Re: [USRP-users] Sampler clock common or not?

[Koyel Das (Vehere) via USRP-users]

Hi Marcus,



I am looking at the following link


https://files.ettus.com/schematics/x300/x3xx.pdf


and as you said same clock is being used by both the boards. NI has asked for 
some time to find it out.


Regards,

Koyel


From: Koyel Das (Vehere)
Sent: Monday, September 17, 2018 12:57:29 PM
To: Marcus D. Leech; Koyel Das (Vehere)
Subject: Re: [USRP-users] Sampler clock common or not?


Hi Marcus,


I spoke with NI and they are saying that in USRP 2954 R there are two LOs for 
two daughter boards which are not locked to any master oscillator. They are not 
phase locked but we can provide a reference clock if we want them to run 
coherently. He said in 2955 R LO sharing is there so we can share single LO for 
4 daughter boards.


Regards,

Koyel


From: USRP-users  on behalf of Koyel Das 
(Vehere) via USRP-users 
Sent: Monday, September 17, 2018 12:12:32 PM
To: Marcus D. Leech; 'USRP-users@lists.ettus.com'
Subject: Re: [USRP-users] Sampler clock common or not?


Hi,


I don't have a splitter currently. So to use it I need to wait until it arrives.


I din't find uhd_fft in gnuradio. I am using gnuradio-companion.


However, I see similar issues faced by some other people:


https://lists.gnu.org/archive/html/discuss-gnuradio/2008-04/msg00029.html


Regards,

Koyel



RE: [Discuss-gnuradio] Compare Phase 
block
lists.gnu.org
My VOR receiver is almost there, but not quite... My code is below. I've 
demodulated the first 30 Hz signal from AM, and the second from the FM 
subcarrier.





From: Marcus D. Leech 
Sent: Sunday, September 16, 2018 9:28:55 AM
To: Koyel Das (Vehere); 'USRP-users@lists.ettus.com'
Subject: Re: [USRP-users] Sampler clock common or not?

On 09/15/2018 11:54 PM, Koyel Das (Vehere) wrote:
I just forwarded in another thread the processing I am doing. Have you received 
it?

Regards,
Koyel
Yes, but I'd rather not have to analyse your MATLAB code in detail.

You can use the --phase-relations  option on the uhd_fft utility for doing 
two-channel phase-coherence measurements in real-time.

Unless something is horribly broken, two UBX cards on the same underlying USRP 
platform WILL operate with mutual coherence, with
  some initial phase-offset, just due to the way synthesizers work.  But that 
phase offset will carry-forward in time, with only very minor
  phase noise perturbations just due to the physics of semiconductors, etc.



From: Marcus D. Leech 
Sent: Sunday, September 16, 2018 9:05:29 AM
To: Koyel Das (Vehere); 
'USRP-users@lists.ettus.com'
Subject: Re: [USRP-users] Sampler clock common or not?

On 09/15/2018 11:32 PM, Koyel Das (Vehere) wrote:
I don’t have access to the flowgraph currently but it’s the same flowgraph I 
sent earlier in my other previous email for 2955 R. I have UHD with two 
channels, each channel connected to a file sink. One channel is set to TX/RX, 
another to RX2. I am sampling at 1 msps. I am capturing data from two channels, 
storing in file and then reading them in matlab and cross correlating them to 
get phase difference but the phase difference for any particular frequency bin 
is showing deviations with time. Deviations are huge sometimes +-pi.

Regards,
Koyel
I would say that you're doing something wrong in the post-processing.

I routinely use USRPs of all types for interferometry in radio astronomy, where 
the long term phase noise has to be very low, or your
  data is rubbish.



From: Marcus D. Leech 
Sent: Sunday, September 16, 2018 8:48:26 AM
To: Koyel Das (Vehere); 
'USRP-users@lists.ettus.com'
Subject: Re: [USRP-users] Sampler clock common or not?

On 09/15/2018 10:38 PM, Koyel Das (Vehere) wrote:
Hi,

Thanks for the information. Do I have to set any property in the gnu radio uhd 
to phase lock the two oscillators to on board clock or they are by default 
locked to it?
Nothing special to do.  It happens automatically.

