So I've always used rcbs dies, I couldn't find a 300 prc in rcbs so I got a hornady. They want it set till the shell holder just touches. But in a side not it says it can be backed up for different type of chambers. In the rcbs die there isn't any way to get it wrong. Once the shell holder meet the die it can't go any farther, the internals can't be moved other than the de priming pin. I haven't found any type of better instructions on backing it out. But the measurements all seem right so I don't think it's a resizing issue. Just seems like junk brass.
Any die set can "get it wrong". Ideally, you want to set the die to move the shoulder back 0.001-0.003" only. Typically, the "screw it in till it touches" method will give excessive head space, as they build the dies to work for the
shortest chamber that might be encountered.
My OEM 300 WM barrel had a particularly long chamber, and back before I measured such things, I followed the die instructions. On the third round of reloads, I got separations just above the belt, with Federal brass. I assumed it was bad brass. I then bought a bunch of MAI (Norma) brass.. After 2 reloads, I bought a comparator set, and started measuring things.
Using the standard die instructions, I was bumping the shoulder back 0.008"!. I have the RCBS case measuring station, and was able to confirm significant thinning at the web,
after only two firings.
Since then, I've been bumping that same MAI brass ~0.002" on resizing, and they've taken another 3 firings. A few are culled each reload for excessively thin webs, but most are still good for at least another firing or two. Had I not put the first two firings with excessive head space on them, I bet they'd have been good for 8-10 firings (I anneal necks).
Anyway, I would strongly suggest measuring your
actual headspace with a comparator, on your once fired brass, then again on your sized brass. My guess is you've got excessive headspace (shoulder setback).