Hello Barry:

On the B200mini, you can transmit (on the TX/RX port) and receive (on the
RX2 port) at the same time.

The minimum sample rate depends on the lowest master clock rate and the
maximum decimation factor.  I don't recall these values off the top of my
head, but I'll go double-check them.  The maximum sampling rate is 61.44
Msps.

--Neel Pandeya

On Wed, 28 Oct 2020 at 19:45, Barry Duggan via USRP-users <
usrp-users@lists.ettus.com> wrote:

Hi Neel,

Thanks for the quick response! As a clarification:

• for half-duplex operation, is the receiver muted (and protected)
  during transmit?

• I could not find the sample rates on your website.

Barry Duggan KV4FV
https://github.com/duggabe

On 10/28/20 7:57 PM, Neel Pandeya wrote:

On 10/28/2020 09:07 PM, Barry Duggan via USRP-users wrote:

The receiver is switched to the RX2 port whenever the ATR logic
determines that transmission is occurring.  It's not clamped or anything,
just switched to the RX2 port.  Which if that port has no antenna, or
is terminated will protected the receiver.  The receiver WILL
receive your own transmissions--because the RX/TX switch isolation is
only about 30dB.

The master clock rate determines the allowable sample rates, and the
master-clock rate can be a very large number of discrete
values between something like 5Mhz and 61.44MHz.  The maximum
decimation is, I believe, 512.

So, the to-the-host sample rate has to be a proper integer fraction of
the master clock rate, and the master clock rate is programmable.
So the answer to the question 'what sample rates does it support' is
"geez, man, really a lot".

Now, having said that, I did at one point in the last few years see an
equation that describes the allowable master clock rates.
The allowable master clock rates aren't exactly "arbitrary", but
pretty large.  My tendency is to pick a clock rate that is 4 times
my desired sample rate.  Most of the time, that works.  Except when I
choose a sample rate that happens to yield a required
master clock rate that isn't allowed by the clock PLL.  For me, that
is rare.  For others, that may not be.

On 10/28/2020 09:07 PM, Barry Duggan via USRP-users wrote:

https://files.ettus.com/manual/page_usrp_b200.html#b200_mcr

It says "ANY value between 5MHz and 61.44MHz", but in reality, it has to
be integer, and there are rates that don't work because of the
PLL architecture--I just can't remember what those restrictions are.

But even if the step size is 10kHz, with integer decimation implemented
in the FPGA, that's really a lot of different supported sample rates.

Normally UHD will pick a master clock rate for you, given the desired
sample rate.  It will try to pick the highest multiple of your
desired sample rate that is also a a multiple of two. Sometimes,
you'd rather it not try to be "clever", so you specify the master clock
rate as a device argument and set sample rates accordingly.

If you change the UHD printout from MHz to Hz, it looks like the step
size is around 0.4 Hz at 23 MHz (and MCR doesn't have to be an integer).
Using a DVB-T2 sample rate of (8000000.0 * 5) / 7 and adding 0.01 Hz for
each run (with 4X MCR)

[INFO] [B200] Asking for clock rate 22857142.977143 Hz... ((8000000.0 *
5) / 7 + 0.03)
[INFO] [B200] Actually got clock rate 22857143.028110 Hz.

[INFO] [B200] Asking for clock rate 22857143.017143 Hz... ((8000000.0 *
5) / 7 + 0.04)
[INFO] [B200] Actually got clock rate 22857143.028110 Hz.

[INFO] [B200] Asking for clock rate 22857143.057143 Hz... ((8000000.0 *
5) / 7 + 0.05)
[INFO] [B200] Actually got clock rate 22857143.028110 Hz.

[INFO] [B200] Asking for clock rate 22857143.097143 Hz... ((8000000.0 *
5) / 7 + 0.06)
[INFO] [B200] Actually got clock rate 22857143.028110 Hz.

[INFO] [B200] Asking for clock rate 22857143.137143 Hz... ((8000000.0 *
5) / 7 + 0.07)
[INFO] [B200] Actually got clock rate 22857143.028110 Hz.

[INFO] [B200] Asking for clock rate 22857143.177143 Hz... ((8000000.0 *
5) / 7 + 0.08)
[INFO] [B200] Actually got clock rate 22857143.028110 Hz.

[INFO] [B200] Asking for clock rate 22857143.217143 Hz... ((8000000.0 *
5) / 7) + 0.09
[INFO] [B200] Actually got clock rate 22857143.028110 Hz.

[INFO] [B200] Asking for clock rate 22857143.257143 Hz... ((8000000.0 *
5) / 7) + 0.10
[INFO] [B200] Actually got clock rate 22857143.427032 Hz.

The claim of "any sample rate" is not that far off.

Ron

On 10/28/20 18:54, Marcus D. Leech via USRP-users wrote: