Linrad to MAP65 IQ communication

[Henk]

Hi all,

Linrad 03.12 is working fine with my Perseus SDR.
MAP 65 IQ does not receive any data from Linrad.
Using a single computer with Win-XP.

After reading all the communication messages of the last year (a huge
pile...and very confusing) I came to the following conclusion and
settings :
send raw data in 16 bit format
IP 127.1.1.0 (comment of Riein, W6SZ in which he announces succesfull
communication),
Port 50020 (comment of Joe, K1JT : MAP IQ listens on 50024)

These settings did no provide positive results.
I did try 127.0.0.1 and port 50000 combinations with above settings,
also with no result.

In Linrad port and IP are configurable. It is unclear to me how to
check the IP and port settings of MAP IQ.
Most probably I am missing something else as well.

Is there any "manual" or "step-by-step description" how to connect the
two?
That would save some mail/questions.

Thanks for any advice

73 Henk

PS: I can not use the "Linrad / MAP package" on the site of Joe
because that Linrad 03.05 version does not work with Perseus.

[Leif Asbrink]

Hi Henk,

> Linrad 03.12 is working fine with my Perseus SDR.
> MAP 65 IQ does not receive any data from Linrad.
> Using a single computer with Win-XP.
>
> After reading all the communication messages of the last year (a huge
> pile...and very confusing) I came to the following conclusion and
> settings :
> send raw data in 16 bit format

No, send timf2 data (blanker output) in 16 bit format.

> IP 127.1.1.0 (comment of Riein, W6SZ in which he announces succesfull
> communication),
> Port 50020 (comment of Joe, K1JT : MAP IQ listens on 50024)

When you set the base port to 50020 linrad will send timf2
data on port 50024 and that should be clear from the network setup screen.
(raw data in 16 bit would be on port 5020)

The noise blanker in Linrad can run on 16 bit integers as well as
on 32 bit float numbers. Which one is selected by the parameter
"First backward FFT version" set it to 1 for 16 bit arithmetics.
You must set "Enable second FFT to 1" to enable the blanker
and get a second time function (timf2)

Without second FFT, the baseband is derived from timf1 through
a back transformation of the first FFT.

> These settings did no provide positive results.
> I did try 127.0.0.1 and port 50000 combinations with above settings,
> also with no result.
>
> In Linrad port and IP are configurable. It is unclear to me how to
> check the IP and port settings of MAP IQ.
> Most probably I am missing something else as well.
>
> Is there any "manual" or "step-by-step description" how to connect the
> two?
> That would save some mail/questions.

The important parameters are in the par_xxx file where xxx
is wcw, cw, hsms or ssb whatever mode you want to use for
running MAP65IQ.

> PS: I can not use the "Linrad / MAP package" on the site of Joe
> because that Linrad 03.05 version does not work with Perseus.

You may copy the set of par_* files from Joes 03-05 package
and use them in the latest Linrad version.

The you will have to go through the global parameters setup
(to configure the Perseus) and you will have to go through the
network setup because the par_network has more entries now.
You may use a text editor to make the new par_network the same as the
old one for those entries that are present in the old file.

You may also compare your old and new par_userint files to see if
there are any differences that you do not understand.

73

Leif / SM5BSZ

[henk]

Hi Leif and all,

Thanks for the info. I (finally) got things (almost) up-and-running on the
same PC.
Data is "seen" by MAP65

However there is still a Linrad errormessage which I can not get rid off:

rx A/D speed error: 190452Hz (nominal 95238Hz)
buffer error (fft3) + counter
After a while this changes to
buffer EDA sync error

The S-meter is full-scale (all red) so my assumption is that it has to do
with input signals.

Changing first FFT amplitude from 1000 to 10 does change the S-meter a bit
(still above 9 though) but does not cure the errors.

If someone can point me in the right direction that would be much
appreciated.

73 Henk
PA0C

PS
What did I do to get Linrad en MAP connected:

A short description, maybe usefulll for others. The following was done:

installation of the combined MAP/Linrad package from Joe's site
move of all contents of the Linrad directory to a temporary directory
(linrad old)

installation in the (now empty) directory Linrad of :
Linrad 03.12 (site of Leif)
the latest .sbs files and perseususb.dll (site of Microtelecom)
the palir-01.dll (site of Leif)

Start Linrad and install the perseus as rx, using U
install the audio output.

Check if Linrad is working as a hf receiver, producing AF output

set the following in Network (N)
1 50024 (50020 on screen)
2 127.0.0.1
8 On
rest = off

Stop Linrad

Compare the following files from the MAP65/Linrad directory and the Linrad
old directory
par_network
par_ssb
par_userint

and make all parameter setting between brackets of thes files in the Linrad
directory the same as in the Linrad old files.

