[Amps] Parasitic Oscillations in RF Amplifiers

[rlm]

>> Note:  Years ago, I had an amplifier which had a burned bandswitch.  I
>fixed
>> the bandswitch and fired it up.  Same problem.  I then decided it was
>> parasitics, so applied all of Rich Measures "cures".  I then fired up the
>> amplifier and had exactly the same fireworks.
>
>I've found the same thing in testing, but the cause was original components.
>WB8BFS reported the same thing here with a TL922, Jack's had a relay
>sequencing problem. A K5 had a gas-tube lighting arrestor in the feedline
>that was intermittent. Both had installed "the kits" before discovering the
>cause.
>
>There are literally dozens of causes of arcs and sparks.
>
>> I do have objections to the approach that almost all amplifier problems
>are
>> parasitic related.  No doubt parasitics can cause some problems.  Indeed,
>I
>> understand some of the tubes replacing the Eimac 3-500Z may have higher
>> gain, causing a marginal amplifier to have parasitics.
>
>The interesting thing is the tubes have less gain, not more, in most cases.
>Not that grounded-grid gain has much to do with VHF grounding of the grids
>and resulting stability.

**  A dipmeter shows that a directly-grounded 3-500Z grid is not grounded 
throughout the VHF range.
-  Is there an amplifier owner out there who has a dipmeter and a SB-1000 
or AL-80_ who would be willing to measure the dip-freq of the 3-500Z's 
grid?
>
>What seems to have become lost is the fact that slight changes in HV have
>more effect on gain than differences from tube-to-tube, 

**  I have seen 5v-fil tubes (3-500Z) which were operated at c. 5.8v 
(Henry 1KD-5) that had about as much gain as a cathode-driven 304TL -- 
i.e., c. x4 instead of the usual x 12 - 14.  

>and that
>instability is a complex function of many things the smallest and least
>critical of which is tube-to-tube gain variations.

** So why do only about 60% of TL-922s cook their parasite suppressors 
and burn their bandswitches at the points where a VHF signal eminating 
from the anodes would be the most likely to arc across open contacts?  
(the 10m, 15m L contacts and the 160m, 80m Tune-C padder contacts)
http://WWW.VCNET.COM/measures/bandswitch.922.JPEG
>
>> It is difficult for me to understand how an amplifier that has been
>working
>> faithfully for 40 years suddenly becomes a poor design because there is a
>> failure.  And, it is also difficult to understand that no matter what the
>> symptoms, the cause and cure is always the same.
>
>Any very simple diagnosis or theory, even if wrong, makes many people
>comfortable and satisfies them. This makes it appear that anyone with an
>antalytical approach is just "causing problems" except to other who think
>the same way. It's a lot like arguing religion.

**  Somewhat.  However, organized religion is unique in that science is 
seen as the enemy.  Also, with organized religion, murder and even 
genocide is used a bit more often to settle theory squabbles than they 
are use among the denizens amateur radio. (N5PS was an exception)
 
 - "Men never do evil so completely and cheerfully as when they do it 
from religious conviction." -   Blaise Pascal (1623-1662) 
>
>> Frankly, I enjoy thinking and solving problems.  If all there was to working
>> on amplifiers was to install a special parasitic suppressor, I would
>> probably lose interest in repairing them.
>
>I like to look at the positive side. I could buy measurement equipment and
>textbooks cheap, because few would have any use for them.
>
**  It might be beneficial for you to take a course in AC circuit 
analysis, Mr. Rauch.

>Enough of this, why not make some measurements Colin?
>
>73 Tom

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