I am working on an arc flash study for a system with a multi-section main switchboard. I have created the model in ETAP. I modeled each of the three sections as a separate busses linked in order by very low-value Impedances (since you can't link a bus to a bus directly in ETAP; I've also modeled it with an appropriately sized conductor of negligible distance). Each section is a separate compartment, with the first section with the main breaker from the utility being different dimensions than the other two sections.

The arc flash analysis in ETAP can't determine a FCT for the first section with the utility main breaker. I'm guessing this has something to do with not being able to isolate from the back-current coming to the bus from the second section of the MSB. However, it seems like that problem should apply to the other two busses for the MSB as well, but ETAP did find a FCT for both of those.

Is there a better way to model a multi-section main switchboard? Is there a workaround for this issue if anyone else has encountered it?

Unless there is some specific reason you are trying to create separate bus sections to represent each breaker compartment, there is no reason to do so for arc flash. Unless it is arc resistant switchgear, you cannot take credit for separate compartments. The incident energy value is going to be the same for the whole line up. The main breaker is excluded and the next upstream protective device is considered the clearing device.

Model it as one bus.