KXB3080 - Same output power on 40m as 20m - How to fix?

[Martin Gillen-2]

Hi, Elecrafters.

I'm hoping someone can help!

I completed my KXB3080 modification tonight but during testing I
discovered that I am getting pretty much the same transmit power out
(1.8W with 10V) on 40m as 20m (according to the KXAT1 this is the only
watt-meter I have) into a 50 ohm load.  I took the board out and
squeezed the green turns on L2 of the LPF1 board together as tightly
as I could and replaced the board back into the case, but it didn't
make any difference I still get the same power out on 40m as 20m.  I
understand that this may be an indication that the harmonic
suppression is not working correctly as I should be getting about .6W
lower power on 20m (I certainly was before the mod!)

I placed a scanner across the room (with a 2m rubber duck antenna)
from the KX1 driving a magnetic loop antenna tuned to 7040kHz and
transmitted on 7040kHz.  The scanner showed full scale deflection.  I
then tuned the scanner to the 1st harmonic 14080kHz and the scanner
'S' meter was maybe 4/5ths deflected.

I know that isn't very scientific, it may be telling me the harmonic
suppression isn't good or maybe it's not.  I don't have anything
sophisticated like a spectrum analyzer to do a proper analysis of the
relative strength of the 1st harmonic versus the fundamental.

The LPF1 went pretty much the way the manual said it should.  I'm sure
it's getting power as I see a 60mA draw on 80m.  The only difference
between my LPF1 and the manual's picture is that my LPF1 PCB doesn't
sit flush to the KX1 PCB because the turns on L1 and L2 extend
slightly past the edge of the LPF1 board.  It is about 1mm away from
the KX1 board, but the turns of L1 & L2 are flush to the board.

Does anyone have any suggestions about how I could start tracing this
fault?  Should I try compressing the red turns on L2, or the green or
red turns on L1?

Thanks,
Martin.
VA3SIE.
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Geoffrey Mackenzie-Kennedy-2]

Martin VA3SIE wrote:

> I placed a scanner across the room (with a 2m rubber duck antenna)
> from the KX1 driving a magnetic loop antenna tuned to 7040kHz and
> transmitted on 7040kHz.  The scanner showed full scale deflection.  I
> then tuned the scanner to the 1st harmonic 14080kHz and the scanner
> 'S' meter was maybe 4/5ths deflected.
>
> I know that isn't very scientific, it may be telling me the harmonic
> suppression isn't good or maybe it's not.  I don't have anything
> sophisticated like a spectrum analyzer to do a proper analysis of the
> relative strength of the 1st harmonic versus the fundamental.

-------------------------------------------------------------------------

Martin,

Your magnetic loop antenna could be confusing your measurement using the
scanner because a magnetic loop exhibits narrow bandwidth, and in normal use
would act as a filter reducing the radiated level of the 20m harmonic.
Before I am flamed on this I know that there are many 'buts' and 'ifs' :-).

When using your scanner it would be better to drive the Tx into a dummy
load, and position the scanner plus rubber duck so that you get * no more*
than half scale deflection of its 'S' meter on 7040 kHz.to avoid any
possibility of overloading the scanner, then compare that reading with the
scanner's 'S' meter reading on 14080 kHz. It would be better to have the Tx
in its case when doing this to avoid any direct radiation from the PA etc
being picked up by the scanner. This measurement will give you a useful
indication but not a precise figure.

Sorry but I am unable to comment on the KXB3080 since I do not have one.

73,
Geoff
GM4ESD



_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Don Wilhelm-3]

Martin,

You may be reading too much into the instructions.  It is not required that
the power on 20 meters be lower than on 40, it just usually happens that
way.  The fact that you have the same power output on 20 as you have on 40
is not necessarily an indication of a lack of 2nd harmonic suppression.

Look at the schematic and you will see that there are two 1000 pF capacitors
added to the LPF on 40 meters, and the cutoff frequency when tuned to 40
will be lower than when tuned to 20.

The green winding is used on 40, 30 and 20 while both the red and green
windings are used on 80.  Normally the red winding should be squeezed a bit
and the green winding spread out somewhat.  Count the number of turns to be
certain L1 and L2 are wound correctly. Count the number of times the wire
goes through the center of the core rather than counting around the outside.

I would not trust the front end of a scanner to determine the relative
strength of a 2nd harmonic.  If you wish to do that, put a tuned circuit
between the antenna and the input to the scanner to give the front end of
the scanner some selectivity - it likely is wide open for broadband
coverage.

The KX1 power display has a lot of granularity, but you can obtain a smooth
relative reading from the KXAT1 by reading the voltage from the forward side
of the wattmeter with a DMM.  One easy to reach place for this voltage is
pin 5 of P3.

73,
Don W3FPR

> -----Original Message-----
>
> I completed my KXB3080 modification tonight but during testing I
> discovered that I am getting pretty much the same transmit power out
> (1.8W with 10V) on 40m as 20m (according to the KXAT1 this is the only
> watt-meter I have) into a 50 ohm load.  I took the board out and
> squeezed the green turns on L2 of the LPF1 board together as tightly
> as I could and replaced the board back into the case, but it didn't
> make any difference I still get the same power out on 40m as 20m.  I
> understand that this may be an indication that the harmonic
> suppression is not working correctly as I should be getting about .6W
> lower power on 20m (I certainly was before the mod!)
>
>

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Martin Gillen-2]

Hi, All.

Thanks to all who responded about the power output being the same
between 20m and 40m.

I was able to run some tests with a recently calibrated HP8591A
spectrum analyzer, and I thought I'd share my results.  (I also had a
chance to calibrate the KXAT1 it was reading about 0.25W high).

My battery voltage was:  9.1V key-up and 8.2V key down.

I was hooked up to the analyzer through a recently certified precision
10dB attenuator and a recently certified low loss coaxial feedline.
(I've adjusted the fundamental power below by 10dB for the pad).

               Fundamental           2nd             3rd
 4th           5th
Band            Power            Harmonic     Harmonic     Harmonic    Harmonic
              (Adjusted Pad)        Delta           Delta
Delta        Delta

 80m     30.6dBm (1.15W)     -45dB          -66dB            -76dB       -76dB

 40m     29.4dBm (0.87W)     -39dB          -52dB           -59dB        -64dB

 30m     30.4dBm (1.09W)     -40dB          -70dB           -72dB        -71dB

 20m     30.1dBm  (1.02W)    -60dB          -66dB           -70dB        -69dB

Observations:

The harmonic rejection is within specification, at least at 1W out.  I
assume that the level of harmonic suppression (delta) will remain the
same regardless of the power out, I'd be interested in comments on
that assumption as I did not have the right connector to run the radio
on 12V to hand when I made the measurement.

I do NOT have the 20m power being 0.3-0.6W lower than 40m (any more -
it was before the KXB3080 mod), in fact I have almost the opposite
(0.15W more power on 20m than 40m), and I assume that difference would
be more pronounced at higher output powers.  The manual states that my
harmonic rejection may therefore be sub-optimal:

"If both bands have adequate output power but 20 meters is higher, you
should adjust the windings of toroidal inductor L2. To do this,
"squeeze" the turns of L2 together slightly so that they occupy only
about 70% of the core's circumference. With L2 at its optimal turns
spacing (for best harmonic rejection), 20 meter output will be about
0.3 to 0.6 watts lower than 40 meters."

The worst case harmonic rejection seems to be 40m with the 1st
harmonic being -38.7dB down from the fundamental, but that's close
enough to the specification (-40dB) I think.  Should I be trying to
squeeze my turns to improve it?  I'm not sure.

I did compare my results to John AE5X's measurements for his KXB30 KX1:

  KX1                    2nd             3rd                4th           5th
Owner    Band    Harmonic     Harmonic     Harmonic    Harmonic
                           Delta           Delta             Delta        Delta


  AE5X    40m     -40dB          -53dB           -60dB        -66dB
VA3SIE   40m     -39dB          -52dB           -59dB        -64dB

  AE5X   30m      -35dB          -63dB           -71dB        -68dB
VA3SIE  30m      -40dB          -70dB           -72dB        -71dB

  AE5X   20m      -58dB          -60dB           -73dB        -70dB
VA3SIE  20m      -60dB          -66dB           -70dB        -69dB

So the KXB3080 low pass filters apparently offer superior harmonic
rejection compared to the KXB30 except on 40m where it's slightly
worse, although in practice the 1dB difference is probably close
enough.

Perhaps I could improve the 40m harmonic rejection by squeezing turns,
but it probably isn't worth it for 1dB.

I hope others can benefit from these measurements.

73
Martin.
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com