[Amps] AL-1200 Intermittent HV Short

[R.Measures]

>Three times in three weeks, when I turn the AL-1200 on, there is a flash 
>and a bang from the RF compartment. This happens immediately at power up. 
>Once it blew the diode that connects the b- to chassis (I replaced it with 
>a 5408).  Once it blew the 15 amp main fuse. The other time, it did no harm.

•  Ed -- Some possibilities:  1. fruit flies.  2.  baby spiders.  3. tiny 
leak in coffee cup that occasionally sits on top of amplifier.  4.  cigar 
ashes.  5.  intermittent VHF parasitic oscillation.  If it seems that 5. 
could be the case due to process of elimination, it's probably due to the 
fact that the 3CX1200A7 has c. 3x as much feedback-C as its successor, 
the 3CX1200Z7.  Although feedback-C is helpful in a VHF-oscillator, it is 
somewhat less than helpful in an amplifier.  Unfortunately, the ._ Z7 
requires a different socket and a different filament potential (6.3v), so 
swapping over is hardly 'plug-n-play'.  An easier $olution may be to 
reduce the parallel-equivalent-resistance of the VHF suppressor.  
http://www.somis.org/Rp_comp.html
- note - Rp is the parallel-equivalent resistance of the VHF parasitic 
suppressor. Rp is significant because VHF voltage gain is basicly equal 
to (Rp x Mu).  Thus, more VHF-Rp means more VHF amplification in an HF 
amplifier. Less VHF-Rp means less VHF amplification in a HF amplifier.
>
>Each time, the amp works fine after I shut it off and let my pulse slow 
>down. 

•  The good news is that 3CX1200A7s are less prone to a filament to grid 
short during a parasite than are 3-500Zs.  
>
>I can't find the point of the arc.  

•  I have never been able to find an arc-mark after what appeared to have 
been a parasitic event.  My guess is that such an arc is too brief to 
melt aluminum.  One way to test for a parasitic visitation is to measure 
the subsequent DC-ohms of the parasitic suppressor.  If the ohms are way 
too high - like more than 50% out of tolerance, a parasite is suspect, 
especially if the resistor body shows little signs of overload.  
- the logic -  Since the typical suppressor is shunted by 0.001-0.01 DC 
ohms, and c. 50nH, neither DC nor HF energy could be likely to damage a 
typical suppressor resistor.   

>I looked all over the place, inside 
>the cover, at the chimney clips, etc etc. I can't find anything resembling 
>pitting or carbon. And since I see a flash, I assume it can't be tube 
>related.

•  Indeed.  One popular theory is that the big bang is made by the tube 
arcing due to 'barnacles', however, any schlub who has ever tested a 
vacuum cap. knows that this theory  is a sea-story since an arc in a 
vacuum hardly makes a sound.  
>
>The amp worked without incident for 3 month or so. Now this. What has 
>changed. The "only thing" is that I changed the chimney from a 3-500 style 
>chimney the correct 3cx1200 glass chimney.  No, I didn't bend the chimney 
>clips. 

•  Does you amp have a "neutralization" circuit?
>
>(Incidentally, I changed the chimney because it seemed to me that lots of 
>air bypasses the cooling fins with the larger chimney.)
>
>Ameritron says only thing to do is 'send it back', but since it is 
>intermittent, I'm afraid it will come back 'no trouble found' or 'no 
>trouble found, damaged by UPS on the way back."  

•  UPS Brown Label damage is also intermittent -  unless the package can 
withstand a 30-foot free-fall onto a concrete slab.  --  "UPS" is 
correctly pronounced: OOPS!

>Has anyone seen this one before, or have any hints?     
>I'm hoping to nail this before it kills something major like the operator, 
>or even worse, the transformer.
>
>Thanks for entertaining a vague, intermittent question.
>
>Ed N1TS
>
>"Life is hard, so why shouldn't amps be?"
>_______________________________________________
>Amps mailing list
>Amps at contesting.com
>http://lists.contesting.com/mailman/listinfo/amps
>