> From: wrt@eskimo.com (Bill Turner)
To: <amps@contesting.com>
> Date: Wed, 27 Aug 97 00:37:58 +0000
> A question about possible VHF parasitic oscillations in a VHF
> amplifier:
>
> Using a grounded-grid 8877 on 50 MHz, with the tube output capacitance
> serving as the only tank capacitance, is there a need for a parasitic
> suppressor?
There usually isn't even a need at HF with that tube Bill, if the
grid is properly grounded. Tubes that use a cone type grid
connection and a short grid length, when grounded directly to the
chassis, are most often unconditionally stable.
That's because the grid design maintains a very low impedance all the
way up to the frequency where the tube has little or no gain.
If you look at tubes, you'll see this is true. Tubes like the 811A
and 572 with long thin leads are very unstable, so is the 3CX1200A
and D7 (the Z7 doesn't even need a suppressor, and even works on
two meters) and 3-1000 and 4-1000.
Tubes like the 8877, 3CX3000, 3CX5000 and so on are almost always
totally stable without any precaution in a good layout.
Other tubes, like the 3CX800 and 3-500Z are in between.
> The tank circuit would be comprised of only the tube
> capacitance and a coil connected directly from anode to B+, with the
> usual bypass capacitor, of course. Tuning would be done by
> compressing the coil, and output coupling would be a tuned link as was
> typical before pi-networks became common.
Short of owning a network analyzer, the only way to tell if you
will have problems is to try it. If you do have parasitics, you
should easily be able to correct them.
The key is using a large continuous groundplane and
surround the grid ring with finger stock bolted directly to
the groundplane. No leads from the output circuit should go below the
chassis, into the area of the the tube. Every VHF amp I have built
(and HF amp also) using the 8877 has been totally stable. It's a
great tube if built properly, it's actually difficult to make one
oscillate without doing it intentionally!
> It would seem that the VHF resonant circuit present at HF using
> a pi-network simply is not there and no suppressor would be needed.
> If I were to use a separate tuning capacitor, then the unintentional
> VHF circuit would appear again, right? Or am I making a fundamental
> error here?
The grid is the dominate element, ground it well and problems
disappear. At 50 MHz and above, I'd use a flapper cap rather than a
conventional cap to keep Q and losses down.
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
|