AT+W32K=1,0 Wakeup - Strange Timing Behavior



I’m trying to send a Fastrack Supreme into a low power consumption state - specifically I’m trying to achieve the Slow Idle Mode without the DTR signal (AT+W32K=1,0). I have successfully gotten the power down to ~3mA using this command (I also send an AT+CFUN=0 prior to the W32K), but have been experiencing some strange behavior when trying to wake up the device using a +CALA alarm or an adl_tmr (note, not an TCU/IRQ timer). Basically the modem seems to only check for timer expiration every 5 minutes, regardless if the timer was a CALA or an adl_tmr timer. If the modem is supposed to wake up 30 seconds after I put it to sleep, it wakes up 5 minutes after I put it to sleep. If the modem is supposed to wake up at 5.5 minutes after I put it to sleep, it wakes up at the 10 minute mark. I have achieved similar results with higher value times.

I can write my software to accommodate this (undocumented?) behavior, but I worry that the modem won’t always behave in this manner. The documentation does say that “Wavecom Wireless CPU decides on its own to leave 32K low consumption at each GSM/GPRS paging occasion.” Unfortunately, the documentation does not describe paging. Since my modem must continue operating without continuous cellular coverage, I do not want to rely on an undocumented “paging occasion.”

Is this expected behavior? Has anyone else seen anything like this? I’ve searched the forums without luck.

  • Fastrack Supreme 20
  • Open AT Embedded Software Suite package 2.21.0.RC1
  • Open AT OS Package
  • Firmware Package 7.3.0.a.00
  • C-GPS-OPUS-I Open AT Plug-in Package
  • WIP Open AT Plug-in Package



“Paging” is part of the GSM system.

If you put your fastrack near some sensitive audio kit, you will hear the “deee-dit-de-deee-dit-de-deee” noise periodically - that’s “Paging”.

You could see if that lines up with the times at which your app wakes…?