[AMPS] smoked TL922

[Rich Measures]

...snip...
Mr. Erhorn:  The following seems to be designated as something you said, 
however, I said it.  

>I would solder the fast-acting 0.25A 250v fuses between the grid pins and 
>chassis ground.  .  
>
>>Seems you were bitchin' about a similar suggestion last week. 
>
>The suggestion of mounting grid fuses on the rear panel was not mine.
>
>>In any cases fuses should be in some easily and quickly accessible location. 
>>Personally, I prefer the chokes since as current absorbers of minor arcs
>>from gas, etc they will not blow or collapse. Once the event has passed
>>the amp works as before. 
>>
>It might be interesting to high-pot such a tube immediately after the 
>suspected gas arc, just to see whether the tube is gassy or not -- unless 
>you subscribe to the rauchian vanishing gas theorum.  .  
>
>It definitely is true that in the vast majority of cases, after one or two 
>BANGs early in life, the tube continues on for a normal lifetime of normal 
>performance, just as if nothing had happened.  If the cause wasn't gas or 
>some other anomaly eliminated by the arc - if in fact it was parasitic in 
>nature - what killed the parasitic during the BANG? 
Since there is virtually no load on the anode during a vhf parasitic, an 
anode circuit arc to ground often occurs, during which the anode 
potential decreases to a low level, which stops the oscillation.

>Especially if 
>parasitic suppressor R increased as a result?
Calculating the effect of increasing Rs involves susceptance and 
admittance calculations, which, for yours truly at least, is slow going.  
>
>Also, it's much more common in our experience for a new tube to BANG when 
>in standby with full cutoff bias applied, rather than when running RF 
>keyed or key-down.  How to explain that?
Beats me.  I'm no amplifier 'expert'.  Murphy said nothing is as easy as 
it looks.  
>
>>>>In extreme cases when the tube arcs or shorts you will also take out the
>>>>Zener D-2, and Bias diode D-1 and caps C-3 and C-26.
>>>
>>>If a gassy tube arced between the anode and the grounded grid, how could 
>>>current flow in the cathode bias Zener?  
>>>-  With zero volts bias between the grid and the cathode, the tubes draw 
>>>around 300mA of cathode current.  I don't see how 300mA could blow the 
>>>cathode bias Zener, Carl.  
>>
>>Which goes first...the grid choke or the zener?  
>
>There is typically a big bang, and subsequently the blown choke and the 
>shorted Zener are discovered during the repair process.
>
>>And if C3, C26 and D-1
>>also blow would that not assume a rather high voltage thru that path?
>
>Agreed, Carl.  However the current path of an anode to grounded grid arc 
>is not through components in the cathode.  
>
>How about the fault current return path from ground back to HV neg? I 
>don't know the particular circuit you're discussing, but if HV supply 
>negative isn't bolted to ground (i.e., if the bias supply or zener is 
>between HV- and ground), the natural fault current return path is from 
>ground via bias supply (zener or whatever) back to HV neg. 

-   The fault current from the grid end of the alleged arc is to the 
chassis, through the grid current meter shunt resistor (which should be 
glitch diode protected), through the anode current meter shunt 
resistor(which should be glitch diode protected), and back to the 
negative HV  .  The bias zener is not in this path.  

>A 200 or 400 
>surge-amp shunt diode (big enough to survive several discharge cycles of 
>the HV filter cap through the limiting R) is typically used to protect the 
>bias supply. But you knew that.
I do not know that because I know that the discharge path is not through 
the bias zener. The glitch diodes protect the current shunts and the 
current meters.   
Rich...

R. L. Measures, 805-386-3734, AG6K   


--
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