Mick H's CNC Router Build

Mr. Mick H. He has been following the build and making great progress. He, just like the other fellow builders have added their own enhancements, which really makes them leaders in building these CNC machines. He noticed that the table, at the sixe of 24" x 48" will deflect a little as the gantry reaches the mid point of the table. He took his reinforcement to another level, shall we say he "spiced it up a bit". He also went the gluing route, as opposed to a mechanical fastened build. With his clean and precise workmanship, as noted in the images, it is a very smart way to go. Remember, if gluing is a consideration, your measurements will need to be somewhat precise to insure a tight fit around the rails.

From Mick (mostly verbatim): I decided to follow your plans initially and use your original dimensions, then once the machine is up and running, cut new parts for machine No2. I am now at the stage of a 90% complete build as far as the mechanical components go, leaving the addition of the X-Axis lead screw & all 3 motor mounts as the final items on the mechanical checklist.

Upon assembling the cutting bed and end supports, along with the addition of a bottom sheet of 4' X 2' MDF, then adding the gantry, I have observed a significant amount of sag within the cutting bed when the gantry is moved toward the centre of the X-Axis. I haven't measured exactly how much sag is present, but it is visually apparent...a few mm approximately.

This in itself is not a problem, these situations arise to test us mentally and I think I may have a solution...or two.

a) Re-design the cutting bed to utilise a torsion-box and extend the gantry sides downwards to compensate for the extra depth. (An increase in build time/cost, due to cutting/glueing-up the torsion box pieces and waiting for the glue to dry and re-shooting video footage for this process. Also, motor mounts would have to be moved, which would necessitate re-shooting footage for the X-Axis end plates etc.

b) Use the design as-is, plus an extra 16 skate bearings, nuts & bolts, and some more angle alu.

I'll elaborate on b)...

What I propose, is to utilise the machine as per the original design, but add extra linear bearings and angle aluminum on each side of the length of the cutting bed. The lower span members of the gantry, beneath the cutting bed, could benefit from linear bearings above & below riding on aluminum angle attached to the bottom sheet and the underside of the cutting bed. This way would eliminate any sag in the bed, provided that the machine was placed on a level surface in the first place.

The more I think about it, the more I'm inclined to make up a torsion box bed...I'm currently sizing up how many 2" squares and 48" X 2" longitudinal rails I will need. I calculate 9 rails and 136 squares should give me a pretty solid bed that I can utilise on a second machine. Additionally, I will have to redo both end plates to allow for the extra height - around 7", and the original dimensions of the gantry sides will suffice using new material and the originals as templates shifted into alignment for the mod...there is plenty of height above the Y-Axis rail that I'm glad I hadn't trimmed yet!! Another 3 boards should get me where I want to be.

I like the idea of the hardware stored beneath the machine, it makes for a neater setup, but I'm not sure about the keyboard & monitor accessibility (he's talking about this setup under "Pimp My CNC"). Are you making any allowances for dust protection/extraction once you start making chips...eg a vacuum attachment? There was a neat idea posted on CNCzone, utilising brushes nabbed from the guy's wifes ice scraper...something to think about, but before I jump ahead I need to sort out a motor/driver kit. (From Patrick: I do plan on video-ing a vacuum attachment).

The image shows linear movement on all 3 axes, with the addition of lead screws and cutting bed end plates to be added.

Z-Axis image shows where I have reduced the amount that the router attachment plate will stick out from the Z-Rail.

Note that the grooves for the linear bearings are routed out completely, slightly undersize, then the corners are cut with a Tenon saw at 45 degrees to give a snug fit. Please note also, that the bearings use double nuts...this is not because I wasn't happy with your idea of tapping into the angle, but due to wanting to get building and settling for the 19mm X 19mm X 1.5mm angle available at the hardware store locallly (B&Q in the UK) In addition, the angle used on the edges of each axis is 15mm X 15mm X 1.5mm, for your reference, in order to reduce the amount of excess material overhang.

90 percent mechanical build.jpg shows the addition of the X-Axis end plates and lower support, along with lead screws on Y & Z. Note the supporting blocks beneath the cutting bed...this is where the sag was observed when the blocks were removed.

Soooooo, I decided to see how much sag was present. (Never actually measured the exact amount though)

Shimmed bed.jpg shows the cutting bed levelled in relation to the bottom support, at approximately mid-way along X.

Shimmed close up.jpg shows match-marks made as a reference point, before shims removed.

No shim sag.jpg shows how far the bed flexed when the shims were removed. As I didn't actually measure the sag, I can only guesstimate it to be aroung 4.5 to 5mm from the pics.

At this point I had a little thinking to do. The sandwich idea seemed quite attractive as an initial fix, but I'd prefer a more permanent option that can be utilised on another build. Swallowing the extra material, along with the extra labour, I have chosen to go along with a torsion box bed. This is a method heralded for it's rigidity, using only simple fillers such as cardboard or similarly slim materials as spacers, in conjunction with an adhesive to bond the filler to each upper/lower surface.

Interior doors are commonly constructed using a cardboard honey-comb structure, bonded between two outer skins.

Maybe I've over-engineered the idea here, but I have chosen to use what I'm currently working with...and that is MDF in handy sheets of 4' X 2'.

Firstly, due to the dimensions of the cutting bed and the linear bearings, a slight reduction of width is required to allow a little clearance. Initially, I had planned to use a 2" spacing between each beam and spacer, but that would have required what I previously calculated to be 9 longitudinal rails and over 130 spacers. I have had a re-think about the way torsion boxes work and have concluded that rails perpendicular to the axis, with fillers in between would be a more viable option. So, I have chosen to go with a wider spacing, at 3", and use only two longitudinal rails which will be visible from outside the bed.

Spacing block example.jpg sort-of shows the use of a template to aid in the construction of a torsion box. As previously stated, I have chosen to use 3" spacings, so the spacing block with the four screws is constructed to achieve this dimension. The 2" high longitudinal rails are laid, then a lateral rail butts against the X-Axis end plate (The pic doesn't really show this...but assume the piece with pencil scribble is butted against an end plate). The exposed top edges are tacked, using a staple-gun and a filler piece is butted against the right hand side of the spacer block. The spacer block is then moved up and over to the next empty cell at the right of the recently-added filler, another filler is added. tacked in place, then the process repeats until the width of the bed is populated. Move up a row and repeat.