Frequency correction tables for WSE
Leif Asbrink
> antenna ports, the frequency shown on linrad screen is not the same.
> For example, if I inject a -100 dBm CW signal at 144.12900, Linrad
> shows the frequency of the signal to be at 144.13065, Is this an
> error in linrad software or in the WSE oscillator?
The error is 1.065 kHz and it is the summed error of all WSE oscillators
(as well as the Delta44 oscillator.)
The WSE units are designed for extremely low sideband noise, not for
frequency stability. There is no temperature compensation so you should
be able to see the frequency drift easily.
See fig 3 here:
http://www.sm5bsz.com/linuxdsp/rx10700/osc.htm
and figs 3 and 4 here:
http://www.sm5bsz.com/linuxdsp/rx70/osc.htm
The WSE units require stable temperature for good stability
unless you add a PLL to each unit.
> My signal source
> is from a very accurate signal generator whose accuracy has been
> verified independently.
The WSE units are designed to allow phase locking from a reference
source. They all oscillate on a multiple of 100 kHz so a simple
PLL that samples at 100 kHz could be used. I have not designed such
a unit because I have not had the need for it myself.
Since you have an accurate signal source you should change the
frequency error tables in wse_sdrxx.c. They look like this:
int rx2500_fqcorr=0;
int rx10700_fqcorr[4]={0,-100,180,-60};
int rx70_fqcorr[5]={-230,-100,-100,-100,-300};
int rx144_fqcorr[4]={0,-50,-160,-130};
These were the values needed here to make my system display
correct frequencies half a year ago. There is no user interface
within Linrad to set these corrections.
Once you installed the compilers and compiled from source once
you will find that a re-compile with a changed wse_sdrxx.c is
immediate.
First inject a signal near 2.5 MHz into rx2500. Note the error in Hz,
change rx2500_fqcorr accordingly, recompile and verify.
You should find that the error now is near zero or twice as large.
In the latter case, change sign.
Then inject a signal near 10.675 into rx10700, set the center frequency to
10.675 and check the error. Change the first value (0) in rx10700_fqcorr
and re-compile.
Then repeat at 10.7, add or subtract your error to -100 and put in
as the second number. repeat at 10.725 and 10.750
Then enject near 70.0, 70.1, 70.2 and 70.3 into rx70 to calibrate it.
Finally 144.0, 144.5, 145.0 and 145.5 into RX144.
I introduced the error compensation fairly recently and it
is not documented anywhere (yet) as I recall.
73
Leif
Mario A. Natali, I0NAA
I am glad to communicate that I was fianlly able to set-up and run Linrad
under UBUNTU 8.10 !!!
I have also the full procedure . I tried few times from scratch and
everyting works well. The procedure is really foolproof and do not require
much LINUX know how.
Now the fun part with parametres setting starts !
73 de i0naa, Mario
f6dro
could U please share your procedure?
tnx and 73
Dom
Mario A. Natali, I0NAA
please find the procedure attached.
I also attached the libvg.sonfig file that, in my case, was modified to
avoid one error thanks to the inputs of Leif.
The installation refers to SDR-IQ ; if you do not want SR just skip make
sdr14
Please let me know if this is clear !!!
Best 73 de i0naa, Mario
f6dro
73
Dom
Mario A. Natali, I0NAA
Forgot to mention ( .... and to report on the procedure !! ) before
launching ."/xlinrad "you should put "sudo". As an alternative if you do not
like to type "sudo" all the time, you can put " sudo -s" and you become root
and you will be able to do everything .... of course need to be careful !!
Let me know if it will work !!
73 and good week end.
I0NAA, Mario
----- Original Message -----
From: "f6dro" <f6...@wanadoo.fr>
To: <lin...@googlegroups.com>
Sent: Saturday, March 21, 2009 12:03 PM
Subject: [Linrad] Re: Linrad set-up under UBUNTU 8.10
>
Gabriel - EA6VQ
antivirus is blocking the data exchange between Linrad and MAP65-IQ after 40
seconds. I mean that from second 40 to the end of the minute MAP65-IQ
shows the red "No data" message. At minute 0 it starts receiving data again
for another 40 seconds and so on....
