Thanks The_Shadow for your advice.  I left my Dillon dies stock, and instead made a push-through sizer die on the lathe that sizes case body and even head down to new round specs.  That fixed a LOT of my case forming issues.  My process became to first size and decap in a 10mm carbide sizing die, then the push-through mounted in my press I use for 50 BMG (No lube needed!), then size in the 9x25 sizer, expand and prime, charge, seat, then crimp.  Getting that area around the case head to resize was key to me, and the push-through die did the job.

Right now, Midway has VV N105, which is very similar to both Power Pistol and AA5 in stock. 

I am kind of a "noob" in the 9x25 community, though I have been a 10mm handloader for around 20+ years.  My original plan was to make my own dies with my mini lathe, but I scored a set of the Dillons and skipped that work.  Step one was making a carbide 12mm rod in to a 9x25 case image at 0.005 inches smaller diameter by using a diamond grinding bit in a Dremel mounted to the lathe tool post.  Step 2 is grinding grooves and shaping and sharpening it in to a reamer. 

Needless to say, making carbide tools at home isn't for the faint of heart.  But it can be done...

I am considering opening up the neck of my factory Dillon 9x25 dies on the lathe a little, as the neck does get overly squeezed. 

And hey, if I fail, I still have that carbide rod that I started turn-grinding that has a 1.5 inch section ground down to 11.3mm...  Getting there only tore up two diamond bits...

I fired up the mini lathe and turned an old rusted 9mm Luger sizing die in to a push-through sizer for 10mm Auto cases.  I chucked it up and drilled through the die with a cobalt 10mm drill, then used a boring bar straight in to open the bore to exactly 10.80mm (0.425 in).  Then I ran the bar in to a depth of 21mm at 10.9mm diameter, and used the cross slide set at 2 degrees to taper the 10.9 to the 10.8.  Finally, I flipped the die around and bored from the top to 10.9mm and a depth that half the case can sit in.  Of course, I cut a chamfer on the input side of the die with a 1/2 inch bit to guide the cases in easily.

It works great.  No lube needed and surface finish of the sized cases looks really good!

I could have gotten the same results with a Lee FCD gutted, but this way I salvaged a rusty die and saved buying one, and got to play tool-and-die maker in my garage for a couple of hours.  I call that a Win.

Nickel plated cases are notorious for splitting when over-worked.  Without necking down they will still split far more frequently and at a lower number of loadings than plain old brass in the 10mm.  And you just can't anneal the nickel layer like you can with brass. 

I've just done my first box of 9x25 rounds and did get my shoulders pushed back just a little bit too far when I formed the cases, which caused me to have a couple of light strike misfires.  I used the empties from rounds that did fire properly to readjust the sizing die to just barely touch the shoulder and that problem won't happen again.  After all, a fireformed case is the best way to adjust the die to your gun's chamber.

As for the neck splits, my theory is that those are caused by overworking the brass without annealing it at the right step in the process.  Perhaps the best time to do an anneal on these cases is right after necking them down the first time - that will prevent the splits seen when expanding them. 

My load workup was guided by Quickload software, and this cartridge actually seems to match the software almost perfectly.  I used Sierra 115 grain SportMaster JHP's loaded at 1.26 inch and worked up to 11.5 grains of Power Pistol powder.  Average velocity of that load was 1688 fps, low of 1671 and high of 1701 with mixed headstamp "FC" and "Blazer" brass cases with small primer pockets, and CCI 500 small pistol primers.