Can the external reference be gps? What settings need to be done in UHD to sync 
the two oscillators to gps?
There's a good knowledge-base article on this and many other topics here:

https://kb.ettus.com/Synchronization_and_MIMO_Capability_with_USRP_Devices



Actually I am receiving a stationary signal using two ports TX/RX and RX2 but I 
am not getting a constant phase difference with time. It can happen if two 
oscillators are not synced .
The starting phase between the two sides will not be anything in particular.

But after that, they'll have a constant phase difference.  What are you using 
to measure the phase different between the two channels.

How is the signal delivered to the two channels?

Can you share your Gnu Radio flow-graph?



Please help.
___

[USRP-users] custom E310 bitfile weird new error

[Jason Matusiak via USRP-users]

I had a system that worked pretty well for creating a new bitfile for my E310.  
I ran through all the steps again so I could make sure that my notes were OK, 
but now I get an error when I run a probe on my E310 using that bitfile.
 
I copied over the bitfile and replaced 
/home/root/localinstall/usr/share/uhd/images/usrp_e310_fpga_sg3.bit with it.
 
I source the env, then I run uhd_usrp_probe:
 
root@ettus-e3xx-sg3:~# uhd_usrp_probe 
[INFO] [UHD] linux; GNU C++ version 7.3.0; Boost_106400; 
UHD_4.0.0.rfnoc-devel-702-geec24d7b
[INFO] [E300] Loading FPGA image: 
/home/root/localinstall/usr/share/uhd/images/usrp_e310_fpga_sg3.bit...
[INFO] [E300] FPGA image loaded
[INFO] [E300] Initializing core control (global registers)...
 [INFO] [E300] Performing register loopback test... 
[INFO] [E300] Register loopback test passed
[INFO] [0/Radio_0] Initializing block control (NOC ID: 0x12AD1000)
[WARNING] [RFNOC] Can't find a block controller for key KeepOneInN, using 
default block controller!
[INFO] [0/KeepOneInN_0] Initializing block control (NOC ID: 0x0246)
[INFO] [0/FIR_0] Initializing block control (NOC ID: 0xF112)
[WARNING] [RFNOC] Can't find a block controller for key FIFO, using default 
block controller!
[INFO] [0/FIFO_0] Initializing block control (NOC ID: 0xF1F0)
[WARNING] [RFNOC] Can't find a block controller for key FIFO, using default 
block controller!
[INFO] [0/FIFO_1] Initializing block control (NOC ID: 0xF1F0)
[ERROR] [AD936X] CODEC loopback test failed! Expected: 0x7C2024A0 Received 
(TX/RX): 0x7C2024A0/0x
Error: RuntimeError: CODEC loopback test failed.
  
 this was built on my host by running the command: 
./uhd_image_builder.py keep_one_in_n fir_filter -d e310 -t E310_RFNOC_sg3 -m 4 
--fill-with-fifos
  
 So it appears to me that it built all the blocks I requested, but the CODEC 
loopback failure is a new one for me.  Since this build process worked fine for 
me the previous day, I have no idea why it wouldn't work a second time (unless 
it could be some sort of weird fpga timing issue?).

Extra info:
1 - The uhd hash is: eec24d7
2 - The fpga-src hash within uhd is: 9c8c2ba
3 - gnuradio's hash is (this shouldn't matter since I am not even running GR 
yet): e4acf4f
4 - gr-ettus' hash is: 51e8828
  
 ~Jason
___
USRP-users mailing list
USRP-users@lists.ettus.com
http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com

Re: [USRP-users] N310 timeout before streaming complete

[Rob Kossler via USRP-users]

Hi Michael,
Were these issues resolved in a recent commit (included in 3.13.0.3 RC)?
Rob

On Thu, Sep 6, 2018 at 2:12 PM Michael West  wrote:

