Whenever a new device is included, several features are defined in communication with the hub:
- The master/slave structure with several sub-devices (each with a device name and number) and end-points
- The parameter structure (assuming there is a template) with default values
- The basic associations to the HC2
- The network connections of the new device
When all these are done, a "Configuration complete" is reported
My question is related to the first item on this list, the basic structure. I have experienced that the sub-device (slave) structure is not consistent. As an example, inclusion of the HeatIt Z-Trm2 thermostat may end up with either 8, 10 or 11 sub-devices, "configuration complete" being reported in every case. If a soft reconfigure is done later, some slave units may disappear, while others are added. The sub-device names may end up being duplicated (e.g. "xxx.2.1" for two separate slaves). A full reconfigure avoids this, but may again end up with a variable device structure. I have experienced similar problems with other devices than the Z-Trm2 as well.
This makes me wonder: When a device is set up at inclusion, is there no verification step in the protocol to assure that all sub-devices have been defined correctly? I suspect that jumbled structures may appear when communication between the hub and the new device is noisy or unreliable. But in such cases, error handling and verification should correct any errors and keep working until the basics are correct.
This question therefore is for anyone who knows something about the basic protocols of the HC2, or maybe the z-wave protocol itself. Any experts out there who can clarify this? For normal users, this is obscure.
PS: I post this in the HC2 forum, as it is a generic question about functionality. I assume the question is also relevant for the HC3, or even to the z-wave protocol.