Really, the effective-stress friction angle phi' never equals zero or anything close. So-called phi=0 analysis just uses an undrained strength that is not affected by changes in total stress during rapid loading. If you are looking at long-term performance, you can take advantage of phi' in compacted or dense material, or material above the water table. Only if you are looking at rapid loading of saturated NC to lightly OC clay would you actually set phi equal to zero in the analysis. It is possible that you would need to consider undrained loads on the wall if the wall was built, the backfill became saturated, then the load was applied very rapidly, perhaps by a loaded dumptruck driving close by. In most cases that I can think of, you wouldn't worry about this.
I don't like the term "phi=0 analysis" because it is semantic nonsense. It should be called "undrained-strength analysis."
