> Well... the other issue of course is _COST_. If a
> reliable
> system is obtainable without investing in a sophisticated
> circuit
> that's a good reason to stick with the simpler method.
As long as it works the same.
> As for "hard faults should be handled in the anode",
> I parse that statement as being a design philosophy.
It is a recommended design standard that appears many
places. Not a philosophy.
> (by Hard Fault do you mean arc? I think so.)
> Well here we get to a minor point of contention.
>
> The textbooks we've been examining, with your generous
> help,
> don't cover what happens when a grid under arc is floated.
> So... we don't have good data on what happens, except for
> the experience of folks who say things work out ok.
Anyone who has taken basic physics or worked with
electronics in any technical depth should know what an arc
does. Pipeline welders even understand arcs.
Once an arc starts, the sustaining voltage becomes very low.
This means to stop a grid from arcing to the anode the grid
has to rise to nearly the anode potential.
Now, we have Will claiming the grid goes negative. If the
grid went negative, the arc would get worse...not better. We
need somehow to bring the grid up positive to anode
potential to reduce current and quench the arc. In order for
the grid to go to anode potential, the grid has to move
significantly positive with respect to the chassis (and
cathode). Once the grid is even remotely far enough positive
to reduce the arc, it is so far positive it almost certainly
has exceeded the breakdown voltage of the grid to the
cathode. Now we have a grid, arcing to an anode, that is
also arcing to the cathode or at least sucking all the space
charge away from the cathode. Once the cathode is exposed to
positive ions, it instantly suffers some amount of damage.
The entire idea of pulling the grid off ground to reduce
grid-to-anode fault current is, quite frankly, very poor
design.
> If it were dangerous to the tube.. I would think the
> experience
> stories would be different than what has been reported
> here.
Why? Why do you think a Ham staring at an amplifier panel
would understand the statistics of making a mod over a large
scale field sample?
We did statistical analysis of exciter failures caused by PA
tube faults, and the number of incidents of reported
destructive damage to exciters was higher in triode amps
with grids that floated. We analyzed a few hundred failures
known to be arcs, and there were zero exciter failures with
hard grounded grids. There were about a half dozen with
grids tied to ground through resistors.
We could actually simulate that failure in the lab.
> Or, maybe the position of the pro-fused-grid bunch would
> be
> that arcing is less likely with a fused grid altogether...
> I'm a little shakey on this point.
So are they, and they are shaky on several other points as
well.
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|