I tried adding both Linrad and MAP65-IQ to the exclusion list of the
antivirus but it still fails.
So far disabling the antivirus has been the only way to avoid the problem,
but I don't like this solution at all becasue I use the same computer for
connecting to Internet and Kaspersky is also my firewall program....
Any ideas?
73. Gabriel - EA6VQ
_________________________________________________________
Web-Site: HTTP://www.vhfdx.net <HTTP://www.vhfdx.net>
VQLog 3.1 (build 56): HTTP://www.vqlog.com <HTTP://www.vqlog.com>
Real-Time Propagation maps: HTTP://www.vhfdx.net/spots/map.php
<http://www.vhfdx.net/spots/map.php>
_________________________________________________________
Gabriel - EA6VQ
It was as simple at defining a new "localhost" network with the IP 127.0.0.1
and add it to the "trusted networks"
73. Gabriel - EA6VQ
_________________________________________________________
Web-Site: HTTP://www.vhfdx.net <HTTP://www.vhfdx.net>
VQLog 3.1 (build 56): HTTP://www.vqlog.com <HTTP://www.vqlog.com>
Real-Time Propagation maps: HTTP://www.vhfdx.net/spots/map.php
<http://www.vhfdx.net/spots/map.php>
_________________________________________________________
> De: lin...@googlegroups.com [mailto:lin...@googlegroups.com]En nombre de
> Gabriel - EA6VQ
> Enviado el: lunes, 23 de marzo de 2009 18:24
> Para: lin...@googlegroups.com
> Asunto: [Linrad] MAP65-IQ and Kaspersky antivirus
Joe Taylor
When running programs with real-time requirements like the
Linrad/MAP65 combination, you do not want other programs
trying to control their inter-communication without your
knowledge or permission. I know nothing about the Kasperksy
program, but I strongly recommend disabling any virus and
spyware protection, etc., from your Linrad/MAP65 computer.
Who knows what these programs may try to do?
Disconnecting the computer from the outside world (at least
while you are using Linrad/MAP65-IQ) is much better
protection against malware, anyway. :-)
You are probably already far beyond needing it, but I have
just posted a draft "Quick Start Guide" for MAP65-IQ on the
WSJT home page. The direct link is
http://physics.princeton.edu/pulsar/K1JT/MAP65-IQ_Quick_Start.htm
Request to anyone who has been using MAP65-IQ:
1. Please upgrade to r1102, if you have not already done so.
2. Post your comments and suggestions on this reflector.
3. What problems did you have getting r1102 to work?
4. What else should be included in the Quick Start Guide?
-- 73, Joe, K1JT
Joe Taylor
If it works well, great! Just watch out for times when the
virus scanner may decide to make heavy demands on your CPU,
when Linrad needs the CPU...
-- Joe, K1JT
Jacques RAMBAUD
have now to find a solution for the hardware.
I am considering to use a downconverter 144 to 28 MHz and then I see several
people using an SDR-IQ or Perseus to generate the IQ signals , How about
using instead a Softrock V9 ??
73
Jacques F6BKI
rei...@ix.netcom.com
Hi Jacques,
I am using Linrad with a V9 with Si570 LO and pre-selector ahead the V9
and it is working great. I also have been using SDR I/Q though not with
Linrad/Winrad at another station and I like the Softrock at least as much
as the SDR I/Q.
Using a preselector helps with the SI570 spurious signals The SI 570 LO is
not as clean as one would like. ( compared to the x-tal LO )
The softrock requires more care with avoiding ground loops etc
then the SDR I/Q. The softrock apprach also requires a soundcard.
( Delta44 in my case )
As a stand alone 10 meter receiver, this softrock has the same
sensitivity on 10 meter as a Kenwood TS-850.
