1kW 100kV Tetrode Valve

Must it be a single valve?

--Europium

Yes these valves in a single block and very heavy. These are used for radio station RF generators in several kW power range and operate on 50kV to 120kV anode voltage. These were also used in X-ray power supplies.

The only mfr of high-power valves I'm familiar with is Eimac. They make some real monster valves, lemme tell ya! Here's a link: http://www.cpii.com/eimac/

Dear Europium

I am looking for more bigger valves. Thomson-CSF which is now Thales makes these.

TH5188 was the Thomson-CSF Tetrode. I need somewhat similar one.

This picture is perhaps for Tetrode Model 640 with thanks to Henning Umland.

Copyright © 1997-2006 Henning Umland

Judging from the size of that valve, I'd say just offhand the plate voltage isn't anywhere close to being 100 kV.

Have you explored Eimac's website thoroughly? I ask this because I know Eimac makes CW transmitting tetrodes capable of dissipating well over 1 megawatt. Hell, the graphite screens in these babies dissipate over 45 kW alone!

These bell-jar shaped valves are also servicable at the customer site, so if you lose the heater, for example, you can replace it right there on the spot. Consequently, provisions for vacuum pumps are made so that the valve can be re-evacuated following the service call (and continuously, for that matter, to compensate for outgassing of the valve's internal components at temperature).

You may wish to explore Eimac's website in somewhat more detail, as I seriously doubt the valve in your pic can handle anywhere near 100 kV. At 100 kV, anything that size - valve or not - may be at considerable risk from external flashover (and as it would be highly prudent to design for generous overvoltage conditions, your 100 kV spec is therefore a lower bound).

--E

This person has built 100kV and 120kV system and used this valve. Hence it is for this purpose.

I have looked the web link you have given and they do make high power valves but for not so high voltages. I do not need high currents and only high voltage is required. As Thales is no longer making these for public and Thomson-CSF no longer there, I need to find some one who can make these for me.

I need Minimum 65kV plate voltage and will be happy if 100kV plate voltage Tetrode is available. I need peak current of 100A for about 1us only and then Dc current 200mA in one case and only 10mA in another case.

I think 10mA DC continuous may make the Valve to survive for an year. I will be generating Plasma out of these valves with injected pulses 10us to 20us PW, 1Hz to 1000Hz repeat rates and limited currents.

You wrote: "I need peak current of 100A for about 1us only."

The TH5188 has a peak cathode current of 5A, with a max plate voltage of 85kV and max plate dissipation of around 1kW. It's big brother, the '5186, is a 10kW beastie with a peak cathode current of 10A, max plate voltage of 120kV, and a max plate dissipation of 10kW. Still not even close to 100A. What's a mother to do?

BTW, the valve in your pic is a very different design than the TH518x series I'm familiar with, and looks to be more like an Eimac tetrode valve commonly used in RF transmitters, industrial RF heating applications, and so forth. The TH518x series is specifically designed for pulse operation at very high voltages, and all the valves in the series are quite different than what's shown in your photo. If you've got the TH5188 in mind, you might wish to post a photo of it instead of this valve. I'd say just offhand that the valve shown in your pic has a typical plate voltage more in the 2kV to 3kV range.

You are hopefully planning to operate the valve in cooling oil? It certainly will reduce the potential for flashover, and is probably the only way you're going to eliminate flashover for device of this size at those voltages.

You still have the problem of switching that 100A current. Any ideas how you're gonna solve this? (BTW, that's why my original post asked if you planned to use a single device.)

--E

You may right about the valve 460, which was in the picture may have been used in transformer primary to switch lower voltage and to get greater voltage output.

Are these TH5188 and TH5186 available somewhere and what may be their cost? Where to look for these in stores? Some of my friends have worked on Plasma heating and they used such valves for RF generator in 14MeV Neutron sources. That was 30 year ago. I have never worked with these glass encapsulated devices but sure want to play with them now (very late starter on valves).

I have some applications for 65kV average current 10mA and Peak current 5A and I think This 5188 will be all right for that job. For another application 60kV 200mA average current and 80A peak current and 5186 may be good. Now how to get these valves is a big question.

