I think in principle this should be ok, but with a couple of caveats which should be considered and which there is not sufficient information to answer at present:
What are the relative sizes of the drives and the transformer? I guess as the transformer is kVA and not MVA we are talking about small-to-medium drives.
How stiff is the supply network? Will regeneration cause a voltage shift on a bus common to some other circuit which could be adversely affected?
As the drive is PWM rather than a current-source type, the regenerated power should be fairly clean in so far as the total harmonic distortion will be lower than the current source type, although there will be switching-frequency distortion produced. The coupling reactor will work in your favour here.
The drive manufacturer should be able to tell you whether there will be any interaction between drives, but again this will probably depend on how stiff the supply is and whether the drives will regenerate into a higher-than-normal bus voltage.
You don't say whether this is your own supply you're polluting or the public one. What degree of harmonic pollution will your supply authority accept?
How will your supply metering react to a regenerative load? Are you regenerating onto your internal system, or effectively exporting onto the utility network?
More questions than answers I'm afraid, but hopefully some that will provoke thought, or more likely an argument...
![[wink] [wink] [wink]](/data/assets/smilies/wink.gif)