The shroud will be designed with a center baffle so that if one fan dies the other will not be affected. But it is unlikely they will ever service this unit without shutting the whole system down, so it's not really a concern. I will however, run an interlock on the individual overloads to prevent damaging the drive.
Might be worth putting a fuse in each phase of each motor, unless the two motors are always at the same load. The drive wont know if one motor is unloaded and the other overloaded
I for one would not put fuses downstream of drives running one motor. Aside from the issues of single-phasing the drive output, a fuse blowing has the same effect of opening a contactor (or disconnect) downstream of the drive; potential loss of the transistor. But there are some NEC installation issues to consider for those of us in the US (and Canada, who's CEC is virtually the same) which usually end up in requiring new SCPD devices downstream of the VFD. Read section 450.53 of the NEC carefully when considering this; there are a lot of what-ifs and issues involved, too complex to go in to here.
But here's the upshot: when you have multiple motors running downstream of a VFD and one trips off line, the dv/dt issues with regards to damaging the output transistors are somewhat mitigated by the fact that there is still an inductive load running in parallel, absorbing a lot of the voltage spike issues created by the contacts opening. I have used Motor Starter protectors (IEC manual motor starters) as downstream protective devices for years, never had a problem. They take care of the individual OL protection as well as the short circuit protection, provide the means of disconnect that LionelHutz mentioned, and open all three phases, which prevent single phasing one of the motors.
By the way, to LionelHutz's defense (not that he needs it), I don't believe he meant using the downstream disconnect to open one motor circuit while running. His comment came after ScottyUK's concern about running one fan after the other had failed. So if one motor fails, a lockable disconnect allows the user to RESTART the remaining fan while the failed one is being serviced.
xJAyx
another point to ponder.
You mention that the VFD is going on a ship.
VFD's on ships will generally require some sort of certification from Lloyds Register or ABS otherwise the ship owner might get a little upset you putting it on there. They generally have to be carefully selected with regards leakage current and control isolation. Make sure any sort of RFI filter is disconnected or specially designed for ship applicatiuons otherwise the leakage current will be a problem.
Yes, Jraef you are correct. Put the disconnects inside the enclosure or something similar so someone can't randomly switch them off while running. You really just want it for the ability to apply a lock for service reasons while still being able to restart and run the other motor at the same time.
Re the disconnect: Since it is shipboard, ABS (and other Reg Bodies/IEE Std 45) a required disconnect may be at the controller, but it must be externally operable and isolates the controller as well. So if single disconnects are put inside the controller to isolate motors another may be required externally that disconnects the entire circuit. Unless the feeder breaker is withing the same compartment as the controller.
The same entities require individual motor overload protection (they don't all agree on the settings, but ABS is 100-125% FLA), but you can have the OL relay isolate the drive input.