> Update:  The issue is included on the head of the UHD-3.12 branch, but is
> not included in the v3.12.0.0 release.  So, be sure to checkout the
> v3.12.0.0 tag.
>
> The issue we replicated is the STREAM_CMD_NUM_SAMPS_AND_DONE.  The
> continuous streaming issue is still outstanding, but may be related.
>
> Regards,
> Michael
>
> On Thu, Sep 6, 2018 at 10:57 AM, Rob Kossler  wrote:
>
>> OK. I will give it a shot.  Which issue did you duplicate (the confusion
>> is my fault for discussing two different issues in the same thread): Issue
>> #1 (NUM_SAMPS_AND_DONE) or Issue #2 (CONTINUOUS)?
>>
>> Rob
>>
>> On Thu, Sep 6, 2018 at 1:52 PM Michael West 
>> wrote:
>>
>>> Hi Rob,
>>>
>>> Thank you once again.  We have reproduced the issue and are working on
>>> tracing down the root cause and fixing the issue.  We have confirmed that
>>> the issue exists on 3.13, but we have also confirmed it does not exist on
>>> 3.12.0.0.  So, you can use v3.12.0.0 in the meantime.  To be sure you are
>>> using the correct FPGA image for v3.12.0.0, be sure to run
>>> 'uhd_images_downloader --refetch'.
>>>
>>> Regards,
>>> Michael
>>>
>>> On Tue, Sep 4, 2018 at 3:21 PM, Rob Kossler via USRP-users <
>>> usrp-users@lists.ettus.com> wrote:
>>>
 Today, I confirmed that this issue exists on the head of 3.13 and 3.12,
 but not 3.11.  The example program provided with my previous post can be
 compiled with any of these versions.

 Rob

 On Fri, Aug 31, 2018 at 12:33 PM Rob Kossler  wrote:

> In this post and one other post, I mentioned two issues I am having
> related to N310 streaming:
>
>1. With STREAM_MODE_NUM_SAMPS_AND_DONE, sometimes I get a timeout
>prior to receiving the requested number of samples (this is the issue
>identified in this post). This may be simply dependent upon the number 
> of
>samples requested.
>2. With STREAM_MODE_START_CONTINUOUS, I get errors with repeated
>captures such that after several successful captures, I eventually get 
> a
>streaming timeout and all subsequent captures fail.
>
> So, turns out that this issue is bigger for me than I realized.  I had
> a bunch of trouble yesterday while doing some research experimentation. I
> had selected to go with X310 devices rather than N310 devices because of
> their relative maturity.  Today, I confirmed that my issues yesterday with
> the X310 are the same as I previously mentioned for the N310 (#2 above).
> So, perhaps it is an issue with UHD-3.13 (I did not check any other 
> branch).
>
> I modified the Ettus "txrx_loopback_to_file.cpp" code to include a
> "for loop of 50 iterations" and changed the streaming mode to be
> continuous.  The modified source is included as an attachment (a 'diff' of
> my code to the original with show the minor changes I made).  I attached a
> console log of the output messages when run with my X310 which shows both
> the command line parameters I used as well as the resulting errors.  Note
> that everything is going as expected through Iteration 5, but starting at
> Iteration 6, there is no end-of-burst (EOB) and starting at Iteration 8, a
> timeout occurs prior to receiving any samples.
>
> Please let me know if you have any questions.  This is a pretty big
> issue for me and will prevent me from using 3.13 until addressed.
>
> Rob
>
>
> On Fri, Aug 24, 2018 at 2:46 PM Rob Kossler  wrote:
>
>> Hi,
>> This post is perhaps a continuation of a previous post "Problems with
>> MPM 3.13 on N310", but I wanted to change the subject to reflect this
>> current issue.
>>
>> The issue is an Rx streaming timeout.  It can be easily duplicated
>> with a stock Ettus example.  Below you will find the console log which
>> includes the command line parameters.  Note the following:
>>
>>- I requested 31250 samples
>>- There is a error "Timeout while streaming" indicated at the end
>>- The final file size of 119340 indicates that 29835 samples were
>>received
>>- I do not have any reason to believe that the specific command
>>line arguments below are needed to duplicate the issue.  I just didn't
>>bother to try other ones.
>>
>> In the other thread, I mentioned that my Rx timeout issue had gone
>> away after switching to streaming mode STREAM_MODE_NUM_SAMPS_AND_DONE.
>> However, the issue below occurs when using that mode so it will be a
>> problem for me.
>>
>> Let me know if you have any questions.
>>
>> Rob
>>
>>
>> irisheyes9@irisheyes9-Z240-SFF:/media/SSD_RAID/multi_pol$
>> txrx_loopback_to_file --tx-args="addr=192.168.61.2"
>> --rx-args="addr=1

Re: [USRP-users] Sampler clock common or not?

