Circuit board milling tooling

[synapsis]

My homebuilt CNC machine is getting close to actually having an endmill put into it. Just need a bed, a coupler, and some dust control.

I'd like to use it to mill PCBs because I'm just about over the chemical etch process. Plus the UV treated boards are almost twice the price of just copper clad boards.

What tooling do you recommend? I'll be doing surface mount as well through hole.

[Experimentonomen]

That construction does not look very sturdy, for pcb milling you need the absolute sturdiest machine you can get, you also need zero backlash = your acme/allthread leadscrews arent anywhere near good enough.

What you have is barely good enough to make a cheap chinaclone 3d printer.

[fcb]

You've built the mill now - try it.

But I'd not worry too much about dust-control at this stage, tape some copper-clad down and go for it!

All CNC mills have backlash (exotic linear motors excepted), and they all have a mechanism to deal with it (usually in software).

[Experimentonomen]

Ball screws have undetectably small backlash.

And your upright for the Z axis is gonna flex like a rubber band. Its the equavilent of cranking the quill out all the way on a bridgeport mill.

[madshaman]

How are you ensuring that the spindle is the same distance from every point on whatever bed you install?  That is, how did you ensure the z-axis is perfectly orthogonal to the xy motion?  I can't see clearly in the picture how you can adjust the alignment.

[fcb]

Ball screws have undetectably small backlash.

Depends on your application if it is 'undetectable'.  My last machining centre had very expensive ball-screws, still had backlash compensation. For machining PCB's (basically no cutting force), the OP could use "rubberbands" to preload the lead-screws.

And your upright for the Z axis is gonna flex like a rubber band. Its the equavilent of cranking the quill out all the way on a bridgeport mill.

I doubt it in this application. Whilst I wouldn't to do anything vaguely heavy (machining) with the OP's mill as the neck doesn't look too rigid, I can't see why it would affect PCB milling.

How are you ensuring that the spindle is the same distance from every point on whatever bed you install?  That is, how did you ensure the z-axis is perfectly orthogonal to the xy motion?  I can't see clearly in the picture how you can adjust the alignment.

Quite simple, use the mill to machine it's own bed flat, or at the very least get it to machine the sacrificial slab that the PCB is going to sit on.

[RodG]

How are you ensuring that the spindle is the same distance from every point on whatever bed you install?  That is, how did you ensure the z-axis is perfectly orthogonal to the xy motion?  I can't see clearly in the picture how you can adjust the alignment.

Quite simple, use the mill to machine it's own bed flat, or at the very least get it to machine the sacrificial slab that the PCB is going to sit on.

Another way is to install a probe and use "Auto Levelling" software to modify the G Code.
http://www.autoleveller.co.uk/ is what I use on my rickety machine.

[synapsis]

The neck is mounted with slotted joints and made from 2" x 3" solid aluminum. I weigh 155 lbs and can hang off the slide and it will deflect a couple thou at the most according to a micrometer. Laterally I haven't tested yet since the whole mill slides on the table. The whole thing is made from solid aluminum. I can still triangulate it later.

The lead screws are preloaded for backlash. I have a linear encoder design on another bench that will go on in the future.

The bed hasn't been built yet, and will be leveled by the mill.