Thank you for the answers and suggestions guys. I appreciate them very much. I think Bazzer is right on to my problem. I have been also thinking if I am a parts maker or machine convertor and definately the simple answer is that I want to use the machine and not spend that much time building and tuning it. I love my machines and it is fun to take care of them and upgrade and look for parts etc but really in the cnc business I would love to focus my energy towards learning the process itself.
Then again I love the Emco machines and got a lead of a VMC-100 mill just today. It would be in original condition supposedly working and goes for around 1500e. The price point is ofcourse one motivator that could drive me to try the conversion business but then again I am really not that excited about that. I think the wabeco machine would be a perfect candidate for me as I've liked the bf-20 mill very much for what I do and that is model engines and everything related to model airplanes. Basically only for the hobby purposes and in general tiny stuff.
Since you said that anything to ask go ahead so I would like to get some more clarification on the wabeco models. There is a cheaper 2-1/2 interpolation model and then more expensive 3D interpolation model, what would one need and what are the most significant differences? All in all the wabeco starts out at 5k (without anything extra or shipping etc) and it seems that that would be at the very top of my liking for the price range so this drives me towards for looking at the used market again and that brings me back to the pretty slim pickings.
What programs is one supposed to use with Wabeco cnc mill? Is it so that you download and learn to use like Fusion 360 or Mach3 and can then export those directly to the machine nccad ?
While Vectric products tend to be router-centric, they do have a very good but somewhat dated intro sort of document that isn't vapid or stupidly simplistic.
For 2.5D work, I used Vectric VCarve for much of my drawing and CAM needs for quite some time. Still use it with my routers, still like it.
Basic quickie flow is:
1) Make a drawing you will convert to G-Code with some form of CAM. These days 3D CAD (Fusion360, FreeCAD, Alibre, Solidwork) tends to dominate the market, but for many folks 2D CAD such as LibreCAD, QCAD, and like are sufficient. You can't run a DXF or STEP file on a machine, you need to feed it through a CAM program. Some packages integrate drawing and CAM functions in a single package, or have add on packages for CAM.
2) Using the CAM application, generate G-Code for your controller - This is where you worry about feeds and speeds, chip loads, all that stuff. Some like MeshCAM will make reasonable assumption for a lot of these details, others will make you specify everything. It's also where you need a post processor for your machine, a file with the rules of how G-Code is generated by the CAM package to work with a specific dialect of GCode and the subtle or horrific details of how some vendors make a mess of things. Some CAM packages have hundreds of post processors available, some claim to do "standard g code", some have a few post processors for very generic or popular hardware. If you really, really have to, some post processors are script files, so you can build you own. My impression is this is the sort of stuff you would rather avoid. Me too.
3) Get the G-Code into your controller and run it. Your controller may be a dedicated hardware device such as FADAL, a software application such as Mach3/4 or linuxcnc, or a combination hardware / software solution such as the Centroid Acorn offerings. Whatever controller you have, it's not a bad idea to set the Z zero point a few inches (or a hundred or more mm) above the table and cut air the first time, better to find a problem cutting air than metal. It's a lot nicer to load the G Code over a network of from a thumb drive than it is to dribble it in over a serial port, assuming you have a serial port on your CAD/CAM computer. The first thing to test on a CNC machine you are firing up for the first time is the E-Stop. Odds are you'll need it at some point.
Quite a few folks are running VMC100 machines with fairly straightforward retrofits, might be a nice middle ground between a full manual to CNC conversion and fighting the somewhat unique Emco syntax and G Code variations (yup, I'm being nice and I have used Emco CNC stuff. Tore out the original electronics / controllers and plopped in nice modern electronics that speak "reasonably normal" G Code!)
For my clock shop needs, I tossed together a 4 axis Sherline mill with a PMDX 4 Axis driver box. Ran it with LinuxCNC. Did most stuff in Vectric, oddball gears were generated as DXF using Gearotic, the DXF was fed through Vectric for CAM, and it just worked. I've build CNC machines from scratch and via conversions. It is sometimes nice to just assemble from clean new standard components, turn it on, and have it work. 15 years later I still have the machine in my shop, it still works. Best part was I think the entire build of hardware and installation/configuration of software was a two day effort. Always more to learn, but very quickly I had predictable motion in all axes and was cutting parts.
Best of luck deciding how to proceed,