Hi Duane, I am copying the instrcution I gave before to Hans, and they worked for him. I don't think anything has happened to change the procedure. The modems on the bottom are all operating in the LF band, and have node IDs: 2 - the MARS attached node (ATM-90x, always awake) 3 - the first independent node (compact modem, in hibernation) 4 - the second independent node (compact modem, in hibernation) When you attempt communications, it will be best to configure your deck box with node ID 2 (S-register 18 should be set to 2), the same as the MARS attached node. In other circumstances, this would be unwise, but due to the limited nature of this setup and planned commands, having two nodes with the same ID will not cause a problem. Also, set S13=4, which raises the verbosity, which gives some more information if communications fail, but nearly work. In the following scripts, "" will indicate node ID "", which should be replaced by the single digit "3" or "4", depending which modem is being addressed. The standard modem wakes up instantly on receiving an incoming packet, but the compact modem is in a hibernate mode and needs an explicit wakeup command, after which it remains active until commanded to hibernate or 3 minutes of inactivity occur. The compact modems are also in datalogger mode, so data from the BMFC is stored internally, rather than sent out acoustically. Thus a desired startup sequence would be: "ATW" - wakeup node , takes about 30 seconds, ends with "OK" or "Response not received" "ATR" - range command, verifies that the node is awake. Try this once even if the wakeup command got "Response not received" If the ATR command results in a message "Error in header", that is an indication that the subsea modem has sent a response (therefore it is awake), but the response was too weak to be properly received. If reliable communications are established, as evidenced by a proper range response, then try: "AT$BN" - to get a report of how many bytes have been stored in the data log Try: "AT$BD" - to transition from datalog mode to command mode "ATEC,EW15=1" - commands node to enter data mode "ATED,R2" - commands node to make a two day report If a data report actually comes back, check the lines labelled "Li top" and "Li bot", to confirm they are at least 3.8V, and the line "Cap" should be at least 1.9V. Note: on 2/9/12, node 3 had a cap value of 1.849V, and node 4 had a cap number of 1.829V. If these voltages are still lower than 1.87V, do NOT do the 8 day report below. If the voltages are good, then do: "ATED,R8" - to get a full 8 day report (the entire buffer) transmitted acoustically After the report, close it down by doing: "AT$BE" - to place it back in datalogger mode "ATED,R8" - to get a full 8 day report placed into the datalogger "ATEC,L" - to put it back in hibernate mode (note this will always get "Response not received") If the voltage is well over 3.8 (>=3.9V), and the datalogger is not too large (the size returned by the AT$BN, I am still thinking about what "not too large" means), we could do: "AT$BT" - to dump the contents of the datalogger If the datalog is successfully dumped, it can be cleared with the command: "AT$BC" - clear data log Please give me at least a few hours warning on when you will be attempting this, so I can make sure I will be available for any questions on my cell phone, 978-761-4884. I will also connect to the MARS attached node, to try to observe what is happening. If you have any questions, please give me a call, Michael