. .
  1. #1
    Neale's Avatar
    Lives in Plymouth, United Kingdom. Last Activity: 7 Hours Ago Has been a member for 9-10 years. Has a total post count of 1,672. Received thanks 292 times, giving thanks to others 10 times.
    This will be my second CNC router, and is to replace a current JGRO-based design which has a few of my own mods. Fed up with MDF moving every time you look at it, but it taught me quite a lot (and does work!).

    I'm requesting a sanity check on my initial thoughts, please. Basic intention is to cut plastics/wood; previous projects have gone from chunky hardwood profiling down to engraving a design on a small wooden box. I shall also need to do some 3D carving in wood. It would be good to be able to do "light" profiling of aluminium, if only to help when I get round to the CNC conversion of my vertical mill, but not a major design driver. I don't want to over-engineer, which is breaking the habit of a lifetime, but I'm trying to be sensible. Initial sketch attached to start dimensioning. I'm thinking 50x50x3 steel welded construction. I might well end up building two, one for me and one for my son, hence the design could sit on a wooden bench (with restricted speed/accel, we realise) for him and a custom floor-standing steel frame for mine. Overall cutting area big enough to go round a 1200x600 sheet, but the spindle will overhang the end of the bed to allow cutting on the edge of panels.

    I'm fed up with the DIY approach to rails and leadscrews (although my homemade delrin anti-backlash nuts on the JGRO are one of its finest features!) and intend to use profile rails and ballscrews.

    I shall reuse some of the more useful bits of the JGRO. Principally, these are 3x NEMA23 3Nm steppers, M782 drivers (bought shortly before they became obsolete and digital drivers took over), 68V power supply, and ZP5A-INT BOB. I am also using a water-cooled 2.2KW spindle with VFD. The MDF Z platform of the JGRO is sagging under its weight...

    First decision to start getting some realistic costs together is about rails and ballscrews. I intend twin X screws outside the upper X rails, single motor with belt drive. I'm thinking 2010 ballscrews for X and 1605 (1610?) for Y and Z. What size rails should I use? 20mm for X and 15mm for Y and Z, or would 15 or 20 do for all 3 axes?
    Click image for larger version. 

Name:	initial draft design.jpg 
Views:	312 
Size:	303.7 KB 
ID:	11345

    I have been looking at UK suppliers for pricing of rails and screws, as well as BST Automation, although it's quite difficult finding the right bits on the BST site - I shall need to send off for a specific quote once I have dimensions. Chai does not seem to stock profile rail - is that right? Unfortunately my little Myford lathe cannot handle ballscrew diameters like this so I shall have to pay someone else to machine them which certainly seems to put up the UK prices considerably.

    All constructive comments welcome! I feel a bit daft asking about something that seems so ordinary compared with a number of the machines described here but I can't really find a "reference" design similar enough to copy, although most of my ideas have come from reading threads on this forum so I don't think I'm ridiculously out of line.
    Last edited by Neale; 22-01-2014 at 12:04 PM.

  2. #2

    I used 50x50x3 for my frame, I knew from the outset this was a bit thin for tapped holes required to mount the profile rails. My intention is to mount (epoxy glue probably) a piece of 6mm flat strap into the upper box section so that my threaded holes will then be 9mm deep.

    I don't see any triangulation on your frame looking at the end view.

    Chai will machine the ball screws to your specification, I asked him to extend the plain area at the end in order to mount a pulley with the shaft going right through, as you say it was cheaper and easier to get these from Chai already machined. I also could not see that he sold profile rail so I bought it in the UK because it was quick and easy. I used 20mm and 15mm as you mention but I believe 15mm would have been okay all round.
    Last edited by EddyCurrent; 22-01-2014 at 01:00 PM.

  3. #3
    Neale's Avatar
    Lives in Plymouth, United Kingdom. Last Activity: 7 Hours Ago Has been a member for 9-10 years. Has a total post count of 1,672. Received thanks 292 times, giving thanks to others 10 times.
    Thanks for the comments, Eddy - I guess that my ideas are sufficiently unexceptional that no-one else has commented. Not surprising - I think I've nicked most of them from this forum anyway!

