# HG changeset patch # User Mychaela Falconia # Date 1518488727 0 # Node ID 630617d5ece329a5c42a8cdd50389eef764778c6 # Parent 5a7c174acd3261977394e4cee2e552f968ac6426 txlevels: updated README diff -r 5a7c174acd32 -r 630617d5ece3 txlevels/README --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/txlevels/README Tue Feb 13 02:25:27 2018 +0000 @@ -0,0 +1,13 @@ +Two sets of Tx power level calibration profiles have been created so far: +rf3166-ideal and rf3166-fcdev3b. As implied in the naming, both sets of +profiles are meant for GSM MS designs using the RF3166 PA as used by Openmoko, +following OM's triband board design for the RF section. The only difference +between the two versions is in the target dBm number for the maximum power +level for the low band (EGSM or GSM850): it is 33 dBm as the GSM 05.05 spec +calls for in the rf3166-ideal-900 profile, but the rf3166-fcdev3b-900 profile +has it shifted down to 32.2 dBm, similarly to how TI did it on their I-Sample +(LoCosto) boards. The rf3166-ideal profile set works for recalibrating +Openmoko-made GTA02 devices, producing even better Tx power levels than OM's +original (set to unshifted GSM 05.05 spec targets), but on our own FCDEV3B +boards we are not able to get the full 33 dBm Tx output in the low band, hence +the rf3166-fcdev3b profile set with the shifted target. diff -r 5a7c174acd32 -r 630617d5ece3 txlevels/fcom1.README --- a/txlevels/fcom1.README Tue Feb 13 01:11:41 2018 +0000 +++ /dev/null Thu Jan 01 00:00:00 1970 +0000 @@ -1,24 +0,0 @@ -The name "fcom1" stands for "FreeCalypso hardware, Openmoko-based, version 1", -and the set of Tx power level calibration profiles named "fcom1" has been -created empirically by the Mother of FreeCalypso to produce successful Tx level -calibration runs on our current FCDEV3B boards with our current CMU200 and -cabling setup. - -These fcom1 Tx power level calibration profiles should ONLY be used for -(re)calibrating FreeCalypso-made hardware, i.e., FCDEV3B boards and potentially -other closely related FC hardware designs in the future; these profiles are NOT -expected to be correct for recalibrating Openmoko's own FreeRunners or for -calibrating other (non-FreeCalypso) Openmoko-derived GSM devices that retain -the internal antenna and the Murata SWD RF test port used by Openmoko, as -opposed to the SMA used on our FCDEV3B. - -For anyone interested in one of those latter use cases, a different set of -profiles would need to be created; the Mother is not currently able to create -those profiles with a good confidence of correctness despite owning several -Openmoko-made FreeRunners and the appropriate Murata MXHS83QH3000 adapter -because the insertion loss of the latter adapter is not precisely known. -Knowing the total insertion loss along the full path from the DUT to the CMU200 -is a prerequisite for determining the correct profiles for Tx power level -calibration, and when Murata SWD/SWF RF test ports are involved, this cabling -path includes both the cable from the CMU200 ending in an SMA and the adapter -from Murata that goes between the SMA and their switch connector.