[Marcus D. Leech via USRP-users]

On 09/17/2018 02:42 AM, Koyel Das (Vehere) wrote:


Hi,


I don't have a splitter currently. So to use it I need to wait until 
it arrives.



I din't find uhd_fft in gnuradio. I am using gnuradio-companion.


However, I see similar issues faced by some other people:


https://lists.gnu.org/archive/html/discuss-gnuradio/2008-04/msg00029.html


Regards,

Koyel

uhd_fft should be one of the command-line commands available to you if 
you have a proper install of Gnu Radio.






RE: [Discuss-gnuradio] Compare Phase block 


lists.gnu.org
My VOR receiver is almost there, but not quite... My code is below. 
I've demodulated the first 30 Hz signal from AM, and the second from 
the FM subcarrier.





*From:* Marcus D. Leech 
*Sent:* Sunday, September 16, 2018 9:28:55 AM
*To:* Koyel Das (Vehere); 'USRP-users@lists.ettus.com'
*Subject:* Re: [USRP-users] Sampler clock common or not?
On 09/15/2018 11:54 PM, Koyel Das (Vehere) wrote:
I just forwarded in another thread the processing I am doing. Have 
you received it?


Regards,
Koyel 

Yes, but I'd rather not have to analyse your MATLAB code in detail.

You can use the --phase-relations  option on the uhd_fft utility for 
doing two-channel phase-coherence measurements in real-time.


Unless something is horribly broken, two UBX cards on the same 
underlying USRP platform WILL operate with mutual coherence, with
  some initial phase-offset, just due to the way synthesizers work.  
But that phase offset will carry-forward in time, with only very minor
  phase noise perturbations just due to the physics of semiconductors, 
etc.





*From:* Marcus D. Leech  
*Sent:* Sunday, September 16, 2018 9:05:29 AM
*To:* Koyel Das (Vehere); 'USRP-users@lists.ettus.com 
'

*Subject:* Re: [USRP-users] Sampler clock common or not?
On 09/15/2018 11:32 PM, Koyel Das (Vehere) wrote:
I don’t have access to the flowgraph currently but it’s the same 
flowgraph I sent earlier in my other previous email for 2955 R. I 
have UHD with two channels, each channel connected to a file sink. 
One channel is set to TX/RX, another to RX2. I am sampling at 1 
msps. I am capturing data from two channels, storing in file and 
then reading them in matlab and cross correlating them to get phase 
difference but the phase difference for any particular frequency bin 
is showing deviations with time. Deviations are huge sometimes +-pi.


Regards,
Koyel 

I would say that you're doing something wrong in the post-processing.

I routinely use USRPs of all types for interferometry in radio 
astronomy, where the long term phase noise has to be very low, or your

  data is rubbish.




*From:* Marcus D. Leech  
*Sent:* Sunday, September 16, 2018 8:48:26 AM
*To:* Koyel Das (Vehere); 'USRP-users@lists.ettus.com 
'

*Subject:* Re: [USRP-users] Sampler clock common or not?
On 09/15/2018 10:38 PM, Koyel Das (Vehere) wrote:

Hi,

Thanks for the information. Do I have to set any property in the 
gnu radio uhd to phase lock the two oscillators to on board clock 
or they are by default locked to it?

Nothing special to do.  It happens automatically.


Can the external reference be gps? What settings need to be done in 
UHD to sync the two oscillators to gps?
There's a good knowledge-base article on this and many other topics 
here:


https://kb.ettus.com/Synchronization_and_MIMO_Capability_with_USRP_Devices




Actually I am receiving a stationary signal using two ports TX/RX 
and RX2 but I am not getting a constant phase difference with time. 
It can happen if two oscillators are not synced .
The starting phase between the two sides will not be anything in 
particular.


