ADALM-PLUTO (new DATV SDR)

[g0mjw]

Hi Jean-Pierre,

I don't see quite the same issue with the pluto - but its not as good as it ought to be. I am using Express 1.25p5 so be nice if you can improve matters.

Could you also do something about trapping errors as I end up with a screen full of error windows at times, all asking me to press OK, when something goes wrong. Only option is task manager.

Mike

[F6DZP]

Charles,

(About the TS issue, I have written a message about one year ago.)

About the PLUTO, I have always the same strange clouds, even at SR2200:

PLUTO_437MHz_SR2200_8PSK.jpg (430.34 KiB) Viewed 4773 times

I have made other tests with Evariste thsi aftenoon, same result.

@ Mike

What is the NIM used? I know that a Sharp or Eardatek don't show with the same amplification these kind of anomalies.
So I have indicated that it is what I obtain using a FTS-4334.

Could you also do something about trapping errors as I end up with a screen full of error windows at times, all asking me to press OK, when something goes wrong. Only option is task manager.

I suppose that it is when you are using Minitioune.
It is difficult to fix if I have not a more complete description and screen capture.

Jean-Pierre F6DZP

[G4GUO]

Jean-Pierre,

I am well aware of the TS issue. I have been trying to fix it for 5 years and have been unable to.
There is nothing more I can do.

- Charles

[F6DZP]

Charles,

Your software is fantastic about the different modulation it can do and the wonderfull result it gives.
The DATV express shows me that every day and I suppose that the Pluto and LimeSDRmini will give the same result in few months.

About the TS, it is also important to find a solution because, with your software, these 3 hardwares will give us the best universal tools.

Evariste is working to add an UDP input, this could help us to have some solution using external software or graph for producing the TS.
Perhaps in the future he will have time to fix the issue inside the software.

Well, we are going in the right way.
Thank you again for all you have already done.

73
Jean-Pierre

[G4GUO]

Hello Jean-Pierre,

Yes I am more of an expert on modulation than I am on transport streams.

I just read this document which helps my knowledge a bit more
http://www.img.lx.it.pt/~fp/cav/Additio ... erview.pdf

I have been trying to fix this issue but the problem I have is getting a good
timing reference. Currently the PCR is generated from the DATV-Express symbol rate.
The Audio PTS from the sound capture device and the Video PTS/DTS from the video frame rate
and because they don't use a common clock they drift apart. I can see it happening when
I pass the transport stream through the Harris Analyser software which displays the
various internal buffer usage sizes. It starts off with under-runs then after a few seconds
I get over runs. If I get the video clock stable the sound starts to either over/under run
and if I get the audio right the video goes wrong it is very frustrating.

- Charles

[G4GUO]

Jean-Pierre,

As far as the constellation is concerned it is difficult to tell what the issue is
but I know there is an issue because when I measure the EVM it is far worse
than Express. The Pluto specs indicate it should have a much better EVM/MER than
Express has. I am using EVM as that is what my Agilent test gear outputs.

The only difference between Express and Pluto is the raised cosine filters.
Originally I used an application on the internet to generate the filter coefficients
but the application is no longer available. Now I calculate the coefficients myself
using some maths I got from a NASA website. I suspect the issue may lie there.
When Evariste used my RRC code to generate a filter for the LimeSDR it did
not generate the correct filter shape.

However I would not expect to see the type of error that is showing up on the constellation.
Maybe the equaliser is trying to correct the error and that is masking what is actually happening.

- Charles

[g0mjw]

.

[g0mjw]

Hi Charles,

Hope you manage to fix the timing issue. There must be a solution. Maybe the problem is trying to satisfy too many constraints and something needs to agree to give.

The modulation issue with the pluto looks quite similar to phase jitter or AM/PM. Could it be the phase noise or other noise? If you got rid of all filtering, the constellation should be perfect? Maybe looking at the constellation with and without filtering would help identify where this is coming from. Than again, does it matter? An MER of 36dB is like a titanium kettle. Somewhat over-specified for the task at hand.

Gnuradio has some filtering code that might be useful. Maybe I might also make my usual point about integer overflow, though I don't think its relevant this time.

Mike

[g0mjw]

Charles,


@ Mike

What is the NIM used? I know that a Sharp or Eardatek don't show with the same amplification these kind of anomalies.
So I have indicated that it is what I obtain using a FTS-4334.

Could you also do something about trapping errors as I end up with a screen full of error windows at times, all asking me to press OK, when something goes wrong. Only option is task manager.

I suppose that it is when you are using Minitioune.
It is difficult to fix if I have not a more complete description and screen capture.

Jean-Pierre F6DZP

I am using the Serit with the low noise amplifier. I have not tried my older sharp in a while but I may get it out to see if there is a difference.

Unfortunately the crashing locks up the PC. But what happens is a dialog box pops up saying something like "floating point division by zero" - like this but this is just a random image from google and not it.

http://community.pmail.com/forums/stora ... pmail1.jpg

Then another pops up on top of it slightly offset and before you know it the screen is completely full of them all overlapping. CTRL+ALT+DEL and killing the process is the only way to recover.

Mike

PS - something similar to this, which was achieved by cutting the power to the tuner - but with some other error causing it.

Clipboard01.jpg (438.45 KiB) Viewed 4726 times

[f1faq]

Hi,

Seems there is a 1.25p5 version of Express ?
What is the difference with 1.25p4 I got from DATV Express web site ?
How to get it to pursue my tests ?

73 - Christian
F1FAQ