I can think of two very likely possibilities to explain what you are seeing. First, make sure that the Com or C terminals of the FP-TC-120 and FP-RLY-422 module are not connected. Each modules inputs/outputs are isolated from the intermodule communication bus and thus each other. However, when two modules are cascade wired (connecting the V & C terminals from one terminal base to the next), the isolation between the modules is bypassed. If you have the FP-TC-120 and FP-RLY-422 module's commons wired together, any changes to the common potential on one module automatically effects the other, and can cause common mode noise problems. Depending upon the range setting in the FP-TC-120, the module can handle between 0.5 and 2.0 volts of common mode between channe
s. This leads us into possibility #2, Common mode noise. If one of the thermocouple wires is near some noise source, you may be inducing too much common mode voltage onto the channel and affecting the common potentials of the other thermocouples as the channels are not isolated. Things you can do to fix this problem include using shielded thermocouples, placing thermocouples that will be at different potentials on separate FP-TC-120 modules, or re-routing the thermocouple wiring away from the noise source.
As for question #2, that will depend upon your system. If the system is being controlled by an external computer and having the system powered, but the computer not on (thus the heater would turn on and stay on) could cause the chamber to overheat and be damaged, then no, it is not the correct way to wire the system.
Regards,
Aaron
LabVIEW Champion, CLA, CPI