I have also been using a V7 version for 28 Mhz IF
73 Rein W6SZ
Leif Asbrink
Using Perseus is not yet possible, but hopefully it
will be tomorrow:-)
The Softrock or any other direct conversion radio will be fine,
but you will have to use MAP65, not MAP65-IQ. (And set the
soundcard to sample at 96 kHz)
73 Leif / SM5BSZ
f6dro
in fact , to obtain full adavanrage what we need is a design able to use
fully Linrad , I mean V and H simultaneously and being able to get optimum
signals any time. WSE RX is perfectly OK , but a bit too expensive for the
average ham.We need a " dual channel system for the poor ham"
73
Dom
Edward Cole
It is true that MAP65IQ only decodes a single polarity, but if you
have a dual-pol antenna system, you can switch pol on the nezt
sequence and capture any signals that are coming in on that
polarity. I would guess that one would sample the spectrum with
MAP65 every 5-10 minutes else you would have information
overload. Activity would not change that rapidly that one needed to
capture everything in one minute. Obviously, not as sophisticated as
WSE Rx into dual channel Linrad, but I think it is workable.
I have a dual-pol eme station but will run my 2m Rx signal into a
splitter feeding my FT-847 <--> JT65 (computer#1) and simultaneously
into a DEMI 144/28 xvtr feeding my SDR-IQ --> Linrad/MAP65iq
(computer#2). So I will be able to watch the eme subband on the
selected pol, switching pol as desired.
***********************************************************
73, Ed - KL7UW BP40iq, 6m - 3cm
144-EME: FT-847, mgf-1801, 4x-xp20, 8877-600w
1296-EME: DEMI-Xvtr, 0.30 dBNF, 4.9m dish, 60/300W (not QRV)
http://www.kl7uw.com AK VHF-Up Group
NA Rep. for DUBUS: dubu...@hotmail.com
***********************************************************
Leif Asbrink
> >in fact , to obtain full adavanrage what we need is a design able to use
> >fully Linrad , I mean V and H simultaneously and being able to get optimum
> >signals any time. WSE RX is perfectly OK , but a bit too expensive for the
> >average ham.We need a " dual channel system for the poor ham"
It should be trivial to modify the Softrock so two of them
use the same LO. The second thing needed would be one more
144 to HF converter that is modified to use the same LO as
the other 144 to HF converter.
Why not phase-locking two 144 MHz converters as well as two
Softrocks to a common reference oscillator with very good
stability. That would solve two problems at the same time:-)
> It is true that MAP65IQ only decodes a single polarity, but if you
> have a dual-pol antenna system, you can switch pol on the nezt
> sequence and capture any signals that are coming in on that
> polarity. I would guess that one would sample the spectrum with
> MAP65 every 5-10 minutes else you would have information
> overload. Activity would not change that rapidly that one needed to
> capture everything in one minute. Obviously, not as sophisticated as
> WSE Rx into dual channel Linrad, but I think it is workable.
Yes. The problem is that you will loose 3dB at times when the
signal level is the same in both polarisations. The loss is
a cosine function with -3dB at 45 degrees. Someone might compute
the percentage of time you will loose more than 1 dB. I would guess
it comes out as more than 50%.
In a contest the time loss will matter. Weak stations that answer
your CQ may have to be much more persistant when you listen
in the wrong polarisation 50% of the time and sometimes loose
3 dB.
There is another thing. In case your noise floor is somewhat
degraded by high rates of interference pulses, you might find
that the noise blanker becomes more efficient when it can
use the information from two antennas.
73
Leif / SM5BSZ
Edward Cole
Yes, the ability of Linrad to receive H/V inputs and display to the
resultant pol is an real advantage. I may end up with two 144/28
convertors that may be PLL slaved to my 10-MHz Rubidium Ref. Too bad
that the SDR-IQ appears to be too difficult to lock to an external
ref., else I could obtain a second unit. But if I only run one
SDR-IQ behind Linrad with full dual-pol input I would be able to see
display of all signals, but still have the 3-dB penalty for decoding
JT65 or MAP65iq when pol=45/135.
