Elecraft › [K2]

worse carrier rejection when KPA100 active?

Classic

List

Threaded

4 messages Options

dl0tuh

worse carrier rejection when KPA100 active?

I've built KPA100 for my K2. When doing two-tone test, I was
satisfied with results at 10 watts power (KPA100 in bypass), they
are shown on photo (1).

Then I increased the power to 11 watts, and what I saw wasn't just a
two-tone signal any more, see photo (2). As if carrier rejection
were not enough when running with KPA100.

So I decided to do a single tone test. While looking good at 10
watts (see photo 3), at 11 watt it looked like an AM signal with
small modulation depth (photo 4). I even played with carrier balance
pot on KSB2, and I was able to make it worse, hi. (Then I re-balanced
it at 10 watts, with mic disconnected, using oscilloscope.)

Any ideas of possible reason?

Some comments to setup. KPA100 running into 100W dummy load.
Bias set to 400mA (2mV on .005 Ohm). Oscilloscope loosely coupled
via an isolated wire near the power resistor. Tones were generated from
PC (photo 5). A wattmeter showed (almost) no difference between
10 and 11 watt, so I hope kpa100's power meter is calibrated correctly
(photo 6). For every test, PC output was adjusted so that one ALC LED
was lighting (photo 7). No other changes were made when switching
between 10 and 11 watt.

(1) http://www.mydarc.de/dl2ipu/kpa100/two_tone_10_watt.jpg<https://mail.eu.cargill.com/exchweb/bin/redir.asp?URL=http://www.mydarc.de/dl2ipu/kpa100/two_tone_10_watt.jpg>
(2) http://www.mydarc.de/dl2ipu/kpa100/two_tone_11_watt.jpg<https://mail.eu.cargill.com/exchweb/bin/redir.asp?URL=http://www.mydarc.de/dl2ipu/kpa100/two_tone_11_watt.jpg>
(3) http://www.mydarc.de/dl2ipu/kpa100/one_tone_10_watt.jpg<https://mail.eu.cargill.com/exchweb/bin/redir.asp?URL=http://www.mydarc.de/dl2ipu/kpa100/one_tone_10_watt.jpg>
(4) http://www.mydarc.de/dl2ipu/kpa100/one_tone_11_watt.jpg<https://mail.eu.cargill.com/exchweb/bin/redir.asp?URL=http://www.mydarc.de/dl2ipu/kpa100/one_tone_11_watt.jpg>
(5) http://www.mydarc.de/dl2ipu/kpa100/tone_generator.jpg<https://mail.eu.cargill.com/exchweb/bin/redir.asp?URL=http://www.mydarc.de/dl2ipu/kpa100/tone_generator.jpg>
(6) http://www.mydarc.de/dl2ipu/kpa100/wattmeter.jpg<https://mail.eu.cargill.com/exchweb/bin/redir.asp?URL=http://www.mydarc.de/dl2ipu/kpa100/wattmeter.jpg>
(7) http://www.mydarc.de/dl2ipu/kpa100/k2_ALC_indication.jpg<https://mail.eu.cargill.com/exchweb/bin/redir.asp?URL=http://www.mydarc.de/dl2ipu/kpa100/k2_ALC_indication.jpg>

73! Oleg
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

Don Wilhelm-4

Re: worse carrier rejection when KPA100 active?

Oleg,

It does not look like the pattern is related to the 2 tone test, but is
at a much lower frequency - perhaps related to your AC line frequency.
Are you powering the base K2 from a power supply different from that
used on the KPA100, then check that supply for ripple. Even if you are
using the same supply, check it for DC purity at the current draw at 10
watts and again at the current draw at 11 watts (or greater).

To eliminate the power supply from question, try running the KPA100 on a
sizable 12 volt battery (like an automotive battery) and see if there is
any improvement.

You might also see what happens if you change the tones. The Elecraft
2Tgen uses tones of 700 and 1900 Hz which have greater spacing than the
tones you are using (do not use harmonically related tones).