But after that, they'll have a constant phase difference.  What are 
you using to measure the phase different between the two channels.


How is the signal delivered to the two channels?

Can you share your Gnu Radio flow-graph?




Please help.

*From:* USRP-users  
 on behalf of Marcus D. 
Leech via USRP-users  


*Sent:* Sunday, September 16, 2018 5:10:23 AM
*To:* usrp-users@lists.ettus.com 
*Subject:* Re: [USRP-users] Sampler clock common or not?
On 09/15/2018 01:35 AM, Koyel Das (Vehere) via USRP-users wrote:

Hi,

I am using USRP 2954R. I have connected one antenna to RX2 of RF0 
and another antenna to TX/RX of RF1. Both are used for receiving. 
Can someone tell me if RX2 of RF0 and TX/RX of RF1 are phase 
locked that is the signal received by the two have a consta

Re: [USRP-users] Sampler clock common or not?

[Marcus D. Leech via USRP-users]

On 09/17/2018 06:31 AM, Koyel Das (Vehere) wrote:


Hi Marcus,



I am looking at the following link


https://files.ettus.com/schematics/x300/x3xx.pdf


and as you said same clock is being used by both the boards. NI has 
asked for some time to find it out.



Regards,

Koyel

You might try disabling I/Q corrections in your flow-graph, and see if 
that helps.





*From:* Koyel Das (Vehere)
*Sent:* Monday, September 17, 2018 12:57:29 PM
*To:* Marcus D. Leech; Koyel Das (Vehere)
*Subject:* Re: [USRP-users] Sampler clock common or not?

Hi Marcus,


I spoke with NI and they are saying that in USRP 2954 R there are two 
LOs for two daughter boards which are not locked to any master 
oscillator. They are not phase locked but we can provide a reference 
clock if we want them to run coherently. He said in 2955 R LO sharing 
is there so we can share single LO for 4 daughter boards.



Regards,

Koyel


*From:* USRP-users  on behalf of 
Koyel Das (Vehere) via USRP-users 

*Sent:* Monday, September 17, 2018 12:12:32 PM
*To:* Marcus D. Leech; 'USRP-users@lists.ettus.com'
*Subject:* Re: [USRP-users] Sampler clock common or not?

Hi,


I don't have a splitter currently. So to use it I need to wait until 
it arrives.



I din't find uhd_fft in gnuradio. I am using gnuradio-companion.


However, I see similar issues faced by some other people:


https://lists.gnu.org/archive/html/discuss-gnuradio/2008-04/msg00029.html


Regards,

Koyel



RE: [Discuss-gnuradio] Compare Phase block 


lists.gnu.org
My VOR receiver is almost there, but not quite... My code is below. 
I've demodulated the first 30 Hz signal from AM, and the second from 
the FM subcarrier.





*From:* Marcus D. Leech 
*Sent:* Sunday, September 16, 2018 9:28:55 AM
*To:* Koyel Das (Vehere); 'USRP-users@lists.ettus.com'
*Subject:* Re: [USRP-users] Sampler clock common or not?
On 09/15/2018 11:54 PM, Koyel Das (Vehere) wrote:
I just forwarded in another thread the processing I am doing. Have 
you received it?


Regards,
Koyel 

Yes, but I'd rather not have to analyse your MATLAB code in detail.

You can use the --phase-relations  option on the uhd_fft utility for 
doing two-channel phase-coherence measurements in real-time.


Unless something is horribly broken, two UBX cards on the same 
underlying USRP platform WILL operate with mutual coherence, with
  some initial phase-offset, just due to the way synthesizers work.  
But that phase offset will carry-forward in time, with only very minor
  phase noise perturbations just due to the physics of semiconductors, 
etc.





