Speaking of blocking capacitors (and at the risk of starting a
My 6m amp, which uses an 8877 and a pi network, seemed to suffer
from the same problem as Lance's -- I could get 1500 W out, but
just barely. It seemed strange when I can easily get 1500+ W
out of my 2m W6PO-style 8877 amp.
Since both Lance and I were using two 500 pF caps in parallel
for the blocking capacitors on our 6m amps, and since he reported
improved output by changing to two 100 pF caps in parallel, I
thought I'd give it a try. Only I thought I'd take it a small
step further and use two 75 pF caps in parallel instead of the
2 X 100 pF caps that Lance used.
So yesterday I swapped out the caps, let the amp warm up, and
applied some drive. When I released the PTT, the grid current
meter came to rest on the left sided peg. After some investigating,
I found that one of the filament feedthrough caps was developing
a short to ground (only about 800 ohms resistance to ground).
I wrote off the failure of the feedthrough cap as a coincidence,
replaced the faulty component, and put the amp back together.
Once back together, the amp warmed up and -- good, no more reverse
grid current reading. But as soon as I applied a small amount of
drive and then released the PTT, the reverse grid current was back.
Checking, I found that the same feedthrough capacitor had failed.
This morning I replaced the feedthrough capacitor once again and
replaced the 2 X 75 pF blocking caps with 4 X 120 pF (4 instead of
2 to better handle the current and to significantly increase the
value of the blocking caps). Now the amp tunes up just fine -- no
more blown feedthroughs -- and I'm seeing 200 more watts out with
similar plate current readings and drive requirements (when compared
with the amp with 2 X 500 pF caps).
My question: why did the feedthrough caps blow when I had the
2 X 75 pF caps in as blocking caps? I assume there was RF voltage
getting to the feedthroughs, but I don't understand why. The amp
uses a bifilar wound filament choke and the feedthroughs that failed
were on the transformer side of that choke. And I don't really
understand why I should be seeing increased power out of the amp now.
By the way, I now get 1500 W out with .90 A of plate current, 2600 V
key down, and 40 mA of grid current.
73, Arliss W7XU
Lance Collister wrote:
> Just in case you have been wondering about YOUR 6m 8877 amp...
> I have been using a homebrew 8877 amp (Pi-L output) for the last six years
> (since I
> started on 6m). The efficiency always seemed significantly lower than my W6PO
> amp on 2m. At one point, there were a number of other people with similar
> I was able to make some improvements by making changes to the input circuit,
> but was
> not satisfied with the results.
> In talking last night with Barrie, W7ALW, he mentioned he was thinking about
> changing the blocking capacitor in his 6m 8877 amp before he hooked it up
> again for
> 6m EME. I asked what that was all about. He explained that the ONLY thing
> he had
> ever seen the Elmers on the AMPS email reflector agree on, was that the
> blocking cap
> should be about 4 times the value of the tuning capacitor. I don't know if
> includes the stray capacitance or not, but I know my 1000 pF blocking cap
> (two 500
> pF doorknobs in parallel) sure did not fit that guideline!
> So, since 6m was dead up here today as usual, I decided it was a good time to
> things up and operate. I replaced the two 500 pF blocking caps with two 100
> pF caps
> (just because I happened to have them handy). I might have even gone down to
> a pair of 50 pF caps, but I only had one of those in the bin here. Anyway,
> result was about 10-15% more output, and much smoother tuning! So, even
> there was no DX in my log today, the amp (with the 20 year old tube) is
> better than ever! ;-)
> TNX, Barrie! GL and Happy Holidays to all! Vy 73, Lance
FAQ on WWW: http://www.contesting.com/FAQ/amps
Administrative requests: amps-REQUEST@contesting.com