[USRP-users] USRP B210 not found and power cycle needed

[Francisco Gallardo lópez]

Hi,

We have a remote station with an USRP B210.

We observed that sometimes the remote machine cannot find the USRP B210 (I
think it has to do with uncontrolled power cycles, but I am not 100% sure
about the root cause. The point is that sometimes it happens).

The only way to sort it out is to disconnect the power supply and the USB
cable from the USRP, and then the remote machine can detect it again. i.e.
doing a complete power reboot to the USRP.

It is not a killer because the problem is not always happening, but it is
annoying because it implies sending somebody to disconnect and connect the
USRP B210 cables.

Is anybody experiencing a similar issue? Any idea how to fix it?

The remote machine uses Ubuntu 22.04.3 LTS and the UHD drivers.

Thanks!
Fran
___
USRP-users mailing list -- usrp-users@lists.ettus.com
To unsubscribe send an email to usrp-users-le...@lists.ettus.com

[USRP-users] Re: USRP B210 not found and power cycle needed

[Francisco Gallardo lópez]

Hi,

Actually, that is part of the problem. It happens sometimes, but not
always.
I would say mainly when there is a power outage or similar, but not always.

For the sake of testing, I just rebooted the remote machine. This time it
appears in both lsusb and uhd_find_devices upon reboot.

As said, I am not sure about the root cause. So it is complicated to tell.

When that happened last time, I did not check lsusb, but uhd_find_devices,
and the system was not seeing the USRP. I guess lsusb would also not show
the device.

I will check the thread Eden pointed out.

Thanks
Fran



El mar, 15 ago 2023 a las 16:59, Marcus D. Leech ()
escribió:

> On 15/08/2023 02:43, Francisco Gallardo lópez wrote:
> > Hi,
> >
> > We have a remote station with an USRP B210.
> >
> > We observed that sometimes the remote machine cannot find the USRP
> > B210 (I think it has to do with uncontrolled power cycles, but I am
> > not 100% sure about the root cause. The point is that sometimes it
> > happens).
> >
> > The only way to sort it out is to disconnect the power supply and the
> > USB cable from the USRP, and then the remote machine can detect it
> > again. i.e. doing a complete power reboot to the USRP.
> >
> > It is not a killer because the problem is not always happening, but it
> > is annoying because it implies sending somebody to disconnect and
> > connect the USRP B210 cables.
> >
> > Is anybody experiencing a similar issue? Any idea how to fix it?
> >
> > The remote machine uses Ubuntu 22.04.3 LTS and the UHD drivers.
> >
> > Thanks!
> > Fran
> >
> I've seen this sporadically with B2xx over the years I've been using
> them.  For example, on some hardware, the system cannot
>see the device on reboot, and you have to replug/power-cycle for the
> system to even see it.
>
> When this happens, does it show up in "lsusb" ?
>
> ___
> USRP-users mailing list -- usrp-users@lists.ettus.com
> To unsubscribe send an email to usrp-users-le...@lists.ettus.com
>
___
USRP-users mailing list -- usrp-users@lists.ettus.com
To unsubscribe send an email to usrp-users-le...@lists.ettus.com

[USRP-users] Strange interference

[Francisco Gallardo lópez]

Hi all,

I am seeing a quite strange pattern in an USRP B210, and I am trying to
figure out what it is.

Right now this is more like an academic investigation, as this is not
causing malfunctions, but it got my attention.

So, from time to time I am seeing a tone (20dB above noise) appearing just
in the middle of the band, to be precise ±2KHz around the central frequency
that I configured in the USRP.

I was not paying attention at all, until I noticed that it appears every
day seven times a day, and always (more or less) at the same times. Hence,
this is not random.

Gut feeling is that this is caused by some sort of signal induced into the
cables or locally radiated interference, not intended, of course.

I was trying to further investigate, and for that I was trying to get a
detail blocks diagram of the B210, as I saw the following information in
the B210 website: "If you, however, happen to have a very strong interferer
within half the master clock rate of your RX LO frequency, you might want
to reduce this analog bandwidth. You can do so by calling
uhd::usrp::multi_usrp::set_rx_bandwidth(bw)." B200/B210/B200mini/B205mini -
Ettus Knowledge Base 

Now the thing is, I realized I don't really know what the internal settings
of the USRP are, also in terms of the Analog Devices AD9363. I know that
the clock rate is 32 MHz, but from there I don't clearly know what is the
LO frequency that the USRP set.

Also, I would like to know the setting that the USRP is using for the
Analog bandwidth.

