FreeCalypso > hg > fc-rfcal-tools
changeset 95:b68301b38ff2
doc: Cable-config-howto and General-flow written
author | Mychaela Falconia <falcon@freecalypso.org> |
---|---|
date | Mon, 17 Jul 2017 18:26:30 +0000 |
parents | b443d0665c1f |
children | ea54469409d4 |
files | doc/Cable-config-howto doc/General-flow |
diffstat | 2 files changed, 101 insertions(+), 0 deletions(-) [+] |
line wrap: on
line diff
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/doc/Cable-config-howto Mon Jul 17 18:26:30 2017 +0000 @@ -0,0 +1,47 @@ +One of the critical requirements for proper calibration of Tx power levels (and +of the Rx "magic gain" to a lesser degree) is that the insertion loss of the +cabling setup between your CMU200 and your DUTs (Calypso GSM devices to be +calibrated) needs to be precisely known and accounted for. Actually measuring +the insertion loss of your cables at the center frequency of each GSM uplink and +downlink frequency band with 0.1 dB accuracy is your responsibility; the rest +of this article will merely tell you how to feed these cable loss numbers to +FreeCalypso RF calibration tools. + +In our architecture the software component that takes care of adjusting the +power levels for the cable loss is fc-cmu200d; our fc-cmu200d reads the cable +loss numbers from a configuration file and instructs the CMU200 itself to +perform the actual adjustments. Furthermore, our architecture allows for +multiple cable configurations. Some Calypso GSM devices have SMA RF connectors +for attaching the antenna, in which case the calibration coax needs to terminate +in an SMA, others have MS-147 or Murata SWD/SWF RF test ports, in which case the +calibration coax needs to terminate in an MS-147 or Murata probe connector. If +you are working with different types of Calypso GSM devices, you will need to +switch between the corresponding cable setups, as the cable loss numbers will be +different for each setup: even if you keep the same "main" coax going from your +CMU200 to an SMA and add/remove Murata's SMA-to-SWD/SWF adapter at the end, the +addition or removal of that extra adapter will change the total insertion loss +along the full path, and this change needs to be accounted for. + +Cable configuration files are kept in the /opt/freecalypso/rfcal/cableconf +directory; it is up to you how you name your cable configs, but the following +convention is suggested: + +* If you have a cable setup that terminates in an SMA connector for calibrating + Calypso GSM devices with SMA antenna interfaces, the corresponding cable + config file should be named "sma". + +* If you have a cable setup that terminates in an MS-147 probe connector for + recalibrating Mot C1xx phones or similar, the corresponding cable config file + should be named "ms147". + +* If you have a cable setup that terminates in a Murata SWD/SWF probe for + working with devices like Openmoko GTA0x, the corresponding cable config file + should be named "murata". + +See cableconf-example for an example of the syntax, but please remember that +you are still responsible for putting in YOUR OWN numbers for the actual +insertion loss of your cable setups. + +When you actually run fc-cmu200d to begin a calibration session (see the +General-flow article), you will select the cable configuration to be used +by name with the -c option to fc-cmu200d.
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/doc/General-flow Mon Jul 17 18:26:30 2017 +0000 @@ -0,0 +1,54 @@ +The general flow of using fc-rfcal-tools to calibrate Calypso GSM devices is as +follows: + +* At the beginning of your calibration work session, you need to power on your + CMU200, wait for it to finish booting, and then run fc-cmu200d; the fc-cmu200d + command line takes the following form: + + fc-cmu200d -c cable_name /dev/ttyXXX 19200 + + Change /dev/ttyXXX to the serial port that is connected to the CMU200, give + fc-cmu200d the same baud rate that is configured on the CMU (19200 is the + default), and select the appropriate cable configuration with the -c option: + see the Cable-config-howto article. + + By default the RF2 port on the CMU200 will be used; you can give a -1 option + to use the RF1 port instead, or you can include an 'rf-port 1' or 'rf-port 2' + line in your cable config files to have the port selection go with the -c + selection. + +* In typical usage the fc-cmu200d process you started in the previous step can + persist through any number of individual device calibrations, as on a factory + production line. However, if you need to switch to a different cable setup + with different loss numbers, you will need to kill the fc-cmu200d process and + restart it with the appropriate new -c option. + +* To calibrate an individual Calypso GSM device, you will need to have an + rvinterf process running, talking to the RVTMUX UART channel on that device. + Booting your Calypso GSM device in the correct manner and running rvinterf + with the right arguments is your responsibility: fc-rfcal-* tools will expect + rvinterf to be running already, and will not launch it themselves. + +* With both fc-cmu200d and rvinterf running, talking to your CMU200 and your + DUT, respectively, and with the correct RF coax (matching your fc-cmu200d -c + selection) connected between the two, you can now run the actual calibration + automation programs. For a 900/1800/1900 MHz triband Calypso GSM device, the + standard sequence is as follows: + +fc-rfcal-vcxo +fc-rfcal-rxband 900 +fc-rfcal-rxband 1800 +fc-rfcal-rxband 1900 +fc-rfcal-txband 900 <profile> +fc-rfcal-txband 1800 <profile> +fc-rfcal-txband 1900 <profile> + +The <profile> argument to fc-rfcal-txband commands needs to be changed to the +name of the appropriate Tx levels calibration profile; see the Tx-cal-theory +article for more information. + +Each of the fc-rfcal-* steps finishes by writing the values it has calibrated +into the Calypso device's non-volatile flash file system (FFS); once all of the +steps have been completed, the Calypso device is fully calibrated. After the +calibration procedure the Calypso device should be rebooted before normal AT +command operation is attempted.