To sum things up, it's slowly moving towards the correct direction. You managed to get somebody to realize that the wrong type output transformer was specified, for the load at hand. Two competing engineering views and a offshore conglomerate management company, leaving it up to the plan reviewer to sort out the design deficiencies or suitabilities. Then a difference of opinions on what is premise wiring and what is separately derived....in other words, who is on the hook if things go poorly.
I come at it from a different perspective, what does the equipment need to function safely, properly and reliably, does it meet the code requirements, does it meet the local inspectors requirement, and does the job relieve me of excess liability when it's said and done. Augie has my vote. The local inspector will have the final say, not the competing engineers, installing contractor or the plan reviewer. And if it goes badly the customer will have something to say about it as well.
My view is that the revised transformer connection print is partially correct, but still has deficiencies and omissions. I think at this point the forum has answered your original question about the transformer originally specified and whether it need to be grounded or not. It is not the job of the forum members to do the engineering review for someone else's job, unless they willingly choose to do so. They are not the ones who get paid to take that responsibility. What each party decides is up to them. The forum is here to share knowledge and experience, not to perform someone else's engineering task responsibilities.
These are the things I see as incomplete or undocumented with the design as shown.
The grounding and bonding requirements of the transformer, and the requirements of the load equipment, were the reason the transformer choice was incorrect it the first place.
The transformer secondary needs to provide for short circuit currents, bonding to the grounding electrode system of the building, and provide an equipment grounding conductor to the machine. The VFD drives in the machine will need the equipment grounding conductor, as you surmised, not only for fault currents, but to provide a symmetrical ground reference for the surge suppression devices (MOV) inside the drives and a common point for dissipation of common mode currents which all VFD drives produce. Without this important pathway, properly connected, the VFD drives are likely to have short lives and be problematic in operation. The drawing also lacks a bonding connection for the case and core. It may be left up to the inspector to know and catch these things, if the installer fails to know what to do and why.
The last issue is the connection and protection of the transformer itself. There is nothing shown for the primary protection of the transformer or a main disconnect. The breakers shown on the secondary side may or may not be adequate, based on the choice of the primary overcurrent protection chosen. The transformer output is rated at 36A continuous but the beakers protecting it add up to 55A. There are tap rule requirements for distance and overcurrent protection as well as sizing of conductors, from the secondary. All of these choices as well as the physical layout and location of each item will affect if these things meet the code or machine listing requirements.
If these things were left up to the installing contractor to decide, then you have a part of the premise wiring system, and subject to the code.
If these things were specified, provided, located and prewired by the machine builder, then it would fall under the machine listing.
Your stuck somewhere in the middle as the plan reviewer, between multiple opposing viewpoints and responsibilities. When you add in the contractor and inspector the opinion may change yet again. This is why it's best to spell it out on a print, so every one involved can be on the same page.
Hope that helps.
