[Amps] ALPHA 77 SX UPDATE

[2]

>2 wrote:
>>
>>
>>>2 wrote:
>>>>
>>>>Another consideration is that 2, 8877s have 0.2pF of feedback-C.  Since
>>>>the 8877 is capable of oscillating) well into the uhf range, this
>>>>amount of feedback-C is nothing to sneeze at.
>>>
>>>Two tubes also have double the input capacitance and half the input
>>>impedance, so the feedback situation remains the same.
>>
>>**  Ian -- How do you explain that 2, 3-500Z amplifiers have a greater
>>reputation for squirreliness than 1, 3-500Z amplifiers?
>
>"Reputation for squirreliness"? Right, so you're looking for a real 
>in-depth technical analysis here...

Squirreliness = marginal VHF/UHF stability.
>
>If we've learned one thing on this list, it is that there's a huge range 
>of variability between different amp designs using the same tubes... and 
>even between different individual amps.
>
Agreed, however, more feedback-C is hardly good news because it is built 
into the tube and it is not adjustable.

>But if forced to give a general answer, I'd guess it is probably because 
>multiple-tube amps can have more oscillation modes than single-tube, and 
>also because lead lengths have to be longer.
>
Multi-tube amps also have multiples of the feedback-C for a single tube.  
Without feedback-C, oscillation would not be a problem.  
>>>
>>>> [intermittent vhf/uhf parasites have a reputation for causing
>>>>intermittent hv-to-gnd arcs, so lower vhf-Q suppressors might not be a
>>>>bad idea]
>>>>
>>>If the grid is nailed down firmly - VHF-style, using multiple contact
>>>fingers on the grid ring - then it's questionable whether any VHF
>>>parasitic suppressors are needed in a single-8877 amp for HF.
>>
>>**  So why did the gold-plating on the 8877's grid boil off in Figure 24
>>on my Web site ?
>>
>I have no idea. The evidence shows only that the gold *did* melt 

**  true, and it shows that the top layer of gold atoms boiled 
off/evaporated and condensed into gold melt-balls.

>- but 
>there is not enough evidence to prove *why*.

**  Eimac's W. B. Foote stated that the 8877 development team determined 
that gold occasionally boils off during an "oscillation condition".   
Eimac's Reid Brandon later told *QST* that Mr. Foote was "not authorized" 
to release this information to me.  
>
>
>>>After all,
>>>the same tube is quite stable in VHF amps, where the input and output
>>>are both *deliberately* tuned to the same frequency!
>>
>>In a VHF amplifier, there is a VHF load.  In a MF/HF amplifier, a VHF
>>load is not present in case the amplifier tries to oscillate at the
>>anode's VHF-resonance.
>>
>Nope - if the grid is well grounded, a typical VHF GG amplifier is 
>stable with *no* input or output load, or with any combination of 
>input/output load and phase.
>
**  Does a well-grounded 8877 grid have a self-resonant frequency?  
>
>>>The same of course
>>>applies to the 3CX800 and similar ceramic tubes with a full grid ring.
>>>
>>>However, with two tubes in parallel there is also the possibility of
>>>push-pull VHF resonances, so parasitic suppressors would probably be a
>>>good idea to kill this mode.
>>
>>**  However, VHF suppressors will melt down at VHF.
>
>Say what?
>
**  By design, above 30MHz, VHF suppressors become progressively lossy.  
If a VHF suppressor does not get perditionly hot at VHF, it is not 
performing its job. 

-  R. L. Measures, a.k.a. Rich..., 805.386.3734,AG6K, 
www.vcnet.com/measures.  
end