LMS TxLO noise

This is merely a historical archive of years 2008-2021, before the migration to mailman3.

A maintained and still updated list archive can be found at https://lists.osmocom.org/hyperkitty/list/UmTRX@lists.osmocom.org/.

Alexander Chemeris alexander.chemeris at gmail.com
Sat Jul 21 20:43:27 UTC 2012


On Sat, Jul 21, 2012 at 12:44 AM, Thomas Tsou <thomastsou at gmail.com> wrote:
> On Fri, Jul 20, 2012 at 6:09 PM, Alexander Chemeris
> <alexander.chemeris at gmail.com> wrote:
>> On Fri, Jul 20, 2012 at 10:22 PM, Thomas Tsou <thomastsou at gmail.com> wrote:
>>> I also still have errors with auto calibration.
>>
>> Well, you haven't enabled Rx chain in LMS and thus it's reasonable
>> that Rx-related calibration is failing. You should do "--lms-rx-enable
>> 1" before you try to do anything with Rx.
>
> Hmm. Similar error.

Hum. I just realized that we set register 0x09 to 0x80 in lms_init(),
which means we set RXOUTSW to 1 and pins 113-116 of the chip are
connected directly to the ADC. While this should not make any
difference in theory, it's better to set this register to 0x00 on init
to avoid any external interference.


One more interesting find is that EVB Quick Starter Manual recommends
to power down SPI blocks inside of the chip if they're not used (see
below). Could you try to set 0x09 register to 0x00 after the tuning
and check whether it affects phase noise?

Clock Buffers control
-----------------------------
Enable pins turn the internal clock buffers on and off. These should
be enabled when control of the device is needed, however during
operation SPI clocks which are not being used should be disabled to
reduce the risk of SPI clock spurious.

-- 
Regards,
Alexander Chemeris.
CEO, Fairwaves LLC / ООО УмРадио
http://fairwaves.ru




More information about the UmTRX mailing list