Thanks,
i might just price up my own build to see what it comes out to, I'd want decent motors & drives, twin drive on the long axis with ballscrews probably, maybe HiWin rails. Water cooled spindle of some sort.

So, searching for design tips here, getting my notepad out...

1605 ballscrews ok? Gives a step resolution of 0.025mm at 5mm/200steps and theoretical speed of 3000mm/min at 5mm/600rpm

Never having used ballscrews, how do i arrive at a motor size easily? No point in throwing massive motors at it if not needed - just increases PSU, weight, cost etc. 2, 3, 4Nm??? Bear in mind I want slaved drives on the long axis. Or do i want a single drive with a fixed gantry & moving bed?

HiWin rails I think are good, based on my Plasma build, would give a very rigid motion system I think, even the small ones are good.

Frame built from Alu or steel? Alu would be a milled, bolt together design much like the 6040's steel could be welded and plasma cut.

Any ideas for a bed/cutting surface?

Feel free to chime in with any tips.

If I'm heading out of the real of Mach3 please tell me and i'll head on over to a machine build forum

My first ever retro was a clapped out manual Bridgeport, worked well and the Z was fairly easy to do, below is a pic of what I did. If you decide you may want to go that route then I can give you more details of how I did it.

Hood

Always interested in build details, when you get a chance

My BP is in fair condition i think, the table screws are worn as its slack in middle and binds at the ends so its been tightened up to remove backlash, I guess they get pulled and replaced with ballscrews?

Might be worth a thought, thanks

HIYA DAVE , The first thing YOU have to decide is WHAT TYPE of machine you want to build and exactly what do you expect out of it .

Do you want a router type, a table mill, a knee mill , gantry mill   etc etc .  Don't worry about anything else until you get that decided  .

(;-) TP

FWIW,

Modified / cnc'd quill feed for  the Atlas mill.
Works very well and if really wanted to  can quickly release and use the quill manually.

http://www.machsupport.com/forum/index.php/topic,7396.msg127816.html#msg127816

RICH

HIYA DAVE , The first thing YOU have to decide is WHAT TYPE of machine you want to build and exactly what do you expect out of it .

Do you want a router type, a table mill, a knee mill , gantry mill   etc etc .  Don't worry about anything else until you get that decided  .

(;-) TP

Ah, there goes that logic again

Not sure really, ideally would like the lot - engraving, pcb's, wood carving, making stuff from aluminium, making labels from 2-layer plastic, maybe even bolt the head and bed from my 3d printer on to it:)

Just fancy messing with A: a new machine build possibly, B: increasing my play time with CNC

Probably won't get off the drawing table but seemed a good idea at the time

The first step to design is to envision it and understand
the requirements for each envisioned task and the specs required to accomlish the task.

So spec out the requirements for each envisioned task.
- max and min size of work/ X/y/z/rotary
- accuracy
- feedrates, spindle speeds, HP, etc
to name a few................

For a general machine you end up with comprimises in how you will apply machine usage. If it satisfies YOUR needs that is all that matters.

BUT

A general machine that does everything rather poorly is just
useless!

FWIW,

RICH

Here's a picture of a machine that seems to do everything the guy wanted and more. IMHO, it's way overbuilt and could likely take a 1/2" cutter and 2hp motor but the accuracy form his tests stand up at about 0.1mm repeat which i think is pretty good.

I was pondering over a fully welded unit from 5 or 6mm plate, 4mm wall box section, plasma cut details, similar to my Plasma table but heavier and smaller - the tools used would be limited to 1/4" more likely 3mm or 1/8" and less so the forces in these machines when cutting must be pretty low, from reading it seems engraving and pcb stuff etc has practically no counterforce due to the microscopic tools.

Still undecided, but things i have picked up and noted...
1610 screws on X&Y
1605 screw on Z
3.1Nm motors seem perfectly adequate
The two Y screws can be driven from one motor and a long belt but I dislike that idea due to belt stretch etc and would use slaved motors again I think.
A heavy, rigid frame and gantry.
A bed that resists downward forces etc.
Motion controller - UC100 etc.

The first step to design is to envision it and understand
the requirements for each envisioned task and the specs required to accomlish the task.

So spec out the requirements for each envisioned task.
- max and min size of work/ X/y/z/rotary
- accuracy
- feedrates, spindle speeds, HP, etc
to name a few................

For a general machine you end up with comprimises in how you will apply machine usage. If it satisfies YOUR needs that is all that matters.

BUT

A general machine that does everything rather poorly is just
useless!

FWIW,

RICH

Fully agree, but considering people are making stuff on these dodgy chinese 3040/6040 machines proves a point that, with a little more engineering and better components, a workable unit could be made. With a little more £££ an excellent unit could be made I think.

These water-cooled spindles seem to come in kW sizes and 1.5 - 2.2kW is about the limit it seems, ER11 Collets etc

I think it would be very hard to build a machine that is worse than the £400 import lumps, or even equal, getting far better should not be unattainable.