My real experience is that doesn't occur 50% of the time. However,
when Faraday is very active on 2m considerable percent of the time it is.
I am considering obtaining a K3 with dual Rx. The problem is
obtaining wideband output as the final SDR baseband is
18-KHz. Taping the 1st IF of 8.2 MHz would work but this still
leaves one with the need for a dual channel SDR-IQ or such. Probably
simpler to go straight from 28-MHz into a dual SDR like you
suggest. Or i could be satisfied only sampling 18-KHz of spectrum vs
95-KHz that MAP65 can do.
More study needed ;-)
73, Ed - KL7UW
Jeremy Alexander
Hi Ed,
If one might be contemplating spending even $2000 towards good rx on 2m, Leif's WSE converters and a good sound card would be the place to drop your money, as then LinRad, MAP65 and plenty of support exists already. Plus the simplicity of hardware, you then also have just One software to deal with as far as the receivers.
Doing this myself I had a Huge learning curve, but once the stitches healed, I couldn't think of using anything else currently available. Plus - oh my gosh the cw ability with these things.
Jeremy w7eme
--- On Thu, 3/26/09, Edward Cole <kl...@acsalaska.net> wrote:
Gabriel - EA6VQ
What's the right way to set QRG of my TS-790 from MAP65-IQ?
I have no problem to write the code to set the QRG via CAT, but I don't know
what is the best way to get the operating QRG from MAP65-IQ.
I notice that the operating frequency is in the file azel.dat, so might be I
could read that file every second or so, but I also see a "Set Tx Freq"
button, that I don't what it is used for.
I would appreciate your advice.
73. Gabriel - EA6VQ
Joe Taylor
Your idea is exactly the right thing to do. I have an external program
coded for my own station that takes care of tracking the moon, selecting
the desired band (from 1.8 MHz to 2.3 GHz), and setting the Tx
frequency. This program reads the file azel.dat to get sun/moon
pointing information and the specified QSO frequency.
When you click the MAP65 button "Set Tx Freq", the tx frequency
(actually, the offset from band edge in kHz) is written to azel.dat.
The number following the frequency is set to 1; one second later it
reverts to 0. The external program watches for the second number to
change from 0 to 1; when it does to, the frequency is set to the
transceiver.
-- 73, Joe, K1JT
Kleibe
linrad but the average hardware is not often heard in the linrad
reflector. I think something like UHFSDR from W6DHW
http://wb6dhw.com/Under_Development.html should be a good start point
or something new.
73
Kleibe
PP2KR
KD7HGL (Josh Logan)
I have also been watching the development of this radio and would like
to see it work well with Linrad.
Would it support 2 radios with one si570?
73, KD7HGL
Edward Cole
>Hi Ed,
>
>If one might be contemplating spending even $2000 towards good rx on
>2m, Leif's WSE converters and a good sound card would be the place
>to drop your money, as then LinRad, MAP65 and plenty of support
>exists already. Plus the simplicity of hardware, you then also have
>just One software to deal with as far as the receivers.
>
>Doing this myself I had a Huge learning curve, but once the stitches
>healed, I couldn't think of using anything else currently available.
>Plus - oh my gosh the cw ability with these things.
>
>Jeremy w7eme
Truth is that I probably will not invest any money this year. I can
proceed with what I have which is running my standard H/V switching
JT65 2m station (FT-847) and splitting off the 2m Rx signal to the
DEMI 144/28+SDR-IQ and running Linrad+MAP65iq. Not the ultimate
solution but should make random JT65 operating work pretty good.
I have other needs than just the ultimate eme Rx. The Flex-5000 and
K3 are pretty much the ultimate HF radios (but expensive). I am
preparing to get on 160m and 600m. The K3 attracts from the small
package that can go portable for mw. I may wait to see what develops
in the SDR area. Eventually, being able to run dual-pol into Linrad
is a 2m-eme goal.
On 1296 my single Rx system will work fine since CP is used. I will
be able to run CW/SSB/JT65 with Linrad and MAP65iq.
Waiting may be the smart move?
73, Ed - KL7UW