Surface finish problems

This is my first post but I've been lurking on here for many years, thanks for all the help I've gotten! Ok, here's my issue. I have some lightening holes I'm producing in a crankshaft and they have a very tight surface call out, 4.5Rz. This is an angled hole into an angled forged surface. I use a 3 step process rough drill with a 18.5mm Corodrill, finish drill with a custom 3 flute core drill at 19.5mm, then ream with a custom 7 flute carbide reamer at 20.1mm. I've tried every speed/feed combo known to man and I can't get the finish in tolerance. It is anywhere from 6Rz to 20Rz between the 8 holes all with the same speed and feed. We started using a Flex-Hone which has helped but they are still not in tolerance and we have to hand polish them.

The machine is a Doosan SMX2600s and is new with 1000psi coolant. The material is proprietary but is similar to 4140. Monaghan has been in and talked about burnishing the hole but we can't have a step going into the 3mm radius at the hole bottom. I have a hard time believing it will work. We've also contacted a company about using a diamond hone.

Thanks all!

Helical bore the holes with a 3mm rad 18 mm bull nose mill. Relieve the upper outside cutting teeth leaving the bottom 3/16 or so of cutting teeth alone. 1500 rpm, spiral down at no more that 0.04" depth per revolution around the spiral, chip load per tooth about 0.003"

Without seeing your fixturing this is just a WAG, but I'd look at how rigidly you're able to hold the crank. If it's vibrating while you're drilling, you'd have finish and tool life issues, so it's worth making sure you can hold it in as stiff a manner as possible.

The other thing I'd try is using an endmill to cut an entry pocket to ensure a constant, clean surface for the drills to work on. In other words, if the drills are starting on a rough surface, there's a chance the tip's being bounced a bit before settling, and this might introduce vibration that propagates and hurts surface finish.