[AMPS] 8877 Parasitic Suppression Requirements

Thu, 4 Sep 1997 00:38:19 +0000

> To:            amps@contesting.com

> It is my understanding that the 8877 is such an excellently
> stable tube that it does not require any form of parasitic
> suppressor... when used by itself.....(certainly my AL-1500
> doesn't employ any parasitic suppressor) however, I am interested
> in that event in which there are two of them in parallel.  Does
> that situation require a parasitic suppressor for each of them
> to keep the "loop" between the two plates from forming a VHF or
> UHF "tuned plate" situation?  Please color me curious.

Hi Rod,

The most important thing you can do to insure stability is to use a 
very low impedance grid lead, and be sure the anode does not resonate 
on the same frequency where the grid has a high impedance.

It's easy to measure the grids impedance by doing a measurement of 
transmission loss through the tube, from anode to cathode. Any 
sharp increase in power through the tube indicates the grid is 
offering a high impedance, and that sets you up for having a 
tuned-plate tuned-grid oscillator!

Let me give you the values for two tubes, the 8877 and the 3-500Z, so 
you can get an idea why they behave so differently. Here's the 
transmission loss of an 8877 in a standard Johnson socket with 
four spring clips to ground the grid ring, I'll only give you the 
PEAK feedthrough frequencies and the dB loss. The lower the loss, the 
more likely instability.

108 MHz -40 dB
166 MHz -29 dB
217 MHz -24 dB
315 MHz -20 dB
430 Mhz -23 dB
532 MHz -25 dB
955 MHz -27 dB

Now a 3-500Z in a shielded socket with zero grid lead length:

80 MHz -36 dB
222 MHz -21 dB (broad peak from 160 to 240 MHz)
461 MHz -21 dB
899 MHz -11 dB

You can see the 8877 has much better grid grounding than the 3-500Z.
The 3-500Z tends to oscillate, in this particular socket, at 180-220 
MHz. The 8877, at about 315 MHz. 

If the anode has a high Q, high impedance very near 315 MHz in the 
8877, it will oscillate. If the anode has a high Q moderate to high 
impedance at about 200 MHz in the 3-500Z (with this  socket) it will 
oscillate at about 200 MHz.  

If I did not measure the particular socket you have, or you can 
not determine the anode lead impedance, you would be better off using 
some sort of suppressor. With the 8877, you want the suppressor to 
look like a dominate RESISTANCE at about 300 MHz with the standard 
Johnson socket. 

If you can keep the anode Z low, or park the anode parallel resonance 
away from the grid's high Z frequencies, the tube will be totally 
stable without any suppressor. In most cases, with a good layout, the 
8877 is totally stable without any suppressor at all.

If you look at 8877 PA's  you will see if they use a suppressor it is 
almost always a small hairpin suppressor. The suppressor only needs 
to be effective at lower UHF, and only if the UHF anode impedance 
promotes oscillation (unlikely in a clean layout). 

A 3-500Z on the other hand almost always requires a suppressor, and 
it takes a lot more inductance (less if you ground the grid pins 
directly, more turns if you use long grid leads).

Always install the suppressor right at the anode, and use one for 
each tube in a two tube PA. Use a WIDE SHORT low impedance solid strap 
from the junction of the suppressor to the tuning cap (through a 
good design blocking cap of course).

73, Tom W8JI

--
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