Thread: Here we go again . . . MK4
Thank you for the comments. I understand where you are coming from and I agonised and developed this design over the last 6 months or so, including looking at fixed gantry ( I think you are referring to that blue framed fixed gantry on youtube ?). I looked at different parameters trading this and that and trying to work through all the options. I think this is version 14 !
In the end I'm pretty happy overall with this concept and think it will be pretty stiff. I also thought long and hard about the arrangement of bits to make sure slip planes were in the right direction so that bearings were not trapped and pre-loaded, and that it could actually be built in a particular order without leaving impossible joints to make.
I also wanted it to look as neat as possible so a bit of the design is for aesthetic reasons as much as anything (e.g. curved cover plates on Y axis, metal junction boxes where the energy chains end)
I'll keep you all posted, but this is probably a long build so you'll all have to wait to see if it is any good at the end. I intend to do a few simple cuts on the current machine at different feed, speed, DOC,etc and film them before it is decommissioned. Then do the same simple cuts when this one is complete to compare the finish, sound (when does resonance, chatter start kick in) plus measure the static stiffness.
These arrived recently as an early Christmas present. Another ballscrew for the Y axis, and another pair of 15mm rails and carriages for the Z axis. All the other mechanical and electrical bits will come from the MK3 machine.
I ordered from Fred at BST automation. I should mention how the ordering works for custom orders. Obviously they cannot list every part in every size to I asked for a quote on the lengths I needed. This was a bit less than the website price, which was nice. I was then asked to add various items to the basket which were 'similar' to the parts I wanted. During the checking out I had to add a note to say the actual lengths I wanted, and then wait at the payment stage for Fred to amend the price. I could then confirm the order.
It's a but of a concern that the order shows stock listed parts and lengths, and you have to take it on faith (and Fred's confirmation emails) that the parts you will receive are the ones you asked for.
But I didn't need to worry - everything was exactly the length I'd asked for. They arrived in a cardboard box, not wooden, but it was well packed out.
I immediately checked the ballscrew and it ran dead true to the eye (unlike the previous ones from other sellers which arrived bent). Bearings were a nice fit on the machined ends, and the end machining was nicely done.
Last job was not forgetting to confirm goods received on AliExpress, which then releases the payment to Fred. Note how this is different to ebay. Overall very pleased with my first order from Fred. Would recommend.
Are the ballscrew end bearings any good? The cheap ones I've seen, admittedly >2 years ago, all seemed to be quite poorly made as they exhibited end float, unless you fiddle about adding shims or better bearings. Have they improved?
For you ballnut mounts, I hope your spindle is accurately trammed, else the the surface the ballnut flange mounts on may not be perpendicular to the base within sufficient tolerance.
Looking at your design, I'm fairly sure that the weakest point will be the Y-axis. You've compensated for the small bearing spacing causing racking by using two ballscrews, which is good, however the carriage can still twist about the X-axis when a force is applied parallel to Y, due to the vectical compliance of the 20mm rail bearings. Assuming you got medium pre-load bearings (I hope not zero), their vertical stiffness will be about 274N/um. The cutter is central, so if we assume the force is distributed equally upon the Y bearigs, but in opposite directions, you'll get twice that stiffness as the bearings act in pairs (I think), so 548N/um. Bearing spacing looks like 78mm and about 200mm to the tool tip from the bearings in the Z direction. This means that the deflection at the bearings is amplified at the tool tip by a factor of 200/78 = 2.56. The stiffness of the bearings is therefore effectively reduced by the same factor, so 548/2.56=206N/um. Hmm, too be fair that's a fair bit higher that the ballscrew stiffness (I estimate ~50N/um for the two in parallel) so maybe it wont be dominant.
EDIT: Just noticed that's you've put the ballscrews on the ends of the X and Y carriages, not at the center (of stiffness), so the stiffness will be a bit worse.
Last edited by Jonathan; 01-01-2016 at 04:01 PM.
I've not opened the bearing packets yet, but I'm expecting to have to make up a shim like I did last time. I've been able to make them work on the current machine so am not too concerned.
The spindle is as true as I can get it - note I'm at the mercy of the flatness of the supported rail on X, and on the flatness of the extrusion for the Y rail. Based on things I've made from the MK3 machine I think it will be OK. If not there are ways and means . . .
Double Y ballscrew design went in at the outset because of the reasons you mention - it cannot rack so you can push the bearings closer together. But you are right, and it bothered me for ages, that for loads in the Y direction, where it rotates around X, there is a moment between the cutting tool (reaction) and the ballscrew (input) which is resolved in the bearings. As they are close together this moment is higher, meaning more deflection.