In short, I would like to:

a) Get a diagram like this one, but for the detailed case of B210. Is there
something like this? (File:2920 simplified system diagram.gif - Ettus
Knowledge Base
)
b) Know what is the LO freq that the USRP is setting (I assumed based on
the 32MHz clock rate)
c) Know where I can get the information on the setting the USRP is putting
for the analog bandwidth filter.

Thanks!
Fran
___
USRP-users mailing list -- usrp-users@lists.ettus.com
To unsubscribe send an email to usrp-users-le...@lists.ettus.com

[USRP-users] Re: Strange interference

[Francisco Gallardo lópez]

Hi Marcus,

Thanks, I was actually looking for a simplified block diagram, more than
the schematics.

As per the get_rx_bandwidth. I checked, and it is 56MHz.

I was guessing that this was more or less the config (my sampling frequency
is actually 4MHz). Mainly because after sending the email this morning I
checked some MHz away from my working freq (but still inside of ±16MHz) and
the RFI was actually there and coming into my processing chain via
subsampling.

Hence, it seems it is not the USRP, there is an actual RFI at the end of
the day coming into the USRP RF input.

I may reduce the bandwidth, though.

Thanks!
Fran


El mar, 26 dic 2023 a las 17:14, Marcus D. Leech ()
escribió:

> On 26/12/2023 06:43, Francisco Gallardo lópez wrote:
>
> Hi all,
>
> I am seeing a quite strange pattern in an USRP B210, and I am trying to
> figure out what it is.
>
> Right now this is more like an academic investigation, as this is not
> causing malfunctions, but it got my attention.
>
> So, from time to time I am seeing a tone (20dB above noise) appearing just
> in the middle of the band, to be precise ±2KHz around the central frequency
> that I configured in the USRP.
>
> I was not paying attention at all, until I noticed that it appears every
> day seven times a day, and always (more or less) at the same times. Hence,
> this is not random.
>
> Gut feeling is that this is caused by some sort of signal induced into the
> cables or locally radiated interference, not intended, of course.
>
> I was trying to further investigate, and for that I was trying to get a
> detail blocks diagram of the B210, as I saw the following information in
> the B210 website: "If you, however, happen to have a very strong
> interferer within half the master clock rate of your RX LO frequency, you
> might want to reduce this analog bandwidth. You can do so by calling
> uhd::usrp::multi_usrp::set_rx_bandwidth(bw)." B200/B210/B200mini/B205mini
> - Ettus Knowledge Base
> <https://kb.ettus.com/B200/B210/B200mini/B205mini#FAQ>
>
> Now the thing is, I realized I don't really know what the internal
> settings of the USRP are, also in terms of the Analog Devices AD9363. I
> know that the clock rate is 32 MHz, but from there I don't clearly know
> what is the LO frequency that the USRP set.
>
> Also, I would like to know the setting that the USRP is using for the
> Analog bandwidth.
>
> In short, I would like to:
>
> a) Get a diagram like this one, but for the detailed case of B210. Is
> there something like this? (File:2920 simplified system diagram.gif -
> Ettus Knowledge Base
> <https://kb.ettus.com/File:2920_simplified_system_diagram.gif>)
> b) Know what is the LO freq that the USRP is setting (I assumed based on
> the 32MHz clock rate)
> c) Know where I can get the information on the setting the USRP is putting
> for the analog bandwidth filter.
>
> Thanks!
> Fran
>
>
>
> ___
> USRP-users mailing list -- usrp-users@lists.ettus.com
> To unsubscribe send an email to usrp-users-le...@lists.ettus.com
>
> Francisco:
>
> There's the schematics here:
>
> https://files.ettus.com/schematics/b200/
>
> Nearly all SDRs use direct-conversion, which means the LO is the same as
> the desired RF frequency.  The B2xx series works like this.
>
> Bandwidth can be set with a set_rx_bandwidth call, formally described here:
>
>
> https://files.ettus.com/manual/classuhd_1_1usrp_1_1multi__usrp.html#a4e026819f286e69c48c2e1956d95c6fb
>
> Normally, UHD (the driver) will set the analog bandwidth to something
> "appropriate" given your sample rate.
>
> If you change to a slightly different frequency, does this occasional RFI
> stay at the same frequency?
>
> If you lower the RF gain, does it go away?
>
>
> ___
> USRP-users mailing list -- usrp-users@lists.ettus.com
> To unsubscribe send an email to usrp-users-le...@lists.ettus.com
>
___
USRP-users mailing list -- usrp-users@lists.ettus.com
To unsubscribe send an email to usrp-users-le...@lists.ettus.com

[USRP-users] Re: Strange interference

[Francisco Gallardo lópez]

I just realized my last message was rejected because of the size of the
attachments. I am resending it with the picture in Google Drive.

