> Hmm. Okay. I have used amps (all GG) with these kinds of
> tubes and
> sometimes they glow more than other times. Why is this?
> It seems like it
> has happened when one of these conditions exist:
>
> 1. out of resonance
> 2. overdriven
> 3. bad match
Hi Mark,
The only significant thing that makes the anode glow red is
power dissipation in the anode. That is purely a function of
the DIFFERENCE between plate input power and RF output power
at the tube anode. A reasonable estimate is found by taking
the anode dc power input and comparing it to the RF output
power. This would be under a long term average of power
conditions, not peak power or peak envelope power. This is
because the thermal inertia of the anode averages things
over a few dozen seconds or longer.
If an amp is averaging 1000 watts dc input and averaging 600
watts RF output, the anode dissipation would be 400 watts so
far as long term heating and color goes.
The problem is most meters won't integrate the power over
long (20-30 second) periods, so we often have to use carrier
conditions and know the real dissipation is less.
This is actually the basis for IVS, intermittent voice
service, ratings. The low duty cycle allows tubes with
larger grids (non-plated, like the 3-500Z) or anodes to be
abused without harm. This is why 811A's with 60-65 watts
dissipation can output 200 watts with reasonable life if
operated voice or CW, and why tubes like the 3CX1200 can run
10 or more kW pulse output power. The average power is low,
the elements are large and have thermal inertia, and so the
heating is significantly less than the same dissipation with
longer duty cycles.
FET's can work this way also. I've used small FET's like
MRF150's in pulse applications where two 150W devices ran
over 1200 watts output.
The big Ham myth is that SWR affects heating or life of the
tube. It does not affect a transistor or a tube, unless the
tank circuit is fixed tuned or adjusted out of range. All
our solid state radios, unless they have built in
auto-tuners, require close to a 50 ohm load or linearity and
efficiency can suffer. Tubes are the same, except the
adjustable tank eliminates the need for the tuner.
The only components in the amp that are affected by load SWR
are the loading cap and things from that point downstream to
the antenna. Inductor current can also change, but it is
generally a more minor change. The tube and the tuning cap
never see the SWR **if the tank is properly tuned to the new
impedance**.
Overdrive mainly affects the control grid. The amount of
loading greatly affects the amount of overdrive, because the
tank tuning establishes the load line of the tube. If I
mistune an amplifier with excessively light loading ( too
much capacitance for the power) the loadline is much too
high an impedance. This increases grid current and tank
voltage (tank voltage can be several times dc voltage) and
destroys linearity. It's very tough on components, although
anode heating is much less.
If we load the PA too heavy linearity generally stays good,
but the load impedance on the tube is too low. This
increases dissipation in the anode, but reduces stress on
other components.
Watching the grid current while talking on voice, unless you
have a peak reading grid meter, is almost useless. We really
need to apply a quick carrier with full drive and look at
the grid meter. A 3-500Z is typically approaching rated grid
dissipation at 150mA during a steady carrier, although that
varies with the particular operating conditions. The 3-500Z
is not sensitive at all to exceeding the grid dissipation
for short periods because the grid is a material that can
withstand a very high temperature, almost incandescence.
On the other hand a tube like the 3CX800 or 3CX1500 can be
ruined with only seconds of excessive grid current, and the
damage accumulates over time even if it isn't ruined
instantly. That's why any tube like the 3CX800 or 8877,
using a metal oxide cathode, requires a very fast electronic
grid overload trip circuit. I'd personally never own or
build an amp with a MOx cathode without a positive acting
electronic overload, not having one is just plain stupid.
I've got three Ten Tec amps here now with blown 3CX800's
from excessive grid current during operation. One amp owner
had an antenna fail, and before he saw the LED flashing he
cooked the grids. That's an 800 dollar problem that would be
easy to repeat over and over caused by lack of a $1 op amp.
73 Tom
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|