Start Linrad, start Map65. MAP should now receive data.

[henk]

Hi all,

It is my intention to use two SDR chains in an X-pol system and Linrad +
MAP65.
Now I have the possibility to get an SDR-IQ next to my Perseus SDR.

Because I could not find any info I have the following questions:
Is Linrad capable of using two different types of SDR's at the same time.
Does Linrad "output" the data of these two rx-es in such a way that MAP65
does not even notice that two different rx-es are being used.

My guess would be that due to different bandwidths, sampling rates, etc. it
is not possible, but you'll never know...
In case it is possible, are there any restrictions in comparison to the
normal same-type-of-rx set-up.

Thanks for the info
Henk PA0C

[Leif Asbrink]

Hi Henk,

> It is my intention to use two SDR chains in an X-pol system and Linrad +
> MAP65.
> Now I have the possibility to get an SDR-IQ next to my Perseus SDR.
>
> Because I could not find any info I have the following questions:
> Is Linrad capable of using two different types of SDR's at the same time.

NO.

You may run two instances of Linrad, one on each hardware, and then
run two instances of MAP65. In case the signal is aligned with one
or the other of your antennas you would have full sensitivity.
At times when the signal is right between with 50% of the energy
in each polarization you would loose 3 dB.

> Does Linrad "output" the data of these two rx-es in such a way that MAP65
> does not even notice that two different rx-es are being used.

No, one instance of Linrad can only read from one hardware.

> My guess would be that due to different bandwidths, sampling rates, etc. it
> is not possible, but you'll never know...

You can set both units to 95.2380952 kHz. You can also calibrate Linrad
for them to give identical frequency responses, but the sampling clocks would
not be locked to each other. You could fix that e.g. by locking both
units to a GPS standard. Still the phase would be undefined because the
digital decimation would not be synchronized, but that would not affect
sensitivity, you would just not be able to evaluate the true polarization.

In practical terms. It will be far too difficult. Wait for a while. Two
channel SDRs will become available. There is the WSE converter chain since
many years, but new systems based on VHF sampling will surely become available.

> In case it is possible, are there any restrictions in comparison to the
> normal same-type-of-rx set-up.

???????????
I do not understand this question.

73

Leif

[Leif Asbrink]

Hi Henk,

> Thanks for the info. I (finally) got things (almost) up-and-running on the
> same PC.
> Data is "seen" by MAP65
>
> However there is still a Linrad errormessage which I can not get rid off:
>
> rx A/D speed error: 190452Hz (nominal 95238Hz)
> buffer error (fft3) + counter
> After a while this changes to
> buffer EDA sync error

These are fatal errors. Try the "U = A/D and D/A setup for Rx"
and select 95.2 kHz for the Perseus. Then, after exit to the main menu,
make sure that you save the settings by pressing "W" ON THE MAIN MENU.

Three different files are involved: par_userint, par_perseus and par_ssb
When doing manual changes in those text files you could introduce some
inconsistency that leads to the errors you observe.

Please pack all the par* files into a zip file and send to me off list.
I will probably get the same problems when using those files and perhaps
I can add some error exit that would give an appropriate error message.

> The S-meter is full-scale (all red) so my assumption is that it has to do
> with input signals.

With the above errors you should not expect anything top work...

73

Leif / SM5BSZ

[Earl Shaffer]

Hi Henk.

 

There are a number of SDR's that use a sound card like the Delta 44.

The Softrock series is one example.

It should be an easy matter to use the same local oscillators to drive

both channels. I am using a home brew scaled down version of the

WSE converters and this is what I do. This solution gives full polarity

resolution.

 

" In case it is possible, are there any restrictions in comparison to the
> normal same-type-of-rx set-up."

 

For full polarization resolution, the only posible solution is for both channels

to be using the same equipment. Both channels must be in phase.

It's a little bit like the problem of listening to the same SSB signal in

two receivers. You will hear the signals go through zero beat and move

in and out of phase, but never stablize in phase with each other.

The 4 channel output of a Delta 44 sound card can be input to one instance

of Linrad to give two channel (V/H) operation with I and Q output from the

SDR.

 

73, Jim Shaffer, WB9UWA.

[Joe Barger]

From the SDR-IQ Yahoo Group:

A few people have asked what new radios we are working on. We have two amateur radios in the works. 

 

netSDR is a high performance networked radio that will replace the SDR-14.

 

The SDR-X2 is a dual RX synchronous receiver. It is TCP/IP based so it can be remote. It can be used for adaptive polarity, interferometry, diversity reception, etc. In addition, you can synchronize multiple SDR-X2s at one location or remotelly using GPS. This allows the building of huge multireceive arrays. The price will be very competitive.

 

 

Pieter