Page 1 of 2

vacuum tubes for HVDC?

Posted: Wed Jan 20, 2010 7:16 pm
by kunkmiester
This thought just occurred to me--feasability is a different matter.

People have mentioned some solid state methods for taking the 1.2 million volts or so and getting it down to a reasonable voltage for transmission, or distribution if using HVDC transmission. I thought of a vacuum tube, and wondered if it would be feasible.

Some of the disadvantages of vacuum tubes aren't present--they'd likely be very large, which allows them to be built very durably and be cooled easily, and with a fixed installation, I'm seeing metal tubes with vacuum pumps--you'd actually be able to take them apart and repair them, rather than replace them.

Is there a vacuum tube circuit that would work for voltage step-down, and given the above, would it be any more feasible than the other ideas around here?

IIRC...

Posted: Thu Jan 21, 2010 3:50 pm
by Nik
IIRC, modern MOSFETs' industrial kin are *better* than vacuum tubes except for self-resonant applications.

eg Klystron in microwave oven...

Uh, IIRC, a bunch of people are re-creating 'the valve sound' by making MOSFET assemblies that will 'plug'n'play'...

Now, if you were working in orbit or on Moon where ultra-high vacuum is *free*, then coolant-jacketted vacuum tubes would be a great idea...

( IIRC, LEO needs a wake-shield to be sure, to be sure... ;-))

Posted: Thu Jan 21, 2010 4:19 pm
by kunkmiester
A polywell plant would already have a large vacuum pump capacity, the extra needed for the tubes I can't see costing a whole lot more, making it pretty cheap.

How big are mosfets for 1.2 MV?

Posted: Thu Jan 21, 2010 4:37 pm
by BenTC
wikipedia wrote: Vacuum tubes require a large temperature difference between the hot cathode and the cold anode. Because of this, vacuum tubes are inherently power-inefficient.
...
For most purposes, the vacuum tube has been replaced by solid-state devices such as transistors and solid-state diodes. Solid-state devices last much longer, are smaller, more efficient, more reliable, and cheaper than equivalent vacuum tube devices.

Posted: Thu Jan 21, 2010 7:31 pm
by MSimon
BenTC wrote:
wikipedia wrote: Vacuum tubes require a large temperature difference between the hot cathode and the cold anode. Because of this, vacuum tubes are inherently power-inefficient.
...
For most purposes, the vacuum tube has been replaced by solid-state devices such as transistors and solid-state diodes. Solid-state devices last much longer, are smaller, more efficient, more reliable, and cheaper than equivalent vacuum tube devices.
That is not strictly true. The temperature is strictly a function of the method of electron emission. Otherwise the rest pretty much holds.

Vacuum and ''vacuum"

Posted: Fri Jan 22, 2010 7:27 pm
by Nik
Uh, I'd be careful using the Polywell's pump for vacuum tubes. IIRC, the 'shiny', metallised patch inside a 'tube is where the commercially pumped vacuum has been 'gettered' clean.

Hence my comment about the wake-shield required for LEO...

FWIW, I read somewhere that each Apollo landing's exhaust & out-gassing temporarily doubled the Moon's 'atmosphere'...

Posted: Sat Jan 23, 2010 3:41 am
by kunkmiester
Small glass tubes had a small amount of a special compound put inside, and then after sealing, heating this compound would remove oxygen and the other volatile chemicals in the air, as well as creating vacuum. It worked well enough, and was easier than trying to pump them down then seal them.

With a vacuum pump available, "gettering" the tubes isn't needed--the vacuum and operating conditions are created by pumping them down, rather than chemically nullifying the air.

The polywell will need a massive amount of vacuum pumping to operate, adding a few more horsepower and making use of the vacuum for other purposes shouldn't be that big of a deal. I can't see why sharing the pump system would be a big deal.

Out of my depth vs vacuum tubes...

Posted: Sat Jan 23, 2010 3:42 pm
by Nik
Okay, I'm waaay out of my depth.


