Desktop 1000 mm x 600 mm Gantry Router

[Soyb]

My aim is to build a desktop router utilising 43 mm Kress spindle, it will mainly be used for cutting wood and aluminium plate.

It will be a hobby machine but I would still like it reliable and accurate as I intend to use it mainly for cutting birch ply for RC model aeroplanes and would like them to 'click' together. I will also use it for all the other fun DIY projects.

I don't have a workshop, my only tool is a Stepcraft D600 and 3d Printer, so the design has been tweaked so I can machine most of the parts myself. The only two pieces I will have to subcontract are the top and bottom Z axis plates as I cannot get the vertical accuracy.

I am trying to balance the complexity / weight / accuracy

All the axis are direct drive so there are no pulleys etc

The design utilises :-
1000 mm 1605 C5 twin ballscrews on the Y axis
750 mm 1605 C5 ballscrew for the X axis
200 mm 1605 C7 ballscrew for the z axis
NEMA 23 74 mm stepper motors on all axis
CW5054 stepper drivers
400 Watt PSU
KK01 Breakout board
All the extrusion is standard off the shelf profiles (Oozenest and KJN)
HGR15 linear rails and HiWin carriages on all axis
Low profile t Slot plate (vacuum tables)

Cutting Area should be 850 mm x 550 mm with 70 mm vertical clearance

[CharlieRam]

Try to imagine the cutting forces on the tool and the movement you will get on the sides. There is a gantry force calculator somewhere on this site as well as lots of ideas for gantry designs that work well .....don't be disheartened though, I got to design 14b before I got any real positive feedback

Charlie

Sent from my M2003J15SC using Tapatalk

[Soyb]

Hi Charlie, thanks for the feedback. When designing the machine I considered various issues with possible flex in the various axis. Like you the ability of the gantry to move due to cutting forces concerned me. Although I have not done any complex calcs working through the axis, my logic is......
Z axis - 4 Hiwin carriages on 15 mm rail will easily support 5Kg of the spindle, carriages and plate
X Azis - 4 Hiwin carriages on 15 mm rail, I am using closed sided aluminium extrusion plus the rail, as long as I don't compress the axis when setting up there will be virtually no flex, it needs to support 12kg (see picture of extrusion)
Y axis - again 4 carriages on 15 mm rail, in compression this shouldn't be a problem, the only issue is the sideways cutting forces as you pointed out. I am using 8 mm bolts to attach the gantry plates and they will be embedded in 2 mm pockets as will the carriages and ballnuts. This leaves about 150 mm by 750 mm 'box' shape that has the potential to deflect. I am using 10 mm cast tooling plate. My gut feeling is this will not flex since the cutting forces for a 1 kw spindle will be comparatively light



If there is too much flex I have a couple of get out of jail options
1) add a cross beam, I would prefer not to do this as I am planning to use the space under the machine to house the electronics
2) Add another rail on each side with more carriages, again more complexity, weight and cost





40 mm x 40 mm KJN aluminium extrusion

I am starting to cut soon so I will have to fix the design

Thanks

Ian

[Soyb]

Slight change to design, working on my KISS principle (Keep it simple stupid) I have removed the longitudinal extrusion between the Y axis plates, the 10 mm plate supporting the machine and Y axis ballscrews will be of sufficient strength.



8 fewer holes, reduced cutting and plate, removed 2 pieces of extrusion.

I was hoping to have two Y axis plates across the whole width of the machine but I cannot cut this length and to buy them will be too expensive so I have opted for a '4 feet' design.

