of ferrite. The epoxy could even also cause shear fractures in the ferrite.

[PartialDischarge]

Now "this" is a proper thyratron, couldn't resist getting one

[drussell]

Now "this" is a proper thyratron, couldn't resist getting one

Indeed!  Now thats a zapper!! 

[T3sl4co1l]

http://www.mif.pg.gda.pl/homepages/frank/sheets/030/d/DCG7-100.pdf if anyone is curious.

Tim

[drussell]

Actually using it for anything could be a bit... errrrr... interesting. 

The heater power on that beast is only 100W but it should be interesting to fire it up for the first time since after the usual 10 minute heater warm-up you're supposed to pre-heat the anode for at least 35 mins using 500 volts, (limited to 6A.)  You can probably get away with a bit less than the full 3000W but need to pass some serious current though it.

Perhaps a couple good sized heating elements in series as a load?

[PartialDischarge]

Rectified AC mains (240 here  ) should be enough to heat that, now I have to solve a more mundane problem which is getting pins that size (8mm diam, 28mm length) to make a tube socket.

[drussell]

Rectified AC mains (240 here  ) should be enough to heat that, now I have to solve a more mundane problem which is getting pins that size (8mm diam, 28mm length) to make a tube socket.

It wasn't the source of power that I was thinking of as being the issue, rather the question of how to limit it to the 6A. 

[T3sl4co1l]

Would expect a sodium lamp ballast to do a good try, give or take exact values of OCV, L and C.

Tim