I started another thread to ask a question in the PCB milling forum however I think my testing deserves it's own thread over here.
I was working on a dedicated mill for doing PCBs and I was looking at interesting ways to design the Z axis. after looking through a whole bunch of interesting designs (rep rap wally) I found the build thread for the Swiss Mak. I saw that he planned to use acetal/delrin on aluminum and had settled on box ways over dovetail ways. It wasn't something I had considered and the more I thought about it the more interested I became. Unfortunately I can't seem to find much information on how to design something like this and I started to do some experiments.
My requirements are milling PCB's at between 12k and 30k rpm mostly with a 500w spindle and .8mm end mills. The work area will be 200x200x30mm. typically the machine will not need to move the Z axis more than 3 mm when milling.
Here are my results so far and they are much much better than what I could expect from what you typically see in a pcb mill (8mm unsupported rails).
I'm waiting for my 'mak to be built and arrive, interested to see how long those acetal liners are going to last under metal cutting loads. But for a PCB machine where it's mostly pretty gentle, it should last forever. The only other thing is working out how to keep your carriage on the rails - one advantage of dovetails is that they hold the two together.
Be interested to see how you get on.
The guy doing the Swissmaks is a decent and approachable fellow. Although (hopefully) he's flat out making machines at the moment so replies might be slow, I'd be quite surprised if he didn't respond helpfully to questions or requests for advice you sent him via private message.
I was able to get the spindle mounted and moving. If I set a 123 block on the spindle I get about 0.005 mm deflection. If I plug in a 0.8mm cutter and .2 mm depth of cut at 12k rpm on fswizard I get .07 kg of cutting force. The run out on the spindle is about 0.015mm !!!
I believe my spindle will be limiting me much more than my z ways. I plan on converting it to just be a drive motor.
This design seems to work ok but I'll go ahead and list the issues.
1. HDPA cutting board doesn't seem very stable. Is 1" thick commercial HDPA more stable?
2. This is likely way over sized for cutting PCBs.
3. My ratio of length to width should target 2:1 however the 2:1 rule for bearing load should also be accounted for. I just happened to have this block of aluminum lying around.
4. the tapered gibs only provide adjustment for one axis of motion constraint. I wonder if it is worth it to also have the other axis of motion constrained by a tapered gib vs my current shimming approach
5. I need a better way to hold tapered gib adjustment.
6. The whole thing can be made much smaller adhering the plastic surfaces to an aluminum substrate in strips/pads. I would have tried this but I happened to have the 1" thick HDPA handy.
7. HDPA isn't all that slick compared to acetal/delrin or PTFE. I understand Rulon 142 is PTFE and I wonder if generic versions can be sourced?
8. I should have relieved the middle less so that it could act as a lubricant reservoir.
9. I'm still not sure how to keep the PCB dust out. I can make wipers but i doubt they'd help much.
10. I wonder what rule of thumb should be used when calculating the size of the keeper gib "L"s.
11. I wonder how thick the gib should be at the thin end? I used 8mm here. This weakens the keeper gib but since it is acting like a sping it may help dapen vibraiton and reduce play.
Now, while this has been a fun learning experience, it is way cheaper (unless you are not valuing your time) to go and buy a HR20 rail/block and be done with it. Since most people building hobby CNC machines are seemingly gluttons for creating lots of work for themselves this is the perfect method!
I'm wondering why box ways don't seem to be discussed as an option in most of the build threads here?
DIY box ways
Reply with Quote
