>>From what I gather, loosing the grid from ground is not
>>something
> you want to do in the course of normal operation. As a
> fault condition
> it may or may not cause other things to happen. And that
> makes
> sense... the grid is your method of control, cut it loose
> and you
> no longer have control. It could float around near ground
> potential.
> It could head straight for plate potential.
Correct. In smaller low voltage tubes that are gas free, it
will go negative. That's how contact bias works.
In larger tubes, especially HV tubes, there is always some
residual gas and positive ions in operation so it tends to
go positive. If you read the text I posted
http://www.w8ji.com/images/Amplifier/Bias%20and%20grid%20fuse/Electronic-Communication-te.jpg
There is an italacized warning at the start that it works
only with high mu very small level tubes only. Several books
have simlar warnings or cautions. You have to read the
entire text.
> Here is the conclusion I'm coming too:
>
> Setting the grid adrift, if done in a system where there
> are other
> controls on plate overcurrent, doesn't mean certain doom.
Correct
> And there are some conditions where it might make sense,
> it seems to
> have a higher sensitivity to certain kinds of faults that
> otherwise
> require more sophisticated detection methods
> (speculation).
No, you have been caught up in a big deception. Someone has
been trying to hard sell the idea that all protection
circuits are bad because they run 50A through transistors.
If you search back through the threads it turn out that
claim, having been made for many years, is entirely false.
The person making that claim never even correctly read a
schematic.
Fuses and especially resistors are significantly less
reliable than electronic overload circuits for soft faults,
and hard faults should be handeled in the anode.
> And to me there is some engineering beauty to the idea of
> cutting off
> the element that is suffering... everything everyone has
> quoted so far
> seems to agree that loosing the grid will stop grid
> overcurrent.
> But you then no longer have a working system so you might
> as well
> shutdown HV and do a complete reset.
No. The point you miss is if there is a hard fault, cutting
the grid loose almost ceratinly does nothing. It simply
transfers the problem to the cathode system. The only way to
stop anode flashover is to remove anode voltage. It is
dangerous to the tube and other components when a grid under
hard fault is floated, and it is also very likely the grid
is well enough insulated to handle anode voltage without
breaking down.
> It's just another fault-control tool in the toolbox.
> So whether or not you want to kill faults using that tool
> or some
> combination of other tools is a design issue, not a moral
> one.
That's right. As long as you include HV fault protection
along with the grid fuse and can tolerate the risk of direct
HV blow-through to the cathode and everything tied to the
cathode, and still have some electronic means for reliable
soft fault protection like an electronic overload, a grid
fuse is perfectly OK.
If you don't include anode over-current and fault
protection, cathode overvoltage protection, and soft fault
protection a grid fuse isn't a good idea.
That's all pretty basic.
Of course if you have soft fault protection in the grid and
hard fault protection in the anode, the grid fuse isn't
needed at all and you don't need to worry about cathode
voltage as much. So if you want to add a grid fuse you
should really add a cathode clamp, and leave the soft faults
and HV fault protection in place. Then you have about the
same level of protection as with anode hard fault and grid
soft fault protection.
73 Tom
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|
