Don't know if this thread is dead yet, but I'll put my 2cents in:
I've worked with a handful of different tipping buckets, and I've noticed something that I'm calling 'instrument interception' with just about all of them. This is the left over rain that sticks to the cone, or the debris screen, or is left in an untipped bucket. This rain eventually evaporates and doesn't get recorded.
Has anyone ever seen how much screen there is on a TE-525? There is probably just as much surface area on the screen as there is on the cone. I've checked on these gages after a storm and seen enough drops hanging on the debris screen to count for a couple of tips. My guess is that in low intensity rainfall events, these tippers are under-representing the total.
I've haven't done the research on this, but I would imagine that a similarly sized orifice on a bulk gage, would probably record more rain than a tipper would.