I'll ask around. I haven't done anything with valves for years, but there are some folks in my building who may know where to get these or comparable valves. In the three-story basement of my building (the other 14 floors being above ground), they used to have a tokomak plasma fusion experiment. Not one of the real biggies you could walk around in, but still pretty formidable. I know they we're switching high currents at some hellacious voltages, but I don't know what they used to do it; whether they used one large device or a number of smaller devices connected in parallel. The Plasma Group is still around, but I don't know what they're working on these days. I'll ask around. No guarantees!

Meanwhile, I'd simply look on the Internet. If somebody sells these valves, they should be smart enough to have an Internet presence.

--E

You wrote: "For another application 60kV 200mA average current and 80A peak current and 5186 may be good."

One thing to keep in mind about valves is that once you reach a point called saturation, you cannot get any more current out of the valve. The cathode in many of the high power valves is made of thoriated tungsten. The geometry and area of the cathode fix the maximum possible current you're gonna get out of the cathode at a given cathode temperature. Without a plate voltage applied, the heated cathode will be surrounded by a so-called 'space charge' produced by energetic electrons "boiling off" the surface of the cathode. These fall back onto the cathode only to be replaced by others boiling off.

If the plate is then brought up to voltage (in the absence of a screen voltage) electrons will begin being pulled to the plate, resulting in a current flow. If you continue raising the plate voltage you will eventually suck all the electrons boiling off the cathode to the plate. As there aren't any more than all of them, you're limited to whatever that current happens to be. As the plate is getting them all, any further increase in plate voltage will not result in greater plate current. The valve is now in saturation. The only way you're gonna get more current is to increase the cathode temperature. Not advisable, and I strongly urge against this. Even the TH5186 with its higher peak current of 10A does not make it a more suitable device for this application since you will reach saturation long before you reach your target current of 80A. To get this current you will either have to use multiple valves in parallel, or find a larger valve (and probably far more expensive than the multiple-valve solution). You need to be aware of the hard limit saturation places on how much current you can ultimately expect to get out of a given device.

--E

This looks a correct picture of TH518X Tetrodes.

http://www.medelex.com/Tetrode_Tubes.htm

Yep. Now you're talkin'!

-E

Somehow the one on the left has always reminded me of a fat polka lady.

--E

Well, and where we can find her now? Looks to me that she may be some good show.

I keep her in my pocket for a rainy day.

If you didn't need such a high rep-rate (1000 Hz?), you might want to consider using some form of Marx generator for your really high-current plasma experiments. The folks at Sandia Labs use a Marx generator in their so-called "Z Machine." I believe it is the world's largest such generator (not that your needs are on a similar scale). The enormous electrical pulse of 50 trillion watts created by this generator strikes a complex target of wire filaments about the size of a spool of thread.

Diving board not included...

http://www.sandia.gov/media/images/Z02.gif

For my part, I've always had a special fondness for things that go "BOOM!" It's when they're not supposed to do that is when I begin to worry...

--E

Marx generator is next idea for me for a show but not for the current need of a highly controlled experiment. I do have something that is solid state with 200 MOSFETs of 1kV each and 30A peak current but I will like to look into valves as well.

How do I estimate the life of the valves and what is the price for such valves in any way? you are keeping lots of secrets so you can let out few if you wish.

As for prices, I'd have to do the same thing you'll have to do: look on the Internet for suppliers and get quotes. As I said before, it has been a long time since I did anything with valves. Therefore any pricing information I would have (which I do not) would necessarily be obsolete - nor do I have specs on these parts anymore. Otherwise I'd be happy to share the wealth!

--E

Do the tops of the valves in that pic look dirty to you? As though the valves had been sitting out in a barn for a long time? Hope you didn't find this pick on some vendor's website!

--E

what is that image?

That is a photo of the Z Machine. The bulk of it is underwater. The "spider web" appearance is due to electrical discharges spanning the surface of the tank as the machine dissipates its energy into the water following the experiment. The water acts as a giant load resistor of sorts. Each of those underwater cylindrical objects (each having a glowing section in the middle) are the energy-storage devices themselves and are about 36 feet (~11 m) long. The tank itself is about 100 feet (~30 m) across, more or less. Google "Z Machine" to learn more.

--E