-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Somebody in the thread at some point said:
| I tried to get an external antenna today but nobody sells them around | here, so I ordered one for tomorrow. It turned up, but it had the wrong connector. I cut it off and soldered it in the external connector for testing. With this external antenna (TomTom 5A00.206) the current detector outputs 2.15V into the antenna source mux, vs "Recommended operating range" of 2.7V "min" on this source select signal. So the select line is out of this "recommended operating range" both ways. They don't show the "on impedence" against Vcont level in the datasheet, but half a dozen times, whenever they talk about isolation or insertion loss they define Vcont as being "2.8V / 0V". I also created a couple of canned UBX packets to control antenna short/open detect, which are attached. I used them like this: echo 0 > /sys/class/i2c-adapter/i2c-0/0-0073/neo1973-pm-gps.0/pwron && sleep 1s && echo 1 > /sys/class/i2c-adapter/i2c-0/0-0073/neo1973-pm-gps.0/pwron && stty -F /dev/ttySAC1 -echo && cat ubx-noant-0 >/dev/ttySAC1 && cat -u /dev/ttySAC1 | grep -v ^$ ubx-noant turns off short and open detect but leaves "Antenna supply voltage control signal", ubx-noant-0 turns even that off. I saw the corresponding changes in antenna status report so I know these commands were understood, but it made no difference to the very bad GPS performance I see here. To my regret forcing Vcont to 3.0V or 0V when on external or internal antenna did not make any difference either. But when I left the thing for 20 minutes, it did find one satellite and the time. Otherwise it finds nothing even on a nice external antenna here at the moment. - -Andy -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org iEYEARECAAYFAkgYqSoACgkQOjLpvpq7dMrkUACeMs+FpaMIyE8SuGc400KgIy8T U4EAn2Uv0wGACbQRrwgca/QrlkNPXa+G =1owl -----END PGP SIGNATURE-----
ubx-noant
Description: Binary data
ubx-noant-0
Description: Binary data
_______________________________________________ hardware mailing list [email protected] http://lists.openmoko.org/mailman/listinfo/hardware
