Where you are more likely to find them is where there was a high leg system to begin with, then came the need for more VA capacity but wanted to stick with the high leg system for various possible reasons.
This is exactly the situation we have run into. A customer has a facility which was previously fed by a 500kVA 240/120V center tapped delta transformer. The facility had 120/240V loads which were served directly from transformer as well as 480/277V loads which were fed via step-up transformers.
The customer wanted to increase service to 750kVA for increased capacity but utility would only provide new 480/277V service. The existing 480/277V loads can now be fed directly from service however all the existing 240V 3-Ph/1-Ph loads and 120V 1-Ph loads must now be fed from new 300kVA and 500kVA Center Tapped Delta secondary's (480V primary) connected to existing distribution panels.
The contractor was going to connect these new transformers in the field and noticed the 5% limitation (and full size neutral conductors way to big for neutral lug) and questioned if there would be issues or not. The challenge now is to try to quantify all of the existing 120V L-N loads that will be fed from these transformers and evaluate weather or not the neutral current will exceed the 5% limitation on these. As you can imagine there is no easy way to evaluate these loads so we are trying to put together as best of an analysis as we can based on panel schedules and perhaps even measuring the existing neutral currents on the (2) panels that will be connected on secondary of these new transformers (panels still in service) to get representation of unbalanced current.
Once we arrive at some understanding of these L-N load profile that then leaves us with evaluating if we consider what we think is the "connected kVA" of the 120V L-N loads or if we can prove that there will always be some degree of balancing/canceling that would allow adequate loading on the transformer.
Converting system to 120/208V was considered but would require replacing several 240V loads that are depended on 3-Ph 240V and thus no longer any option.
Open to any ideas/suggestions anyone has on trying to evaluate this situation.