Care: Your mileage may vary...
http://en.wikipedia.org/wiki/Vacuum_tube

poking around the links found a 'rule of thumb' of one (1) micro-torr as requirement for a 'good tube', plus a warning that this is difficult by pumping due out-gassing etc.
http://www.vacuumtubes.net/How_Vacuum_Tubes_Work.htm

===
FWIW, I googled into...
http://www.sparkbangbuzz.com/
& pre 2007 stuff at...
http://home.earthlink.net/~lenyr/index.html
About half-way down page is 'home evacuated vacuum tube'...

and, slightly OT as no vacuum tubes yet, this fascinating page...
http://home.earthlink.net/~jimlux/hv/hvmain.htm
However, it *does* reference...
http://home.earthlink.net/~jimlux/hv/pasley1.htm
... which, incidentally, lists HV switching devices that are restricted by dual use guidelines. Take note lest MIB query your aspirations...

plus, some cautionary notes on the X-ray hazards of HV...
http://www.celnav.de/hv/hvindex.htm
http://www.celnav.de/hv/hv2.htm

Posted: Sun Jan 24, 2010 3:27 pm
by BenTC
Some interesting stuff in there Nik. I'll file for later reference.
Here is some similar stuff: http://tesladownunder.com/

Posted: Sun Jan 24, 2010 7:22 pm
by chrismb
I don't understand what is being asked here.

HVDC has been around for decades, but they used to use mercury vapour switches to control the HVDC and down-convert it, but mercury vapour valves aren't particularly reliable in this service. It's all done by silicon now.

Posted: Sun Jan 24, 2010 7:26 pm
by MSimon
chrismb wrote:I don't understand what is being asked here.

HVDC has been around for decades, but they used to use mercury vapour switches to control the HVDC and down-convert it, but mercury vapour valves aren't particularly reliable in this service. It's all done by silicon now.
In fact light triggered silicon.

Posted: Sun Jan 24, 2010 11:24 pm
by kunkmiester
I don't understand what is being asked here.
Well, with industrial solid state available at the voltages and such needed, it's somewhat of a moot point. A metal vacuum tube isn't that hard to do though, it will just have a much shorter life than a solid state setup. Getting vacuum isn't hard, since you already have a massive vacuum pump, and making it a bit bigger doesn't seem like that big of a deal.

Most silicon circuits were designed from vacuum tube circuits--triodes to transistors, etc. So, why couldn't the step-down circuits be made with vacuum tubes instead of silicon? That's really the question.

The answer thus far is basically economics--silicon circuits will last long enough, and are available cheap enough, it'd be ridiculous to make the step-down circuit with vacuum tubes. Now, if for some reason you loose the ability to make silicon, it might work, but for now, it's just fun speculation.

Posted: Mon Jan 25, 2010 1:42 pm
by MSimon
Most silicon circuits were designed from vacuum tube circuits--triodes to transistors, etc. So, why couldn't the step-down circuits be made with vacuum tubes instead of silicon? That's really the question.
The voltage drop across the tubes is higher. And the voltage drop across the tubes is relatively constant. It is the current that changes.

So think of it: A 1.2 MV tube is going to have a 1.2 MV drop.

Gas filled tubes do better (thyratrons).

A shame that tube theory is no longer taught. Look up - GM vacuum tubes - for some hints.

Us old farts have one or two advantages over you whipper snappers ;-)

Glad to be of service. For as long as I last.

And - even in gas tubes the voltage drop is going to run on the order of 50 volts or so in the best case. For semiconductors in the saturation region the voltage is going to run from 1 to 5 volts depending on current. For very low currents in MOSFETS the voltage is a linear function of current. They act like pure resistors in that region.

For thyristors the voltage is going to follow a diode function. i.e. about .1 V increase for every factor of 10 increase in current until you get near the device maximum rating. And the .1V/10X is also a function of temperature.

Well that is probably more than you wanted to know.

Posted: Mon Jan 25, 2010 1:55 pm
by MSimon
Now it is true that tubes are constant current devices. And the voltage on the plate can go relatively low and still maintain current. But there is a minimum voltage that will support a given current flow. And that is higher for tubes than semiconductors. One of the reasons is the higher field (millions of volts per m ) vs tubes.

http://ecee.colorado.edu/~bart/book/boo ... /ch3_3.htm

See figure 3.3.7

Posted: Tue Jan 26, 2010 6:54 pm
by Roger
MSimon wrote: A shame that tube theory is no longer taught.
FYI I bought a Fender tube amp for my guitar, the tube amps are coming back because of the quality (warmth, smokeyness of tone) sound generated. I can purchase a variety of tubes to tweak the sound of my amp. This guy is awesome, known for his quality modifications.

http://billmaudio.com/wp/?page_id=2

I hope to get his mods, they are the shiznit, the guitarist from Blondie turned me on to this Fender amp about 9 yrs ago, and now this mods guy.

http://www.facebook.com/profile.php?suc ... =616526389

Somebody still designs and makes tubes, albeit for a super niche market.