@SamKeller I was wondering if you'd be willing to share any photos or diagrams on how you have your 6pack/drivers wired. I'm working on a (close) copy of your machine and i'm getting kinda hung up there. I think I have it right in my head but maybe since you already have it functioning I can check my work against yours? Thanks!

 

hi
I might seems ungrateful, but has anyone typed out a list of parts? I'm in a cast and would love to order this so I have everything by the time the cast comes off, but even just typing this has taken ages. It would help if I could copy/paste
thanks

 

Sam gave a list of parts( if your talking about the cnc router) on page 1 of this thread.

 

I appreciate your reply, and I should have been more specific.
Is there a list thats been typed out that I can copy and paste into search? my arm is in a cast and typing is very tyring. I can't copy and paste part of a picture into search.
If not I'll wait till cast comes off.

 

Thank You Sam!

 

If you were to re-build this machine would you still use the DUAL z Gantry Design? Was slaving the second axis difficult?

 

Hi, I am not sure about this machine, but I am just putting together a different build, with a dual-Z (i.e. XYYZZ configuration). The capability of the control board is a key factor in running two motors on the Z. I am using FluidNC for my build, and having two motors on Y, and another two on Z is trivial, with full independent control and auto-squaring being easy. The auto-squaring needs independent end-switches on the axes for homing, but again that is only a cost and cabling issue, rather than controller restriction. Other cnc controllers may need the servo drivers to be ganged to a single control signal (which risks getting out of sync if steps are missed), or a belt/chain etc. could be used for a mechanical link.

Evan

 

It seems like the blackbox can only handle 3 limit switches, so autosquaring with that controller, isn't an option. I suppose that is why your using fluidnc

 

Amazing, thank you.

 

Yes fluidnc is not the only possible controller by any means. GrblHal as a prime 'alternative' code can support many options, depending on the underlying processor and board configuration.

My machine is a new design and I did not know how well it would actually work until I powered it all up.... For XYYZZ operation with true auto squaring on both Y and Z, you need a minimum of 5 input lines for switches, but may need additional lines too if for some reason you need to be able to differentiate each axis end. I used all 10 endstop switches as being independent lines for my initial trials so that I had options of homing at each end, but most importantly, if (i.e. when) I crashed the machine into an endstop, the machine knew which way to pull away when I restarted. I used fluidnc as it could handle 10 switch lines. Any config change is just a tweak of a text file, so is fast for debugging a new machine design, but GrblHal has the web builder which is quite quick too. GrblHal has some nice features that fluidnc does not have yet though, and I am still mulling over which way to go for the future.

I have now ganged the switches at either end of each axis, so I only need the 5 input lines, but if for some reason I manage to crash into an endstop, I have to manually wind the axis away from the stop to clear the switch before I can power back up again. Thankfully with homing and soft limits, it is rare that I have to manually move an axis as some of the drives are tricky to reach and move by hand.

 

I forgot to mention, as Peter said, if you always home Z before Y, then you can get auto squaring of both axes in GrblHal by ganging one pair of the Y and Z switches. Fluidnc does not like that option and needs independent switch sets on each axis. I have the option of homing in any order though, which is important for me as I have a secondary horizontal spindle and I swap the Y to be Z and Z becomes -Y, changing the homing order.

 

Hello Sam, I like your build.
As I have limited space in my boatbuilding workshop I'm curious whether you think that I can use this kind of machine semi upright - means leaning the cnc machine in a certain angle (lets say about 20 degrees) against a wall too. I'm on the lookout for a cnc mill allowing me to cut e.g. long boat planks from full size plywood sheets (about 6 x 1.5 meters).

I've found something used by a boat warft in Austria, and I guess it is a self built cnc mill too (I added an image just to give an idea what I'd like to build/to have).

 

Swarfer, thank you.
I guess that the motor for the Y-axis must be high-torque too for beeing able to lift the weight of the router, the router's sled and the z-axis motors weight?

 

Which axis is what? It depends on one's assumptions if it's not defined ;-).
In the image I added the X-axis is horizontally, and the y-axis is vertical.

But anyway, I understand what you mean. Thank you for the tip, it is appreciated.

 

Thanks. Good idea, I like it.

 

Since no one mentioned it I will add that you need two motors for the horizontal axis. One at the top and one at the bottom so you don't get racking.

 

For sheets of that size, and for situations where the bed is close to vertical, machines like the Maslow cnc provide an alternative configuration, BUT, it can be a case of 'if there is a problem that is hard to solve, try changing the problem'. Much depends on cost, and in particular, how precice and how fast you need the cutting to be.

 

Thank you both for clarification, and the maslow suggestion as well.
Fully agree with both of you.

My use of axis directions comes from boat building (see img). It is not a der jure, but a de facto standard used by many boatbuilders.
Thank you for telling me, that there's another de facto standard for milling machines. I did not know.
But it's never too late to learn something new .

I know the Maslow4. It is not able to cut the size I'd like to cut.
It's main drawback is the fact that the machine's overall size - this means the belt anchor points in the corners of the whole machine - is way bigger than the effective milling area.
This is dependent on the construction using four belts anchored in the four corners of the installation to move and to position the router on the milling area.
Another limitation is the capacitiy of the spools used to adjust the lenght of these belts. It is limited too.

This means that even if someone has a huge workshop wall, the max. size of the potential milling area is limited by the design of the Maslow4.
A calculation of the frame size/milling area size relationship for the new Maslow4 can be found here
http://lang.hm/maslow/maslow4_frame.html

So - my humble opinion - there's really a need to make something of my own. Depending on the fact that there exsists a rather big amount of (nerby) vertical big panel cutting saws (for horizontal and vertical cuts) I think that it will be possible to create something similar carrying a router sled instead of the circular saw as mounted and used on these machines.

Precison should be whatever is possible (a deviation of a millimeter or so is not a problem), and the speed is not of main concern.
I have some design, CAD construction and build experience (Ive designed and built a car, sveral racing motor bikes, some recumbent bicyles, and am designing, drawing plans for and building small wodden boats now for more than 14 years).
So I'm aware of the problems with sturdiness, forces and the like to be respected when designing, constructing and building a cnc machine. I'm only new in this area, and eager to learn from the best.

And with my nearby 80 years I'm old enough to be aware of the fact that I know just a few, and way too less, all the time ;-)
So every tip and hint is highly appreciated.