[CharlieRam]

Hi Charlie, thanks for the feedback. When designing the machine I considered various issues with possible flex in the various axis. Like you the ability of the gantry to move due to cutting forces concerned me. Although I have not done any complex calcs working through the axis, my logic is......
Z axis - 4 Hiwin carriages on 15 mm rail will easily support 5Kg of the spindle, carriages and plate
X Azis - 4 Hiwin carriages on 15 mm rail, I am using closed sided aluminium extrusion plus the rail, as long as I don't compress the axis when setting up there will be virtually no flex, it needs to support 12kg (see picture of extrusion)
Y axis - again 4 carriages on 15 mm rail, in compression this shouldn't be a problem, the only issue is the sideways cutting forces as you pointed out. I am using 8 mm bolts to attach the gantry plates and they will be embedded in 2 mm pockets as will the carriages and ballnuts. This leaves about 150 mm by 750 mm 'box' shape that has the potential to deflect. I am using 10 mm cast tooling plate. My gut feeling is this will not flex since the cutting forces for a 1 kw spindle will be comparatively light



If there is too much flex I have a couple of get out of jail options
1) add a cross beam, I would prefer not to do this as I am planning to use the space under the machine to house the electronics
2) Add another rail on each side with more carriages, again more complexity, weight and cost





40 mm x 40 mm KJN aluminium extrusion

I am starting to cut soon so I will have to fix the design

Thanks

Ian

It's not so much the deflection as it is the lack of rigidity which will lead to chatter, premature tool wear and tool breakage ect, probably ok for woods but I would imagine you'll struggle cutting aluminium.

Charlie

Sent from my M2003J15SC using Tapatalk

[Clive S]

I would consider 10mm pitch screws on X and Y and 5mm on Z

[Soyb]

I have to admit I had not thought about vibration / resonance, with an open x axis there is a potential for chatter

[Soyb]

Hi Clive, why larger pitch ballscrews? from a newbie I assumed that 5 mm would give better control, less backlash etc

[Clive S]

Hi Clive, why larger pitch ballscrews? from a newbie I assumed that 5 mm would give better control, less backlash etc

Steppers have higher torque at lower speed . Also you will get less whip with a lower speed on the screw

[JAZZCNC]

Hi Clive, why larger pitch ballscrews? from a newbie I assumed that 5 mm would give better control, less backlash etc

The pitch has nothing to do with how much backlash you get and everything to do with the speed and resolution. Clive is correct that for a router that will cut a range of materials you are better with 10mm pitch on X & Y as it covers a much wider range of feed rates for the various materials correctly. Also, like Clive says it allows you to keep the rpm down which gives higher torque where you need it, which is when cutting.

If you are planning on cutting mostly Aluminium then a 5mm pitch would be better as it allows higher resolution, that said 10mm pitch still gives a decent resolution. However, if you are mostly cutting aluminium then you'll need a stronger machine than what you are planning now.
The design you have now is ok for an all-around machine, though it could use a little beefier gantry sides, it would also benefit from using my "L" shape gantry design which will be much stronger than what you have shown. If you do stay with your gantry design I would drop the plate at the rear on the x-axis carriage as you will find it a pain to set up and doesn't add very much strength.

I have built lots of machines using a similar design with higher sides and the design is stronger than people realise, so don't let the naysayers put you off because as an all-round material cutting machine they work very well. I also have customers who run very very successful businesses with my machines which are very similar in design.

Some advice would be to use 20mm rails rather than 15mm as the bearing size is easier to work with and just suits a machine like this better, the cost difference isn't much more and worth the extra. 15mm bearings are fiddly little things.

Use 15mm plate as a minimum for gantry sides and Z-axis. Remember the Z-axis is the most important area as it holds the cutting tool so if it's weak and vibrates then doesn't matter how strong the rest of the machine is you'll get rubbish finish and excess tool wear.

DUMP the KRESS, they are toys compared to water-cooled spindles.

DUMP the KK01 Bob and buy a decent Ethernet motion controller like the AXBB-E and run it using UCCNC, this will be the best investment you'll make.

Lastly but not as important as the controller but still does make a big difference, I would use higher power drives than 50Vdc and if you can afford it go with closed-loop steppers as they are SOOOO much better than standard steppers.

Building a good machine is all about the balance of electronics and components along with a strong design, if anyone is weak the machine will suffer.

Good luck.