Message-ID: <50783F5D.60809@iup.uni-heidelberg.de>
Date: Fri, 12 Oct 2012 18:03:41 +0200
From: =?ISO-8859-15?Q?Stefan_Sch=E4fer?= <Stefan.Schaefer@iup.uni-heidelberg.de>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 6.1; de; rv:1.9.1.8) Gecko/20100227 Thunderbird/3.0.3
MIME-Version: 1.0
To: rsgb_lf_group@blacksheep.org
CC: Vasily Savchenko <ua0snv@yahoo.com>
Subject: LF: WSPR on the 630m band
Content-Type: multipart/alternative;
 boundary="------------050605010206060206030809"
Sender: owner-rsgb_lf_group@blacksheep.org
Precedence: bulk
Reply-To: rsgb_lf_group@blacksheep.org


This is a multi-part message in MIME format.
--------------050605010206060206030809
Content-Type: text/plain; charset=ISO-8859-15; format=flowed
Content-Transfer-Encoding: 7bit

MF!

Last night i built a transmit converter which allows me to transmit 
*WSPR on 630m* now.

Yes yes yes yes yes Mal, CW is preferred! But many OMs go to bed quite 
early i find. Not so their PCs, they run over night, and so my 
transmitter will do this weekend if all goes right. :-) I'm not present 
in the shack anyway, so no chance for CW...

I'm using _*normal WSPR*_, not the slow versions.

Right now the first tests are running and i check if the system becomes 
to warm.

The design of this mixer is quite chaotic but i like it :-) It is a 
modular system...: First, a tone of 14700 Hz is generated by the WSPR 
program. Is is decoupled from anything by an audio isolating 
transformer. This signal is applied to a SA612 mixer. The LO (461 kHz) 
in this test setup comes fom my DDS VFO which is used in the first VLF 
experiments. The output looks very well as expected on the scope. 
Filtering of mixing products is done in my MF meachanical filter which 
is use in front of the receiver normally! I've published the nearly 
perfect filter curve some weeks ago. Of course the sine wave coming out 
from the filter looks very well! This sine wave is applied to an OP AMP 
which is switched as a comparator. I've added a small hysteresis (10 
MOhm / 47k) to prevent an oscillating output when there is no input 
signal. Works well! The output of the OP AMP is fed to a ICL7667 (FET 
driver) which makes a low impedant output. In series with 1 Ohm and 1 uF 
this rectanfgular waveform is applied to a small ferrite toroid with a 
trifilar winding. So this output is again galvanically decoupled from 
the circuit and provides an opposite phase rectangular signal at 475.7 kHz.
Since my PA was/is designed to have a 4f input (1900 kHz -> 475 kHz out) 
i had to make a tap behind the divider, so i have to take care to 
maintain the 50 % duty cycle in the new circuit. The new external drive 
signal from the converter is applied by a simple headphone jack/plug! I 
actually found a 3.5mm stereo-jack with dimensions 10mm x 5mm and also 
found a few holes to add this on the board. So the PA is now able to run 
as before and additionally with this new converter......

The AF signal levels and the LO signal level is critical. Choosing wrong 
values will change the 50% duty cycle which will quickly kill the FETs 
in this PA design. So i'm using an old modified PC power supply that 
will shut down quickly if the current exceeds a certain limit...

After writing this email the PA became hand warm so it looks all fine. I 
will go down to 50% TX duty cycle now and even receive during the breaks 
in WSPR. However the MF filter is now used for the TX so the RX 
performance may be slightly lowered...

*Reports about my transmissions on 475.700 kHz are very welcomed. I'm 
also receiving and uploading to the WSPR database...
Maybe i will run the test until monday :-)*

Best 73, Stefan/DK7FC

PS: https://dl.dropbox.com/u/19882028/MF/stuff.JPG

--------------050605010206060206030809
Content-Type: text/html; charset=ISO-8859-15
Content-Transfer-Encoding: 7bit

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>

<meta http-equiv="content-type" content="text/html; charset=ISO-8859-15">
</head>
<body bgcolor="#ffffff" text="#000000">
MF!<br>
<br>
Last night i built a transmit converter which allows me to transmit <b>WSPR
on 630m</b> now.<br>
<br>
Yes yes yes yes yes Mal, CW is preferred! But many OMs go to bed quite
early i find. Not so their PCs, they run over night, and so my
transmitter will do this weekend if all goes right. :-) I'm not present
in the shack anyway, so no chance for CW...<br>
<br>
I'm using <u><b>normal WSPR</b></u>, not the slow versions.<br>
<br>
Right now the first tests are running and i check if the system becomes
to warm.<br>
<br>
The design of this mixer is quite chaotic but i like it :-) It is a
modular system...: First, a tone of 14700 Hz is generated by the WSPR
program. Is is decoupled from anything by an audio isolating
transformer. This signal is applied to a SA612 mixer. The LO (461 kHz)
in this test setup comes fom my DDS VFO which is used in the first VLF
experiments. The output looks very well as expected on the scope.
Filtering of mixing products is done in my MF meachanical filter which
is use in front of the receiver normally! I've published the nearly
perfect filter curve some weeks ago. Of course the sine wave coming out
from the filter looks very well! This sine wave is applied to an OP AMP
which is switched as a comparator. I've added a small hysteresis (10
MOhm / 47k) to prevent an oscillating output when there is no input
signal. Works well! The output of the OP AMP is fed to a ICL7667 (FET
driver) which makes a low impedant output. In series with 1 Ohm and 1
uF this rectanfgular waveform is applied to a small ferrite toroid with
a trifilar winding. So this output is again galvanically decoupled from
the circuit and provides an opposite phase rectangular signal at 475.7
kHz. <br>
Since my PA was/is designed to have a 4f input (1900 kHz -&gt; 475 kHz
out) i had to make a tap behind the divider, so i have to take care to
maintain the 50 % duty cycle in the new circuit. The new external drive
signal from the converter is applied by a simple headphone jack/plug! I
actually found a 3.5mm stereo-jack with dimensions 10mm x 5mm and also
found a few holes to add this on the board. So the PA is now able to
run as before and additionally with this new converter......<br>
<br>
The AF signal levels and the LO signal level is critical. Choosing
wrong values will change the 50% duty cycle which will quickly kill the
FETs in this PA design. So i'm using an old modified PC power supply
that will shut down quickly if the current exceeds a certain limit...<br>
<br>
After writing this email the PA became hand warm so it looks all fine.
I will go down to 50% TX duty cycle now and even receive during the
breaks in WSPR. However the MF filter is now used for the TX so the RX
performance may be slightly lowered...<br>
<br>
<b>Reports about my transmissions on 475.700 kHz are very welcomed. I'm
also receiving and uploading to the WSPR database...<br>
Maybe i will run the test until monday :-)</b><br>
<br>
Best 73, Stefan/DK7FC<br>
<br>
PS: <a class="moz-txt-link-freetext" href="https://dl.dropbox.com/u/19882028/MF/stuff.JPG">https://dl.dropbox.com/u/19882028/MF/stuff.JPG</a><br>
</body>
</html>

--------------050605010206060206030809--