Hi all.

I have lurked here since varsity when I was building my first little machine, and have slowly been accumulating parts and designing components over the last 8-ish years - as the budget of a middle class family man (with diverse hobbies) allows...

This will be a thread chronicling the design (briefly) and construction (likely protracted) of the machine in question, which shall lovingly be referred to as the Reasonably Rigid Router (RRR) for the duration of this build thread.

I began with a goal in mind, cut all the things, and do it to 0.01mm repeatably - alas, this will likely not be realised...

The design spec evolved to where I am currently:

Machine footprint - 1538x1366mm
Usable cutting space - roughly 1250x950x170mm
I have 3Nm Nema 23 steppers - they will be driven by dm860a drivers at 65v.
2 spindles, one standard 2.2kw Chinese 24krpm air cooled flange mounted router, and a bt30 6kRPM rate spindle cartridge which will be belt driven by a 1.8kw servo.
Ball screw drive on all axes (2x 2005 on X (regret not getting 2010), 1605 on Y, 1204 on Z), with 20mm HIWIN linear guides.

There are many conventional design choices, and some unusual ones (which I am sure will be cause for comment). I suspect I will need to put little AC servos on my axes to get decent performance out of what will be a heavy machine, but the idea was a solid mechanical build primarily, which could have electrical and electronic upgrades applied as funds allow.


Where I began:


Intermediate design using square tube frame elements:


Where I have ended up (showing the servo spindle mounted)

Change to heavy C Channel because I got a bunch of scrap channel for small money.

The bed will be slightly different, with 20mm steel plate drilled and tapped with a bunch of m12 holes for clamping, and 16mm steel plate on the sections that cant be reached by the cutter.

I have started fabrication of the frame elements, and will share some pics of the process shortly.

Very simple FEA indicates an at the tool stiffness of between 4000 and 6000N/mm(10 component spring factors added in series), which according to the late D. Malicky (of this forum), should be ok for Ali.

Thanks for looking.

Luke