[Amps] Fw: Tank tweaking

[Bill Coleman]

I left off some info, target Q is 12.  Have also made some progress.  =20

I stuck in an additional 1000 pF on the plate blocking cap (from 800 pF =
to 1800 pF) & the results are more reasonable.  Am now able to get 1500W =
with reasonable efficiency, but the actual load C value is still low =
(was abt 600 pF, now 1000 pF, should be closer to 1700 pF). =20

The tune C value is right in line with the calculations (calc'd at 298 =
pF, actual 285 pF). =20

Now I need to find a cap or two that will survive on 10M.=20
~~~~~~~~~~~~~~~~
Bill Coleman    N2BC
http://home.stny.rr.com/n2bc
----- Original Message -----=20
From: Bill Coleman=20
To: AMPS=20
Sent: Sunday, September 22, 2002 9:51 AM
Subject: Tank tweaking


A few months ago I posted a query about crummy efficiency on my homebrew =
5CX1500 AB1 amp. I have cured most of the original problems by re-doing =
the input circuit and tweaking the output tank coils. Oh, 'tweaking' =
means creating a lot of scrap copper.

It works well on 10 through 80.   Efficiency on the order of 60%, the =
'real world' tune and load pF are very close to the calculated values.  =
I'm using a spreadsheet I cooked up from the W5FD formulas.

But... I have been chasing my tail on 160M.  The calculated C1 / L / C2 =
values at 1.9 MHz are 298pF / 25.47 uH / 1708 pF.  Actual values are 276 =
pF / 25.53 uH / 600 pF.  Efficiency is off to 48%. =20

The tank coils are 3/16" copper tubing for 10/15/20 and a hunk of coil =
stock for 40/80/160.   Plate blocking may be a bit small @ 800 pF - =
would that have such a huge effect on loading?  The plate choke is an =
RFParts RFC-3 (Ameritron) 225 uH & I don't feel any undo heating.  Plate =
bypass is a pair of 500 pF @ 30KV caps & I've tried adding an additional =
1000 pF 6KV disk ceramic with no effect.

The input is across a 200 ohm grid resistor with a 4:1 balun. It is =
untuned on 160 and 80, tuned on 10 thru 40.

Ideas?

Thanks!   73, Bill   N2BC

~~~~~~~~~~~~~~~~
Bill Coleman    N2BC
http://home.stny.rr.com/n2bc


--- StripMime Report -- processed MIME parts ---
multipart/alternative
  text/plain (text body -- kept)
  text/html
---