What is your opinion if the tap is made to a bus bar or other conductor not listed in chapter 9 tables?
A tap can be made to any type of feeder conductor, per 240.21(B). I just don't think that by definition, the breaker lugs are feeder conductors. They are a part of the overcurrent
device.
Where the tap starts at an overcurrent device connection lug do you have to have a "feeder conductor"? Electrically all the current still passes through the overcurrent device, through the lugs, and is branched to each feeder tap. Is not the overcurrent protection of the tap conductors what the 240.21 rules are all about. Having that one more segment of full sized conductor, even if only a few inches long, is going to make it any more, or less, safer?
If the "tap" started at the overcurrent device lug, you would have a "feeder conductor." However, those "feeder conductors" (lets say the separate sets of #3/0) are not properly protected against overcurrent protection per 240.4 and 240.21. You would have 3/0 feeder conductors being protected by a 400A OCPD. 240.21 says that conductors shall have overcurrent protection
"located at the point the conductors receive their supply." The point that the #3/0 conductors receive their supply is at the lugs of the 400A overcurrent protective device. The have overcurrent protection
AT their point of supply which exceeds the value permitted for similar conductors...rather than
AHEAD of their point of supply. The do not meet the definition of tap conductors in 240.2.
How about installing the two 3/0 conductors on a 400 amp OCPD and a few inches from the lugs install a compression connector between the two 3/0 conductors effectively making them parallel conductors for just a few inches. This would comply with what you are saying but still seems pointless to do so.
Yes, that would be perfectly compliant. The parallel sets of #3/0 have an ampacity of 400. The point that this feeder receives its supply is the 400A OCPD. This complies with 240.4 as far as protecting the conductors at their ampacity, and with 240.21 as far as having an OCPD at the point the conductors receive their supply. Then two sets of #3/0 conductors run from the compression connector to separate devices. Each of these sets has an ampacity of 200. The point the receive their supply is the "tap" location from the 400A feeder. Each set has overcurrent protection (the 400A c/b) "ahead of its point of supply" that exceeds the value permitted for similar conductors. Each set is permitted to have its overcurrent protection (say a 200A c/b) where it terminates per 240.21(B).
Again, I'm not saying that there is any electrical logic to this, or that this is somehow safer, or that there is any specific point to be made or gained by doing this. I'm saying that this is what the language of the code requires. As we all know, the code isn't always entirely logical.
