I have a GC864-QUAD that after several weeks of operation is suddenly (and continually) generating the CME error 23: “memory failure” on power up. This is in response to the at#scfg command. Presumably this is related to the non-volatile memory storage, is there a limit to how many times these kinds of commands can be called before failure or is there another reason for this error?
I don’t think you can reach such a write amount only with this;re load the firmware and see if problem persists.
Reloading the firmware fixed the problem. Thanks Cosmin.
Glad it worked, thank you for notice Wiebe.
I’m getting memory failures on a second module and suspect the same for two more in the field. Updating the firmware fixes the problem but unless the problem is related to an issue with 07.02.204 then this may just be a temporary solution and I’ll have to look for a cause elsewhere.
Some background; the module is controlled by an external uC and is powered down completely between http transfer sessions. Depending on the setup the power may be cycled frequently (as much as every 5 minutes). CGDCONT and SCFG are only set once, all other commands are set at every power cycle.
After the memory failure the module is able to send SMS’s but at#scfg fails continually (actually is was successful exactly once after the first failure but never again).
Do you have any suggestions about what may be causing the issues I’m seeing?
Are you strictly complying with the shut down and power on procedures, exactly as indicated in Hardware User Guide? Maybe you can describe little more your schematic and procedures. My colleagues say it would be useful to try 7.02.206 or 7.02.207 firmware versions.
I see that I am not complying with the shut down procedure. I’m issuing an AT#SHDN with a 10 second timeout but if this fails I power the device (Vcc) down rather then going through the Modem Reset Procedure. I’ll fix this and see if that resolved the issue.
I have a couple of questions regarding the procedures:
If PWRMON is not OFF after the Modem Reset Procedure, must the procedure be repeated (if so how often) or can the power then be disconnected from the module?
There doesn’t seem to be a command time out specified for the AT#SHDN command, what would you recommend?
The latest datasheet (GE864_GC864-QUAD_Hardware_User_Guide_r0) mentions that at 3.4v the ON# pad must be tied low for at least 4 seconds rather then 1. The Modem ON flowchart (page 26) doesn’t seem to cover this. Would it be better to monitor PWRMON while ON_OFF is low rather then after ON_OFF has been set HIGH again?
I suppose the 4 seconds value is correct for lower voltage. They key in these procedures is PWRMON line which must be monitored and decisions taken on its signals, timeouts being only for special cases of locks, which should not appear frequently if at all. At power off I would rather wait at least the prescribed 15 seconds, then proceed with failure procedures.
So, do you see that power off on your modules requires many forced actions? Try with a longer timeout, and check if you don’t have pins left in high levels of voltage, situation known for creating problems.
Hit enter to search or ESC to close
Knowledge Base & Download Zone