73,
Don W3FPR

Amateurfunkstation TUHH wrote:

> I've built KPA100 for my K2. When doing two-tone test, I was
> satisfied with results at 10 watts power (KPA100 in bypass), they
> are shown on photo (1).
>
> Then I increased the power to 11 watts, and what I saw wasn't just a
> two-tone signal any more, see photo (2). As if carrier rejection
> were not enough when running with KPA100.
>
> So I decided to do a single tone test. While looking good at 10
> watts (see photo 3), at 11 watt it looked like an AM signal with
> small modulation depth (photo 4). I even played with carrier balance
> pot on KSB2, and I was able to make it worse, hi. (Then I re-balanced
> it at 10 watts, with mic disconnected, using oscilloscope.)
>
> Any ideas of possible reason?
>
>

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html

dl0tuh

Re: "worse carrier rejection" when KPA100 active?

Don, thanks for your reply!

I repeated my tests with a 12V 7Ah battery connected to KPA100 power terminals as the only power supply of K2. And, to eliminate 2tone complexity, I only used single 1900 Hz tone.

Unfortunately for me, there is no change compared to 20A PSU used yesterday. 10 watts are ufb (photo 3), but 11 watts signal looks much like an AM signal (photo 4). The frequency of the envelope on photo 4 is the tone frequency (well, may be +- 50 or 100 Hz), I tried changing it.

(3) http://www.mydarc.de/dl2ipu/kpa100/one_tone_10_watt.jpg
(4) http://www.mydarc.de/dl2ipu/kpa100/one_tone_11_watt.jpg

I'm not expecting 50 Hz buzz from my sound card, that one I would see at 10 watts, too. Neither I'm expecting any ripple from a battery. Out of ideas by now...

73! Oleg

P.S. Remember photo-effect of power transistors of KPA100, producing abt 0.2V where 0V should be during DC checks? I have a very strong light above my table, hi

Don Wilhelm-4

Re: "worse carrier rejection" when KPA100 active?

Oleg,

Unless you changed the time-base on your oscilloscope between the
photos, I calculate that the period of the ripple is close to 0.11 ms
which would mean a ripple frequency near 9000 Hz., but you will have to
tell me how your'scope time base is set.
It may be possible for your light source to cause that effect if your
light is a florescent type. An incandescent light would produce that at
the 50 Hz line frequency. Put some black tape over the PA transistors
as a test and see if the ripple is reduced.

You will likely have to probe around in the KPA100 circuit to see where
it originates. If the PA transistor base has a clean signal, and your
DC Supply is clean, then I do not have a good answer.
OTOH, if the ripple is present at the PA transistor base you should
probe further to see if it is coming from the bias supply (U7B and Q4)
or from somewhere in the T/R switch area.

If you have the KPA100 mounted to the base K2 when you are doing this
test, be certain the shield is fully in place and that the KPA100 ribbon
cable and speaker cable are dressed well away from the KSB2 board.
Remove the right side panel and tuck those cables up into the area
between the KPA100 heat sink and the top of the Control Board.

73,
Don W3FPR

dl0tuh wrote:

> Don, thanks for your reply!
>
> I repeated my tests with a 12V 7Ah battery connected to KPA100 power
> terminals as the only power supply of K2. And, to eliminate 2tone
> complexity, I only used single 1900 Hz tone.
>
> Unfortunately for me, there is no change compared to 20A PSU used yesterday.
> 10 watts are ufb (photo 3), but 11 watts signal looks much like an AM signal
> (photo 4). The frequency of the envelope on photo 4 is the tone frequency
> (well, may be +- 50 or 100 Hz), I tried changing it.
>
> (3) http://www.mydarc.de/dl2ipu/kpa100/one_tone_10_watt.jpg
> (4) http://www.mydarc.de/dl2ipu/kpa100/one_tone_11_watt.jpg
>
> I'm not expecting 50 Hz buzz from my sound card, that one I would see at 10
> watts, too. Neither I'm expecting any ripple from a battery. Out of ideas by
> now...
>
> 73! Oleg
>
> P.S. Remember photo-effect of power transistors of KPA100, producing abt
> 0.2V where 0V should be during DC checks? I have a very strong light above
> my table, hi
>
> ------------------------------------------------------------------------
>
>
> No virus found in this incoming message.
> Checked by AVG - www.avg.com
> Version: 9.0.733 / Virus Database: 271.1.1/2730 - Release Date: 03/08/10 02:34:00
>
>