Tubesheet _Collar Bolt Design

No.

It might be interesting to know the context and motivation of this question.

No, it's just for testing the shell side, with the bonnet remove you can visually inspect if the are any leaks though the tube and the shell and the tube sheet to shell gasket.

the tube side test will be tested separately with the bonnet and without the shell for the same reason, In other words, see if the are any leaks in the tubes, bonnet and gasket.

This is similar for ASME U stamp testing for test, but the tubes need not be installed for ASME, and both the shell, and tubes would get stamped.

I'm sure it's the same reason for API.

Collar bolts are used where tubesheet employed is extended as a flange. Collar bolts (usually 25% bolts or min. 4 in no. are provided as collar bolts) serve many purposes like

1) it ensures that during disassembly of channel, the gasket between shell flange & tubesheet does not get disturbed. Otherwise bundle has to be removed completely to insert new gasket for shell flange - tubesheet bolted joint. Collar is designed for flange gasket seating condition only.

2) During hydro testing of shell side where channel/bonnet is not in place, tubesheet acts as a flange with the help of collar bolts i.e. test ring is not required in that case and assembly of channel is performed without disturbing tubesheet-shell flange joint. Tubesheet should then essentially be designed for hydro test design case, of course you can use higher allowable stresses in test case equal to 90%-95% of YP as your client spec allows.