Indeed, just for completeness:

The captures below were done with the case open, one with the GPSTXCO
connected and without it:

With it:

https://drive.google.com/file/d/173OqgJsi_7j0BYpE939FipxcMe8onXuI/view?usp=drivesdk



Without it:

https://drive.google.com/file/d/16vPzIC1fLphIlyw-_QG5Yc4pm0fgBC0H/view?usp=drivesdk


The difference is quite noticeable. And if on top of it, you close the
case, you have the perfect storm.

Thanks
Fran


On Fri, Jan 5, 2024, 11:53 Francisco Gallardo lópez <
f.gallardo.lo...@gmail.com> wrote:

> Indeed, just for completeness:
>
> The captures below were done with the case open, one with the
> GPSTXCO connected and without it:
>
> With it:
>
>
> Without it:
>
>
>
> The difference is quite noticeable. And if on top of it, you close the
> case, you have the perfect storm.
>
> Thanks
> Fran
>
> El jue, 4 ene 2024 a las 18:36, Francisco Gallardo lópez (<
> f.gallardo.lo...@gmail.com>) escribió:
>
>>
>> I was guessing that because grabbing the device out of the box makes the
>> interference weaker, I would say that even taking the GPS TCXO out also
>> helps, a bit.
>>
>> Did somebody try soldering some metallic shielding on the board? it seems
>> like it is designed for that.
>>
>>
>> On Thu, Jan 4, 2024, 18:16 Marcus D. Leech 
>> wrote:
>>
>>> On 04/01/2024 10:54, Francisco Gallardo lópez wrote:
>>>
>>> Hi Brian,
>>>
>>> Yes sir, precisely each 80 MHz... You found the root cause.
>>>
>>> Shit, that is a considerable inconvenience. I saw the spike gets
>>> stronger or weaker depending on whether the USRP box is opened. Actually,
>>> it is weaker when opened. I will check whether I can improvise some
>>> shielding directly in the board itself.
>>>
>>> Any suggestions here?
>>>
>>> This STRONGLY suggests that the mechanism is radiative coupling.  The
>>> box acts as a waveguide cavity at certain frequency
>>>   ranges.  Lining the box with carbon-foam RF absorber will help.
>>>
>>> This also explains why my observations show the spur as considerably
>>> weaker--I'm using an ad-hoc box from Hammond that
>>>   is quite a bit larger than the factory box.
>>>
>>>
>>>
>>> Thanks!
>>> Fran
>>>
>>>
>>> On Thu, Jan 4, 2024, 16:00 Brian Padalino  wrote:
>>>
>>>> On Thu, Jan 4, 2024 at 8:52 AM Francisco Gallardo lópez <
>>>> f.gallardo.lo...@gmail.com> wrote:
>>>>
>>>>> As for the USB2, I just tried. It is still there (see attached (you
>>>>> can see the USB type in the terminal))
>>>>>
>>>>> [image: Screenshot from 2024-01-04 13-31-53.png]
>>>>>
>>>>
>>>> 1560 MHz is the 39th harmonic of the 40 MHz reference used in the
>>>> device.
>>>>
>>>> Check to see if you have interference spikes every 80 MHz or so.
>>>>
>>>> Brian
>>>>
>>>
>>> ___
>>> USRP-users mailing list -- usrp-users@lists.ettus.com
>>> To unsubscribe send an email to usrp-users-le...@lists.ettus.com
>>>
>>>
>>> ___
>>> USRP-users mailing list -- usrp-users@lists.ettus.com
>>> To unsubscribe send an email to usrp-users-le...@lists.ettus.com
>>>
>>
___
USRP-users mailing list -- usrp-users@lists.ettus.com
To unsubscribe send an email to usrp-users-le...@lists.ettus.com

[USRP-users] USB USRP with VMWare ESXI

[Francisco Gallardo lópez via USRP-users]

Hi all,

I am trying to get to work a simple signal recorder with an USRP1 via USB.

With a bare metal installation I got no samples drop with fs=8MHz.

Now, with a VM and the USRP connected to the ESXI I need to go down to
fs=4MHz.

I was wondering whether somebody has tried this and knows how to tune the
VM/ESXi in a way to get better performances. I know I cannot expect it to
run as fast as in a bare metal installation, but maybe there are ways to
tune it a bit.

Currently I took out the USB 2.0 controller from the VM and the only
controller available to the VM is the USB 3.1. The ESXI has USB 3.0
physical ports.

In the VM machine I am writing the file to a ram disk.

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