| |
DK8KV > QRP 16.03.00 20:42l 118 Lines 6138 Bytes #-9451 (0) @ EU
BID : G30DB0MKA02H
Read: DL2MHG DL4SHE GUEST
Subj: re:K2 power test
Path: DB0ZKA<DB0GPP<DB0LX<DB0CZ<DB0GE<DB0LJ<DB0MKA
Sent: 000316/1906z @:DB0MKA.#NRW.DEU.EU [HennefJO30PR DL3OE] DB19c1 $:G30DB0MKA
From: DK8KV @ DB0MKA.#NRW.DEU.EU (Walter)
To: QRP @ EU
Hallo zusammen,
es ist kein Problem, mit dem K2 auf 10m einen Output von 10...12W zu
erreichen. Hier ist ein Beitrag zu diesem Thema von Gary Surrency
von Elecraft Support. Ein weiterer Fehler könnte darin liegen, dass
in den LP-Filtern der beiden höchsten Frequenzbereichen durch falsche
Windungszählweise jeweils eine Windung zuviel gewickelt worden ist.
73 Walter, DK8KV
From: Gary L Surrency <gsurrency@juno.com>
Date: Wed, 13 Oct 1999 16:17:00 -0700
Subject: Re: [Elecraft] 10m Power Output
Message-ID: <19991013.161702.-280003.2.gsurrency@juno.com>
Mark,
I get 11-12 watts on all bands, including 10m. First, be sure of your
measuring equipment if you are going by some wattmeter other than the
internal K2 one. Also, check to see if the 1N34A diode in the RF output
detector is good and installed correctly. You can momentarily short the
anode of this diode (D9) to the ground foil of the nearby mounting screw
hole and see if when you remove this short the output goes up. If
everything is working well, the output will then rise temporarily until
the power regulating firmware in the microprocessor regains control. I
understand the new firmware, and that implemented in the SSB adapter,
will allow 15 watts output. It is obvious on my rig that this is easily
possible if everything is working properly. The current firmware in my K2
just throttles it back to 11-12 watts.
Make certain that the secondary windings on T2 cover at least half of the
toroid. I did this by stretching the windings around the core evenly and
then coming off the bottom with the leads close to each other. Then I
bent a right angle in the secondary leads so that they are the same width
as the pads located just outside the outline. Then I formed the leads so
that they pass along the bottom of the toroid and come out near the edge
of the core next to where the secondary pads are. I then made another
right angle bend in the secondary leads pointing downward to match the
pads. You can allow the right angle bends to stretch a little wider if
you like as you insert T2 and route its wires to the correct pads. Tin
the portion of the leads that go through the PCB up to the bend, and then
solder them in.
As Eric or Wayne said, it's OK to have a little bit of lead length on the
toroid. Just make sure to spread the windings out so that the
wire-to-wire capacitance is reduced.
FWIW:
I also tried Nil's binocular core transformer idea for T2. It works well,
but of the two sizes of bino cores I have, neither one seems appropriate.
The smaller one seems too small as it is a BLN43-2402, but I tried it
anyway. I had to use #26 ga. for the 3 turn primary, and #24 for the 2
turn secondary, thus maintaining the 12:8 ratio. This is possible because
of the greater amount of core that the windings are exposed to in the
balun core. More coupling with fewer turns is the result.
Even then, it was necessary to insert a small insulated wire into the
balun core holes temporarily after putting the 3 turns on the primary to
compress the windings enough to get room for the 2 turns of #24 for the
secondary. Once installed, the core did not get warm noticeably, except
on 10m I thought I could detect a bit of warming after 10 seconds at 12
watts. But even the original FT43-37 core gets a little bit warm on
extended operation at full power, so this is probably not a concern with
T2 either.
The power output was equal to or greater than that obtained using the
original T2 wound carefully as just described. It seemed to me that there
was a bit less current draw when using the binocular core however, based
on some quick current measurement I recorded. Perhaps Nils or someone
else will measure his own K2 current demand or others will try this
experiment with us. The tiny bino core I used sits easily within the
toroid outline for T2, and is easy to install once you get the wire
through the binocular holes. The slightly better efficiency may be due to
the reduced amount of copper wire and its associated lower resistance,
along with the known superior magnetic properties of bino cores.
The next larger core I had, a BLN43-202, was too large to fit between
C129 and C130. These caps are just outside of the mounting location for
T2, and they are too close togther to use this core which seems to be a
better choice than the little BLN43-2402. Pity, since it looks like the
exact right size for this application. If C129 and C130 were moved to the
bottom side of the PCB, assuming there is room, it would be perfect since
this larger core would allow just about any wire size within reason to
pass through its core.
I got my bino cores from Dan's Small Parts some time ago, but there are
also some bino cores available from Tech America (now RadioShack.com) in
their catalog on page 38. Some further experimentation on employing bino
cores in place of T2 may be worthwhile, assuming Wayne and Eric haven't
already done so. It is much easier to use bino cores for T2. I have not
looked into replacing T1 with another bino core yet. It would require the
larger core also, since it has a 9:3 turns ratio and I don't think I
could reduce that to 3:1 and still have the required coupling. Maybe, but
I'm not sure. I want to get a book on toroids and especially binocular
cores to become more knowledgeable in this area.
After I was finished with the bino core experiments, I elected to
re-install the original T43-37 transformer after re-winding it
ever-so-carefully.
K2 builders, do not rush through winding T2, as it is fairly critical to
good power output and efficiency as Eric has already posted some time
back. Wind T2 as neatly and accurately as possible. Maybe Elecraft and
some of you other K2 builders will do so more experimentation with Nils
and I to see if binocular cores (or balun cores, if you like) can be a
preferred replacement for at least T2, if not T1. It was an interesting
experiment to try, and I hope to learn a bit more about this type of
transformer thru further testing and reading.
73,
Gary Surrency AB7MY QRP-L #571 Chandler, AZ (near Phoenix)
Read previous mail | Read next mail
| |