*From:* Marcus D. Leech  
*Sent:* Sunday, September 16, 2018 9:05:29 AM
*To:* Koyel Das (Vehere); 'USRP-users@lists.ettus.com 
'

*Subject:* Re: [USRP-users] Sampler clock common or not?
On 09/15/2018 11:32 PM, Koyel Das (Vehere) wrote:
I don’t have access to the flowgraph currently but it’s the same 
flowgraph I sent earlier in my other previous email for 2955 R. I 
have UHD with two channels, each channel connected to a file sink. 
One channel is set to TX/RX, another to RX2. I am sampling at 1 
msps. I am capturing data from two channels, storing in file and 
then reading them in matlab and cross correlating them to get phase 
difference but the phase difference for any particular frequency bin 
is showing deviations with time. Deviations are huge sometimes +-pi.


Regards,
Koyel 

I would say that you're doing something wrong in the post-processing.

I routinely use USRPs of all types for interferometry in radio 
astronomy, where the long term phase noise has to be very low, or your

  data is rubbish.




*From:* Marcus D. Leech  
*Sent:* Sunday, September 16, 2018 8:48:26 AM
*To:* Koyel Das (Vehere); 'USRP-users@lists.ettus.com 
'

*Subject:* Re: [USRP-users] Sampler clock common or not?
On 09/15/2018 10:38 PM, Koyel Das (Vehere) wrote:

Hi,

Thanks for the information. Do I have to set any property in the 
gnu radio uhd to phase lock the two oscillators to on board clock 
or they are by default locked to it?

Nothing special to do.  It happens automatically.


Can the external reference be gps? What settings need to be done in 
UHD to sync the two oscillators to gps?
There's a good knowledge-base article on this and many other topics 
here:


https://kb.ettus.com/Synchronization_and_MIMO_Capability_with_USRP_Devices




Actually I am receiving a stationary signal using two ports TX/RX 
and RX2 but I am not getting a constant phase differ

Re: [USRP-users] Sampler clock common or not?

[Koyel Das (Vehere) via USRP-users]

Hi Marcus,

My colleague suggested not to use TX/RX port and I did that. I just tested two 
sets of data of stationary wifi with both antennas this time connected to RX2 
port as he suggested and then the phase difference deviation goes maximum to 
0.3 rad. I will carry out experiments with more data sets to see if I always 
get same types of deviation.

Regards,
Koyel

From: Marcus D. Leech 
Sent: Monday, September 17, 2018 9:16:13 PM
To: Koyel Das (Vehere)
Cc: 'USRP-users@lists.ettus.com'
Subject: Re: [USRP-users] Sampler clock common or not?

On 09/17/2018 06:31 AM, Koyel Das (Vehere) wrote:

Hi Marcus,



I am looking at the following link


https://files.ettus.com/schematics/x300/x3xx.pdf


and as you said same clock is being used by both the boards. NI has asked for 
some time to find it out.


Regards,

Koyel

You might try disabling I/Q corrections in your flow-graph, and see if that 
helps.



From: Koyel Das (Vehere)
Sent: Monday, September 17, 2018 12:57:29 PM
To: Marcus D. Leech; Koyel Das (Vehere)
Subject: Re: [USRP-users] Sampler clock common or not?


Hi Marcus,


I spoke with NI and they are saying that in USRP 2954 R there are two LOs for 
two daughter boards which are not locked to any master oscillator. They are not 
phase locked but we can provide a reference clock if we want them to run 
coherently. He said in 2955 R LO sharing is there so we can share single LO for 
4 daughter boards.


Regards,

Koyel


From: USRP-users 
 
on behalf of Koyel Das (Vehere) via USRP-users 

Sent: Monday, September 17, 2018 12:12:32 PM
To: Marcus D. Leech; 
'USRP-users@lists.ettus.com'
Subject: Re: [USRP-users] Sampler clock common or not?


Hi,


I don't have a splitter currently. So to use it I need to wait until it arrives.


I din't find uhd_fft in gnuradio. I am using gnuradio-companion.


However, I see similar issues faced by some other people:


https://lists.gnu.org/archive/html/discuss-gnuradio/2008-04/msg00029.html


Regards,

Koyel



RE: [Discuss-gnuradio] Compare Phase 
block
lists.gnu.org
My VOR receiver is almost there, but not quite... My code is below. I've 
demodulated the first 30 Hz signal from AM, and the second from the FM 
subcarrier.




From: Marcus D. Leech 
Sent: Sunday, September 16, 2018 9:28:55 AM
To: Koyel Das (Vehere); 
'USRP-users@lists.ettus.com'
Subject: Re: [USRP-users] Sampler clock common or not?

On 09/15/2018 11:54 PM, Koyel Das (Vehere) wrote:
I just forwarded in another thread the processing I am doing. Have you received 
it?

Regards,
Koyel
Yes, but I'd rather not have to analyse your MATLAB code in detail.

You can use the --phase-relations  option on the uhd_fft utility for doing 
two-channel phase-coherence measurements in real-time.

Unless something is horribly broken, two UBX cards on the same underlying USRP 
platform WILL operate with mutual coherence, with
  some initial phase-offset, just due to the way synthesizers work.  But that 
phase offset will carry-forward in time, with only very minor
  phase noise perturbations just due to the physics of semiconductors, etc.



From: Marcus D. Leech 
Sent: Sunday, September 16, 2018 9:05:29 AM
To: Koyel Das (Vehere); 
'USRP-users@lists.ettus.com'
Subject: Re: [USRP-users] Sampler clock common or not?

On 09/15/2018 11:32 PM, Koyel Das (Vehere) wrote:
I don’t have access to the flowgraph currently but it’s the same flowgraph I 
sent earlier in my other previous email for 2955 R. I have UHD with two 
channels, each channel connected to a file sink. One channel is set to TX/RX, 
another to RX2. I am sampling at 1 msps. I am capturing data from two channels, 
storing in file and then reading them in matlab and cross correlating them to 
get phase difference but the phase difference for any particular frequency bin 
is showing deviations with time. Deviations are huge sometimes +-pi.

Regards,
Koyel
I would say that you're doing something wrong in the post-processing.

I routinely use USRPs of all types for interferometry in radio astronomy, where 
the long term phase noise has to be very low, or your
  data is rubbish.



From: Marcus D. Leech 
Sent: Sunday, September 16, 2018 8:48:26 AM
To: Koyel Das (Vehere); 
'USRP-users@lists.ettus.com'
Subject: Re: [USRP-users] Sampler clock common or not?

On 09/15/2018 10:38 PM, Koyel Das (Vehere) wrote:
Hi,

Thanks for the information. Do I have to set any property in the gnu radio uhd 
to phase lock the two oscillator

[USRP-users] RFNoC TX Examples

[Leandro Echevarría via USRP-users]

Hey everybody,

Is there anywhere I could find C++ examples of a TX stream using RFNoC
blocks? The latest rnoc-devel commit only has two RX examples, and I cannot
seem to replicate the behavior from that.

Thanks,

Leo
___
USRP-users mailing list
USRP-users@lists.ettus.com
http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com

Re: [USRP-users] Software changes for new FPGA registers B205

[Mark Cromwell via USRP-users]

Martin,


I can only find methods to write the user registers, poke32() and 
set_user_register() using the less invasive access method. Is there a 
peek32() or get_user_register() method to read custom registers in the 
user_regs address space?


Thanks,

Mark




___
USRP-users mailing list
USRP-users@lists.ettus.com
http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com

Re: [USRP-users] Sampler clock common or not?

[Koyel Das (Vehere) via USRP-users]

Hi Marcus,


yesterday I got the deviations of 0.3 rad by fluke. Even after receiving using 
two RX2 ports the phase difference is deviating to large values like 2 rad. I 
tried disabling I/Q  corrections by passing 0 to the property field but that 
too didn't work. I think I really now need to use splitter to find out so need 
to wait. If meanwhile I can try something else then please let me know.


Regards,

Koyel


From: USRP-users  on behalf of Koyel Das 
(Vehere) via USRP-users 
Sent: Monday, September 17, 2018 9:23:33 PM
To: Marcus D. Leech
Cc: 'USRP-users@lists.ettus.com'
Subject: Re: [USRP-users] Sampler clock common or not?

Hi Marcus,

My colleague suggested not to use TX/RX port and I did that. I just tested two 
sets of data of stationary wifi with both antennas this time connected to RX2 
port as he suggested and then the phase difference deviation goes maximum to 
0.3 rad. I will carry out experiments with more data sets to see if I always 
get same types of deviation.

Regards,
Koyel

From: Marcus D. Leech 
Sent: Monday, September 17, 2018 9:16:13 PM
To: Koyel Das (Vehere)
Cc: 'USRP-users@lists.ettus.com'
Subject: Re: [USRP-users] Sampler clock common or not?

On 09/17/2018 06:31 AM, Koyel Das (Vehere) wrote:

Hi Marcus,



I am looking at the following link


https://files.ettus.com/schematics/x300/x3xx.pdf


and as you said same clock is being used by both the boards. NI has asked for 
some time to find it out.


Regards,

Koyel

You might try disabling I/Q corrections in your flow-graph, and see if that 
helps.



From: Koyel Das (Vehere)
Sent: Monday, September 17, 2018 12:57:29 PM
To: Marcus D. Leech; Koyel Das (Vehere)
Subject: Re: [USRP-users] Sampler clock common or not?


Hi Marcus,


I spoke with NI and they are saying that in USRP 2954 R there are two LOs for 
two daughter boards which are not locked to any master oscillator. They are not 
phase locked but we can provide a reference clock if we want them to run 
coherently. He said in 2955 R LO sharing is there so we can share single LO for 
4 daughter boards.


Regards,

Koyel


From: USRP-users 
 
on behalf of Koyel Das (Vehere) via USRP-users 

Sent: Monday, September 17, 2018 12:12:32 PM
To: Marcus D. Leech; 
'USRP-users@lists.ettus.com'
Subject: Re: [USRP-users] Sampler clock common or not?


Hi,


I don't have a splitter currently. So to use it I need to wait until it arrives.


I din't find uhd_fft in gnuradio. I am using gnuradio-companion.


However, I see similar issues faced by some other people:


https://lists.gnu.org/archive/html/discuss-gnuradio/2008-04/msg00029.html


Regards,

Koyel



RE: [Discuss-gnuradio] Compare Phase 
block
lists.gnu.org
My VOR receiver is almost there, but not quite... My code is below. I've 
demodulated the first 30 Hz signal from AM, and the second from the FM 
subcarrier.




From: Marcus D. Leech 
Sent: Sunday, September 16, 2018 9:28:55 AM
To: Koyel Das (Vehere); 
'USRP-users@lists.ettus.com'
Subject: Re: [USRP-users] Sampler clock common or not?

On 09/15/2018 11:54 PM, Koyel Das (Vehere) wrote:
I just forwarded in another thread the processing I am doing. Have you received 
it?

Regards,
Koyel
Yes, but I'd rather not have to analyse your MATLAB code in detail.

You can use the --phase-relations  option on the uhd_fft utility for doing 
two-channel phase-coherence measurements in real-time.

Unless something is horribly broken, two UBX cards on the same underlying USRP 
platform WILL operate with mutual coherence, with
  some initial phase-offset, just due to the way synthesizers work.  But that 
phase offset will carry-forward in time, with only very minor
  phase noise perturbations just due to the physics of semiconductors, etc.



From: Marcus D. Leech 
Sent: Sunday, September 16, 2018 9:05:29 AM
To: Koyel Das (Vehere); 
'USRP-users@lists.ettus.com'
Subject: Re: [USRP-users] Sampler clock common or not?

On 09/15/2018 11:32 PM, Koyel Das (Vehere) wrote:
I don’t have access to the flowgraph currently but it’s the same flowgraph I 
sent earlier in my other previous email for 2955 R. I have UHD with two 
channels, each channel connected to a file sink. One channel is set to TX/RX, 
another to RX2. I am sampling at 1 msps. I am capturing data from two channels, 
storing in file and then reading them in matlab and cross correlating them to 
get phase difference but the phase difference for any particular frequency bin 
is showing deviations with time. Deviations are huge sometimes +-pi.

Regard