    I agree that there's some bracing and triangulation missing - across the closed end of the bed (can't put anything across the open end as I want the spindle to be able to cut off the bed - although I guess I could extend the top rails a bit further still and then put in a cross-bar) and a bit of triangulation on the bed wouldn't do any harm.

    I'm a bit nervous about starting a build, ordering ballscrews (and to a lesser extent rails), and then finding that they don't quite fit - seen lots of comments about not knowing exactly what design of support blocks, etc, are going to arrive - so next step will be to do a bit more detail on the drawings, work out exactly what I think I shall need, order the bits, and then tweak the drawings to suit and then start building. Going to be head-down in the 3D CAD for a bit...

    Back of the envelope sums seem to say 2010 for the X axis ballscrews - gives .0125mm resolution with 400 microsteps and 1:2 belt drive from motor to screws, and 6m/min is 600RPM which looks to be well under the critical speed for that length screw. Might as well make Y the same pitch so that would be 1610, and 1605 for Z. Should be fine for wood. I'll probably go for 20mm X rails and 15mm for Y/Z as much as anything because that feels about right (based on zero experience of using them, of course...).

    Back to the drawing board with a bit more focus on detail.

  4. #4
    With only 1200mm travel required, you should be able to use a ballscrew under 1500mm (perhaps 1300mm), in which case RM1610 would be fine. Bear in mind that the torque required to accelerate the ballscrew goes up with the diameter raised to the power 4, so increasing from a 16mm screw to 20mm requires 2.44 times more torque from the motor to accelerate the screw. The ratio isn't quite so bad when you factor in the inertia of the gantry, but it's still significant and best avoided.

    You've currently got the top Y-axis rail longer than the bottom one. If you go down that route, then you can increase the spacing of the bearing blocks on that rail (a bit) without compromising travel.

    Your requirements for this machine aren't remotely ambitious, so you'd be fine with 15mm rails all round, but it can make the design easier, particularly on the Z-axis, if you use 20mm rails. Just draw it and see what you come up with.
    Old router build log here. New router build log here. Lathe build log here.
    Electric motorbike project here.

  5. #5
    Neale's Avatar
    Lives in Plymouth, United Kingdom. Last Activity: 7 Hours Ago Has been a member for 9-10 years. Has a total post count of 1,672. Received thanks 292 times, giving thanks to others 10 times.
    Forgotten about the torque aspect - as I suspect that I am more likely to do smaller fiddly jobs rather than large straight cuts, acceleration is probably more important to me than ultimate rapid feed speed so 1610 makes a lot of sense.

    Y rail position subject to revision at next level of detail - can't work out if it's better to put them on the face of the gantry rails and simplify alignment or on top/bottom as shown which is better geometrically (I think) but more fiddly to adjust/align. Actually, the whole alignment thing puzzles me a bit at present and I need to do a bit more digging around to see how it's done.

    I guess that the rail size thing, especially on Z, is to give a bit more space between moving and fixed plates. I've sketched something out using 15mm rail, and starting with 20mm plates I reckon I could pocket moving plate to take the nut holder, carve a groove in fixed plate for clearance. and still have enough meat left for strength.

    I take your point about bearing block spacing - any useful rule of thumb figures for what I should be aiming at? Current sketch has overall length (between outer edge of bearings on X) as 250mm, which seems reasonable given two ballscrew feed to avoid racking; Y is 150mm (bit on the small side? But as drawn, this could go up to 200mm or so on the top rail); Z is 275mm.

    Thanks for constructive comments (and I take "unambitious" as a compliment in this case - I'm trying to come up with a good engineering compromise rather than overbuild as I usually do!)

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Similar Threads

  1. Replies: 9
    Last Post: 21-02-2014, 08:57 PM
  2. Some initial questions....
    By andy586 in forum Gantry/Router Machines & Building
    Replies: 5
    Last Post: 27-01-2013, 02:00 PM
  3. Replies: 2
    Last Post: 22-10-2011, 11:38 AM
  4. Newbie Initial Wiring/Setup
    By Rikk in forum Motor Drivers & Controllers
    Replies: 3
    Last Post: 28-06-2011, 09:19 AM
  5. Any thoughts on pendants?
    By Robin Hewitt in forum Machine Discussion
    Replies: 5
    Last Post: 27-12-2009, 06:28 PM


Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts