At 08:52 PM 12/12/97 -0800, you wrote:
>>At 08:36 AM 12/12/97 -0800, you wrote:
>>
>>>There are no smallish wires in the the grid of an 8877. There are 100+
>>>rectangular bars. At Eimac's maximum ratings of 4000v and 250MHz, the
>>>grid of an 8877 carries over 40 RMS amperes. // There are no small
>>>wires in the 8877 cathode element.
Small is a very general term...compared to what? With > 4 Joules @ 50 volts
flowing, small Eimac tube wires are small enough to be zapped, according to
Eimac. "Thoriated tungsten filaments are capable of withstanding higher energy
arcs than are the oxide cathode types. However, some of the smaller tube
types such as the 4CX1500A, 3-500Z, and 5CX1500A as well as some high-mu
triodes such as the 3CX3000A7, 3CX10,000A7, or 3CX20,000A7, while employing
thoriated tungsten emitters, have very fine wire grid and cathode structures
and should be protected in much the same way as the smaller oxide cathode
tubes." So you see they are calling the grid and cathode structures in a
3CX20,000A7 "small".....that's why I asked...compared to what? It is not
as if Eimac is trying to sell us a magic black box for $500; they are
trying to get us to invest $1.50 for a resistor to protect our investment
in tubes and external components because they do not warranty arcs. What
would be their possible motive in producing a service bulletin not backed
up by lab tests? Maybe they are trying to increase their Ohmite stock
value in their 401K plan.
>
>This is starting to sound like guess work, Phil, not anything based on
>scientific measurements. - - We have a tube that has a 25
>watt-continuous rated grid which is designed to carry 40A of grid current
Eimac states that a properly designed glitch resistor will limit the peak
current during a fault to 40 amps. This is why your grid dissipation
analogy doesn't wash. IMO, the current available from an unprotected typical
capacitor input filtered power supply is as much as 200 amps! Certainly
well over 100 amps even in a very modest amateur table top unit.
What is the energy delivered by a 4 watt-second charge in 5us?
> However, I have never seen gas vanish in an 8877, so if one has an
>anode/grid arc, one need not worry about damaging the "delicate grid
>wires" - because the tube is KAPUT.
If the energy in the arc is less than 4 Joules, the tube will not be
kaput, according to Eimac. It sounds like guess work because you are
reading my "sound bites" on a long, concise technical paper. You need
to read the whole thing to get the picture.
However folks report hearing a
>'big bang' from an alleged arc inside a near vacuum, and lots of current
>seems to be flowing in the cathode because the cathode bias supply diode
>often shorts, which hardly seems likely because the grid is directly
>grounded.
The electrons are emanating from the negative terminal, plus space charge
neutralization. In all my amp designs, the Zener is right in this path.
Do not get locked up on the grid; the cathode is just as much of a target.
>
>>I am sure some of the smaller glass tubes are closer to disaster at the
>>4 Joule limit than an 8877, ... ... ...
>
>Maybe not. Mr. Rauch and I have run a 3-500Z grid so hot it glowed, yet
>no damage was done.
I say again; run 10 Joules or so through there and report back.
>How would you calculate the size of an 8877 glitch resistor using a 35uF
>filter, operating at 4000v?
The 4000 volts and 35 mfd are the easy part. The follow-on current from
the power supply is the variable, so every case will be different. That
is why Eimac devised the copper wire test so that the average builder
can arrive at the proper value resistor without calculations or test
equipment. They state that 4 Joules of energy (to give us an idea of
what 4 Joules looks like) is stored in a 2 mfd capacitor in a 2000 volt
power supply. So even it would need a glitch resistor to absorb the
energy from the follow-on current.
What I am wondering now, is if the glitch resistor will also limit the
current in a VHF parasitic arc to 4 Joules, and damage nothing?....Hmmmmmm
"Twilight Zone" theme; fade to black........
(((73)))
Phil, K5PC
--
FAQ on WWW: http://www.contesting.com/ampfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|