[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [TCML] Tube question



David,
Ralph has narrated a perfect explanation of what is likely occurring.
There are also impedance and capacitance values associated with 833A/C tubes that should be consistent, but may not be. The Chinese seem to have quality control problems. For example many of their 811A tubes must be upright to run. If you run them sideways with the filament hot the filament starts to "droop" into the grid. This never occurred with the RCA 811A tubes. So an inference can be made that their filaments are different and could easily behave differently.

David, this is most likely what you are seeing (i.e., Occams Razor ). There are other less likely explanations.

You might note that Ralph's explanation defines one advantage of running two 833A tubes in parallel. There are others too, but they are specific to coil design parameters. Two tubes may do nothing for an individual coil. It is all associated with design parameters. Impedance matching and other characteristics can all be calculated and used for optimal tube coil design. You might what to get ahold of the characteristic operating curves for an 833A or C from an old RCA manual. Life just isn't simple. Ask the folks at CERN.
John

John W. Gudenas, Ph.D.
Professor of Computer Science

On Sep 20, 2008, at 1:31 PM, <sparks@xxxxxxxxxxxxxx> wrote:

I remember Tony DeAngelis telling me that he bought a
new 833C tube and got only 16" sparks.  Then he bought
another 833C tube of a different brand and got 22" sparks.
He's not sure if the first tube was defective, or had
different characteristics, etc.

As an old transmitter engineer, let me pop in here from lurk mode, and
say , "Been there, done that, blown out the tubes!"

When using a vacuum tube for Tesla coil service, we want to get the
biggest sparks possible, and make the tube last long enough to get our
moneys worth from it.  Unlike commercial radio operation, where a tube
may be expected to deliver tens of thousands of hours of good service,
if we get a thousand hours out of a tube when it's driving a Tesla coil,
we are probably doing well.

Having said that, if the tube is operated well within the manufacturers ratings, there's no reason why the tube will not provide the same length of service as it would in commercial radio service. But, in an attempt
to get longer and brighter sparks from our coils, there is the almost
irresistable tendency to ramp up the power "Just a bit more," or add
more plate voltage, etc., and we find ourselves ultimately operating the
tube far in excess of one or more of it's maximum limits.

Tubes, unlike transistors, can usually handle short-term overloads
rather gracefully.  Heavy anodes tend to average out the heat pulses
from high current staccato operation, so everything looks fairly normal.
There's a good likelihood, however, that we are far exceeding the
maximum plate current rating of the tube during the "ON" time of the
pulse - and possible exceeding the maximum allowable grid current, too.

The 833 series of vacuum tubes use a directly heated filament for the
cathode.  That type cathode has far lower peak electron emission than
does an indirectly heated cathode.  That is to say, the tube is
"emission limited" in that the cathode cannot supply any more than a
certain number of electrons, no matter how high we raise the plate
voltage.

Using a filament cathode tube in pulse service far above its normal
cathode current ratings is a good way to destroy the tube by cathode
disintegration. It also quickly shows you which tubes cathodes have the
greatest amount of reserve electron emission as they will give you the
greatest spark lengths. Because different tube manufacturers process
their tube cathode assemblies in different ways, it is to be expected
that there will be differences in performance when the tube is operated
at the absolute limit of its capabilities.

So, if a new tube gives you a shorter spark length, you may reasonably
conclude that the problem is that the tubes cathode cannot emit enough
electrons to satisfy the demand that the higher than normal plate
voltage requires.  All that raising the plate voltage will do is to
accelerate the available electrons faster, so that when the hit the
anode, they give up their additional kinetic energy and cause the anode
to get hotter than normal.  While it looks as though we should see
longer sparks, the additional plate voltage does not cause any
additional current to flow through the tube, so the spark length does
not change, it just makes the plate (anode) get hotter.

There are other tubes available that will handle high current pulses far
better than will the 833 tubes.  Most tubes that have an indirectly
heated sleeve cathode, such as are used in hydrogen thyratrons and other
tubes that require large amounts of electron emission are capable of
high peak plate currents.  Another way to get a "good" tube for Tesla
coil service is to use a large directly heated filament tube, such as a
7C24, which can handle about 50 amps peak plate current.

Long sparks to all,

Ralph   W5JGV - WD2XSH/7

_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla

John W. Gudenas, Ph.D.
Professor of Computer Science


_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla