From: Larry East (LVE1@inel.gov)
This is for all of you QRP+ owners who a are not afraid to muck about in the
innards of your great little toy to make it even better!
1. Eliminating keying thump.
When the rig is keyed, a transient is induced on the +12V power buss that
causes a low frequency "thump" in the audio output; this is most noticeable
when using headphones. The magnitude of the thump depends on the "stiffness"
of the 12V power source as well as the low frequency response of the
Some time ago Norbert, DL8BDF, posted a fix for this problem; his solution
was to increase the size of the decoupling cap and resistor to the LM386
I have a somewhat simpler solution: simply change C35 -- the bypass cap on
pin 7 of the LM386 -- form the original 0.1uF to a electrolytic or tantalum
cap in the range 4.7uF to 10uF. The purpose of the bypass on pin 7 is to
decouple the high gain input stage of the LM386 from the power source and
the recommend size is of the order of 4.7uF; 0.1 is just not enough to keep
the keying induced transient from getting through. (Increasing the size of
this cap will also reduce hum from poorly filter power supplies.)
I replaced C35 with a 4.7uF dipped tantalum and for good measure swapped C40
and C41. This swap results in a little more decoupling to the LM386 and a
slight reduction in its low frequency response. These changes resulted in
the complete elimination of the keying thump in my case -- your results may
vary depending on the power supply you use.
2. Transmit/Receive switching click.
There is a slight "click" in the audio output when the rig switches from
transmit to receive. Norbert also posted a fix for this: I haven't
installed it yet because the above described "thump" was much more annoying
to me than the "click". Now that the "thump" is gone, guess I'll now do
something about the "click".
Norbert's fix is as follows:
Break the trace going to pin 11 of U8 on the audio board and insert a 0.1uF
cap; also add a 100K resistor from pin 11 of U8 to ground. This eliminates
DC being applied to C29 every time U8 switches on, which is reported to be
the cause of the "click".
3. Eliminating/Reducing receiver "birdies"
There are several spurious signals present in the PLUS receiver, most of
very low intensity but a few in the S1 to S3 range. There is also a problem
with very strong signals on 80M on or near the frequency used by W1AW
bulletins (can't recall frequency at the moment...) The fix for the "W1AW
problem" suggested by INDEX is to move the shield of the coax running from
the LO to the RF boards to the ground end of L5 on the LO board. This seems
to eliminate the problem, but some reports indicate that it induces more
"birdies" into the receiver. Well, I think I have (at least a partial) cure
for the problem!
There is a "ground buss" on the component side of the LO board between the
microprocessor and associated digital stuff and the local oscillator
circuitry -- the shield of the LO output coax was originally connected to
this ground buss. I decided to see what effect installing additional
shielding between the microprocessor and LO would have, and soldered a strip
of 0.025 x 0.25 inch brass (obtained from a local hobby store) vertically to
the ground buss. I also moved the coax shield to the ground end of L5 and
made sure the coax was routed away from the digital side of the board.
The end result: Most of the "spurs" previously noted on 160, 80 and 40
meters are either gone completely or drastically reduced! Examples: a
"multiple tone" spur around 3845.6 which was originally S3 is now just above
the noise; a similar spur around 7142.2 went from S1 to just above the
noise. In addition, a spur at 10.000 MHz that made WWV almost impossible to
copy is now less than S1. The down side: a spur at 21164.4 went from S1 to
S2 -- oh well, you can't win 'em all!
I did a similar thing (i.e., added vertical shielding) to the two ground
buss strips on the RF board. It appears that this may have reduced
"blow-by" around the xtal filter, but I don't have any quantitative data to
prove it. Anyway, it does not seem to have hurt anything.
72, Larry W1HUE/7
Search QRP-L Archives
QRP-L Archive |
[ 1993 | 1994 | 1995 | 1996 | 1997 | 1998 | 1999 | 2000 ]
This archive was generated by hypermail 2b29 on Fri Jun 02 2000 - 11:29:50 EDT