Unless you really crank it, the poweramp will be clean and IMO, the 'default' preamp the profiler uses when profiling clean sounds isn't very convincing.
If you cause it to distort you might get better results, but you'll have impedance and level problems - i don't know how your effect return is built, but i'm not sure it's identical to a preamp guitar input.
I've had mixed results, in short.
YMMV.
a 'problem' in the setup as you describe it is the loadbox. the speaker will cause the connected poweramp to behave differently than a simlple loadbox can.
Quitty
not sure what you're referring to.
there are loads of great, extremely clean profiles out there from pretty hi end stuff like EvilRobot and /13 that receive nothing but praise.
likewise profiles of vintage Fender amps, there even is a bunch of profiles dedicated to lap steel guitarists, and these guys know their cleans. 
a 'problem' in the setup as you describe it is the loadbox. the speaker will cause the connected poweramp to behave differently than a simlple loadbox can.
Is this a problem ?
If i want to compare, I will use :
-guitar --> KPA --> IR cab simulator (VST plugin, Torpedo...)
and
-same guitar --> head (return effect loop of the ) --> loadbox --> same IR cab simulator
There's a big difference in the cab section response between something that the profiler has recognized as clean and something that has been profiled as distorted with gain on 3.0.
Both would sound fairly clean and both may sound nearly identical if you decrease both gains to 0.0, but the cab section on the 'clean' profile would be very, very different.
It also replaces much of the response of the amp that i feel i find in gain levels under 3.0.
