[pskmail] fldigi v x-76 testing

  • From: Bob Morgan <wb5aoh@xxxxxxxxxxxxxxxxx>
  • To: pskmail@xxxxxxxxxxxxx
  • Date: Thu, 19 Sep 2013 21:43:11 -0500

Rein, et al,

I managed to get the latest patched fldigi version -76 installed on
two similar ubuntu systems, and put them into BPSK1000 mode.
I was doing some acoustic coupling between systems.

First observation once I got a few things sorted out, was that the
receiving machine recognised the signal as QPSK63 instead of BPSK1000.
It would do this regardless of which system was sender and receiver,
in other words it seemed to be a bidirectional problem, and I made it
repeat the behavior a few times, and I would have to go back to each one
and reset things to BPSK1000, and sometimes I had to recenter to 1000 Hz,
for reasons I haven't tracked down yet, one receiving machine wanted to
shift to about 1500 Hz or so, even though I thought I had it locked down.

I then brought up the new jpskmail-1.5.25 on each one, and more or less
saw the same behavior there too, although I didn't test it quite as much,
as it is getting towards the end of the daytime I can spend on it today.
I did switch off the RX RSID switch on both ends and try to reset both ends
to BPSK1000 and 1000 Hz, but I didn't seem to be getting any recognisable
data.  I was using the CQ or POSIT pushbuttons to generate TX data on
jpskmail, and with fldigi by itself, I was just using the TR toggleswitch
to send its bitstream.

I had to shut off the PTT right channel tone so the ptt tone didn't get in the 
way,
up until that point I was guessing that was the only problem, but things
didn't improve aftewards.

I am running fldigi in the audio tx display loopback mode on the waterfall so I
can see what and where I am transmitting, and that part looks OK.
I can see similar waterfalls on the receiving system although I want
more audio gain in my test setup than what I was getting.  When I get some
more time and parts I probably am going to make up a pair of audio cables
just to interconnect the soundcards, as the testing is getting a little
awkward with what I am doing, and as it is, I have to take the laptop
to work to get it beside the desktop.  When I get a little farther along then
it will be time to connect things to radios, but the arrangement isn't ready
for that yet.

Mainly I am curious if there could be a mapping problem between BPSK1000
and QPSK63, since I remember that some RSID codes had to be shuffled around.
That is the reason I tried turning the RX RXID off for the next tests.

I probably ought to have first tried something well known like PSK250 or 500,
but I don't have a whole lot of time to work with it at one sitting,
so I got in a hurry, and the VHF BPSK1000 mode (and later on, the 2x1000 or
4x500 etc modes) are of primary interest to me right now.

Any thoughts?

One other quick question I would like someone to check my sanity on, while
I am thinking about it, is for the typical soundcard mike jack, are the
tip and ring terminals equivalent?  I thought they were monoaural and the
signal was on one of them and a DC bias voltage was on the other one
(I am not sure which I thought was supposed to be tip and ring).
On this laptop, I was seeing DC bias on both of them, and touching
either tip or ring would put plenty of hum noise spectrum onto the
waterfall display.  I didn't -think- it was a stereo mike port,
but maybe I am getting to wonder about it right now.
If it was a standard desktop line-input port it wouldn't have
any mike DC bias on it and it would be stereo, but this is just
a laptop with just a stereo output and a mike input.  When it is time to
wire things up for real it would be good to know which to use, or
might it be better to use a usb sound device instead which has a
regular line input port.  The reason for doubt with a mike jack is
that I was always under the impression that soundcard mike inputs
ran some compression and that there wasn't any way to disable it,
as opposed to a line input jack which was a fixed gain with no agc action.

Other questions:

I was designing an interface circuit for a radio and I am going to either
use the psk signal for ptt, or maybe also use the right channel 1 KHz tone
for the purpose.  Going back to my many years with AX25 packet, it seemed
like a good idea to put a TX timeout timer on it, in case something gets
stuck it won't hog a channel indefinitely.  The next question is what
is the worst case expected maximum duration of a legit jpskmail transmission
in any or the slower modes, so the timer can be set to something longer
than that time.  Any numbers available there?

I also just noticed a short silent gap between the RSID and the PSK carrier,
and I would want the interface to remain in TX for that short interval.
Is that time period known, and does it vary with mode?  Or maybe it
would be better to use the ptt tone although it would add a third isolation
transformer to the circuit?

73 de Bob WB5AOH
Austin TX.


Other related posts:

  • » [pskmail] fldigi v x-76 testing - Bob Morgan