Reversed tuning?

4 views
Skip to first unread message

Andrew Lees

unread,
Oct 6, 2009, 7:06:59 AM10/6/09
to lin...@googlegroups.com
Hi,

I'm trying out linrad with a softrock 80 board, and on comparing the signals between it and a Kenwood receiver, I find that the direction of tuning is reversed, so that signals above the crystal frequency appear below on the linrad display, and vice-versa.  There is undoubtedly  a simple reason for this, I wonder is anyone can suggest it?

I have checked the correct connection to tip and ring of the audio cable.  The sound card is a soundblaster 24bit usb at 96KHz sample rate, and generally things appear to be working very well, there is good image rejection - it's just in reverse!

Thanks for any assistance,

Andrew Lees.

Alex

unread,
Oct 6, 2009, 8:33:23 AM10/6/09
to Linrad
You have to set vertial red lines to another side.

Andrew Lees

unread,
Oct 6, 2009, 9:23:49 AM10/6/09
to lin...@googlegroups.com
Hi Alex,

That appears to correctly select upper or lower sideband for ssb reception, but doesn't seem to affect the macro tuning behaviour.  This is what I would expect from the help - I understand that the red lines define the offset from the tuning frequency of a "digital oscillator" used for demodulating ssb and cw signals.

Regards,

Andrew

Leif Asbrink

unread,
Oct 6, 2009, 4:44:00 PM10/6/09
to lin...@googlegroups.com
Hi Andrew,

> I'm trying out linrad with a softrock 80 board, and on comparing the
> signals between it and a Kenwood receiver, I find that the direction of
> tuning is reversed, so that signals above the crystal frequency appear
> below on the linrad display, and vice-versa. There is undoubtedly a
> simple reason for this, I wonder is anyone can suggest it?

Yes, you have two audio signals I and Q. If you exchange them the frequency
scale would invert.

Whether I is channel 1 and Q is channel 2 or vice versa depends on
the hardware. In Linrad the user is supposed to edit users.c and rename it
to users_hwaredriver.c to adapt Linrad to the hardware in use.
At line 172 you would find:
//************************************************************
// This routine is responsible for setting
// int fg.passband_direction = -1 or +1 depending on whether the
// hardware local oscillator is above or below the desired signal.
//************************************************************
// This routine is responsible for setting
// int fft1_direction = fg.passband_direction
}

With a softrock you always would have the same center frequency
so you can set -1 or +1 to fit your hardware. You might set a sign
that depends on the center frequency because you might use converters
that reverse the received frequency band by having an LO above the
signal frequency.

Once you have changed the source code it is trivial to recompile Linrad
from source code as long as you use Windows XP or earlier. It might be
possible to use Vista in XP compatible mode but I do not know for sure.

73

Leif / SM5BSZ

Andrew Lees

unread,
Oct 6, 2009, 6:10:38 PM10/6/09
to lin...@googlegroups.com
Hi Leif,

That was a very kind way of saying "RTFM", thankyou!

I'm using Linux, so the source is already there and compiled.  The next step is to get an SI570 integrated, so I'll need to look at the tuning area and set it up to send the required commands.

Thanks again.

Andrew
Reply all
Reply to author
Forward
0 new messages