am
Sorry for the late replies everyone, been manic here at work. The good news is that the machine is now up and running and is performing beyond my expectations. I too had issues with the collet size of the Dewalt router and so bought a reducer in the end to take it down to 6mm. I May also purchase a 3mm one soon.

The custom enclosure for the wiring is now complete minus the lid, I awaiting carbon for another job so will use a piece of that.

Thanks for your comment. Hopefully not too late for you but in short, no, I would not change anything at this time. All I did is extend the axis, add additional supports to the base, make some custom cable chain guides/mounts and purchase a taller set of side plates and work from Kyo's plans. For this build and what I needed it for it is working perfectly. It has opened up more options for my in-house fabrication possibilities and has drastically reduced project and prototyping cost, this has really helped my profit margins and project times.

I do think that it will be upgraded and probably changed over time to fit my needs though. Probably an improved spindle and side plates with holes to adjust the anti-backlash nuts. I am working on a laser attachment for it at the moment. More on that soon.

Thanks! My advice would be planning your build and if you are in no rush take time sourcing parts early on. This way you can get reduce your build cost and you won't be waiting on parts to arrive when you get into the build. Take lots of pics and get involved on here, so many people willing to help!

No worries Gary, thanks for your comments. 1500mm, in my opinion, should not be an issue when scaling up this build. There is some torsional flex in the gantry when the carriage is in the centre but its nothing I would not have expected. If anything its way less. There are ways that have been discussed that will reduce this. My initial thoughts are to use two cross pieces with a carriage that sits between them, you would lose some Y travel though. Everything is a trade off. Good tooling and slower cutting speeds are key, let the bit do the work.

Hi Andrea, As mentioned above it really is not an issue for me. I mainly cut Carbon Fibre and Aluminium. From my tests I believe I had to put over 10kg of torsional force on the bit holder before I would get deflection, the size of bits I use would fail way before then. It comes down to the right tool for the job, build a machine to fit your needs.

 

On a 2 flute 4mm flat end mill I use around 16k rpm - (as slow as my dewalt will go), I mist with lube and go slow.

Ideally you want as much cooling as possible, for me this is not a reality at the moment.
I use 1 or 2 flute bits to help with chip size and make sure I clear them out as I cut.
I cut at 40 IPM, I believe others cut much faster.

Some good articles for reference:
CNC Routing Basics: Toolpaths and Feeds 'n Speeds | Make:
Understanding CNC Cutting Tool Speeds & Feeds

And a calculator to help work out IPM:
Milling Speed and Feed Calculator

 

Just logged into work server to check.

Single flute Carbide 3mm end mill
0.5mm depth of cut
600mm min speed
50mm min plunge speed
16,000 rpm

If you have a double flute you would need to try to run at 8,000 rpm etc.......
I burnt up heaps of bits when learning.

 

Been experimenting with lots of different settings lately. A lot comes down to the tool paths you use, the size of bits and what you are cutting.

I use Fusion 360 to generate tool paths, here is my advice for cutting right up to 4mm carbon. I will update once I cut some thicker stock.

Holes up to 5mm
Fusion 360 function: 2D bore
Tool: 2.5mm bullnose carbide PCB endmill (like this 5 Pcs Tungsten Carbide PCB End Mill 3.175mm Shank 2.5mm Cutting Dia: Amazon.co.uk: DIY & Tools)
Spindle speed: 18,000
Feedrate: 500 mm/min
Vertical plunge: 150mm/min
Settings: bottom height -1mm (less if you can get away with it), repeat passes (thin bit seems to deflect slightly so the second pass really helps)

Full depth contour cuts
Fusion 360 function: 2D contour
Tool: 2.5mm bullnose carbide PCB endmill (like this 5 Pcs Tungsten Carbide PCB End Mill 3.175mm Shank 2.5mm Cutting Dia: Amazon.co.uk: DIY & Tools)
Spindle speed: 18,000
Feedrate: 500 mm/min
Vertical plunge: 150mm/min
Settings: bottom height -1mm (less if you can get away with it), single pass.

Remove material but not to full depth
Fusion 360 function: 2D adaptive clearing
Tool: 2.5-6mm bullnose carbide PCB endmill (like this 5 Pcs Tungsten Carbide PCB End Mill 3.175mm Shank 2.5mm Cutting Dia: Amazon.co.uk: DIY & Tools)
Spindle speed: 18,000-14,000
Feedrate: 500 mm/min
Vertical plunge: 150mm/min
Settings: leave 0.5mm radial stock, single pass.

then

Fusion 360 function: 2D contour
Tool: 2.5mm bullnose carbide PCB endmill (like this 5 Pcs Tungsten Carbide PCB End Mill 3.175mm Shank 2.5mm Cutting Dia: Amazon.co.uk: DIY & Tools)
Spindle speed: 18,000
Feedrate: 500 mm/min
Vertical plunge: 150mm/min
Settings: bottom height -1mm (less if you can get away with it), single pass.

 

When I first started cutting carbon I was going too fast with the spindle and too slow with the feed speed and was doing multiple passes. This just created dust and burnt up a lot of bits! Every day is a school day!

 

Thanks Paulo! Design kudos go to @Kyo however - I just modified and enlarged it. V2 will all be my own however

All axis that have wheels use 8 wheels on each carriage.

 

The "short" answer is yes. I found this out when putting it all together. I thought I had updated the spec on my post.
I used a "custom" 3mm spacer to space out the end plates, its late now and I am not near the machine (and I am full of whisky) but will have a look on monday and update the BOM.

 

Hi Benny,
I did not design these, I simply purchased them from @Chris Laidlaw via ebay:

Openbuilds Kyo Sphinx C-Beam CNC Router Plates Stainless Steel Extra LARGE | eBay

Dealing with Chris was a breeze and he was happy to help in any way. Parts arrived in under a week, can't recommend him enough!

I have designed my own version that I will make available once complete and tested, these will be a different size and include cable routeing and a drag chain mount.

 

In my opinion 3mm would not be suitable, especially on the scaled up version. There is just not enough meat for the v-slot wheel mounts to go into. I think they would be fine on the Y axis however so maybe you could adapt them in some way? buy two sets?

@Kyo has supplied the files on his build page so you could import them into your CAD program of choice and make them taller yourself.

In my case it was cheaper for me to buy them from Chris than it was to have them fabricated locally.

No disadvantages to scaling up but now that its in daily use I do have a list of things I want to add or change. Off the top of my head:

• add angled guards to the Y axis to keep debris out of the v-slot wheel tracks
• change Z axis direct drive to belt driven with reduction. This will give better resolution and reduce overall height of machine.
• add an enclosure with dust extraction and suction that will allow me to connect a hose to the spindle assembly with different options for dust/debris extraction/removal.
• replace wooden(MDF) "spoiler board" with a thick stainless steel or cast surface that will include threaded mount points.
• replace router with spindle now that they are so cheap on ebay
• redesign driver / control box and add external controls

 

Yes, exactly that. When I built the machine openbuilds had stopped making them. Now I can just make my own and include a tensioner

Now they even offer a kit!
NEMA 23 Reduction / Stand Off Plate Set

and UK supplier so I may just order one!
V-Slot Mounts : C-Beam Belt Reduction Plate

and while I am at it Belgium!
NEMA 23 Reduction / Stand Off Plate - TronixMaker