Ways around this are to lower the ballscrew and reduce the moment - I tried between the box sections but this makes the gantry even wider fore/aft. I tried underneath, which is very good for forces, but it would get swarf on it and if it hit a clamp it would be nasty. So it went on top. The mori-seki is like that so that sealed it.
p.s. What did you mean here?
". . .Just noticed that's you've put the ballscrews on the ends of the X and Y carriages, not at the center (of stiffness), so the stiffness will be a bit worse. . . "
Bearings: Ok, thanks for the info - I'll continue to give those a miss.
Mounts: Sounds reasonable.
Stiffness: Understood - I wouldn't say Mori-Seki doing something one way is any reason to copy, as the relevance of their reasoning is unknown. Also I'm not sure my calculation in the previous post is any good, as the bearings are close they probably can't reasonably be considered as points, plus I ignored the position of the ballnut.
P.S: See here, from page 53.
Well that link is quite read. OK, I think I get what you are saying there. I did initially put all ballnuts in the middle of the axis, at the centre of stiffness. I hadn't though about it in the way the article suggests, it was just an intuitive start point, but the flange on the ballnut means they need alot more space which pushes everything apart.
So the X axis ballnuts are now on the 'front' of the gantry, whereas the article suggest they should be further back in the middle of the gantry. Problem is that they would hit the X bearings and make everything wider which knocks onto the frame and bed. I've only got so much space and I had wondered about an enclosure in the future so don't want to go past the edge of the table with any parts.
The Y axis ballnuts are also on the side plate of the Y axis, not in the middle of the bearings. Again the flange would push them higher in Z, and they needed much more material under the bearing mounts at each end of the ballnut to support them. It all looked too tall when I drew it.
To summarise I think what the article is saying is that because things will twist and rotate, you should put the ballnut in the centre of rotation (that is rotation of the structural parts caused by moments) so that the nut sees minimum binding loads etc. I think that is a good principle to aim for where ever possible, however I think in practice the machine I've drawn will see relatively small loads at the ballscrew due to this deflection causing a radial / binding load on the ballnut and I'm not too concerned.
Last edited by JAZZCNC; 02-01-2016 at 12:41 AM.
Dean: thanks, noted. If I ever design a machine where their form is not a constraint, I'll consider using them.
Last edited by Jonathan; 02-01-2016 at 03:20 AM.
As I feared the return to work after Christmas means workshop time has significantly reduced. For various reasons I've only been able to get about an hour in there.
But I have been able to get a few sessions on the CAD and this has meant I was able to go over some of the other designs and have one last go at unlocking some of the compromises. The net result is I've developed some of the other ideas and ended up with what I think is a much better design.
Old one for comparison:
Here is the new version:
Just X axis showing gantry, ballscrews, epoxy (orange), and custom ballscrew floating end as the ballscrews were not long enough to mount the standard floating end.
X axis drive and belt tensioning arrangement
The new features are:
Gantry beams smaller
They are now 80x40x5 RHS steel (down from 100x60x5). I'd put too much emphasis on huge sections, whereas with a double beam gantry I could afford to scale them down and still have plenty of stiffness in reserve over a single gantry.
Because they are smaller I was able to re-configure the X-bearings, bring them closer together, and give more travel in X. They are now only 274mm apart which is getting close to my current machine spacing. I was also able to get the X ballnuts at the centre of stiffness, rather than at the ends of the gantry.
Y rails are much lower
With the smaller gantry sections the Y rails are now lower and closer to the tool, as are the Y axis bearings, which all provides more stiffness
As the gantry sections are narrower I was much happier putting the ballscrews on the front and rear faces as there was much less of a bracket required to join them back to the main Y axis, therefore stiffer.
These are now much lower in Z which also puts them much closer to the tool, which reduces the moments, which lowers the forces on the Y bearings and makes the machine stiffer.
I was also able to put the ballnut in the centre of stiffness, rather than on the outer edge of the Y axis. There might be a marginal gain here I really don't know, but it does look nicer.
Y axis bearings spacing
By re-designing the gantry end plates to free up some space I was able to make full use of the linear rail and spread the Y axis bearings out considerably more than before - without losing travel. This should significantly improve the stiffness due to moment inputs when cutting in the Y direction. When coupled with the lower ballscrews there should be a double win here.
In one of my earlier posts I suggested that double ballscrew on Y eliminates racking - whilst this is true for rotations about Z axis, it is not for rotations about the X axis. To eliminate these you could add another 2 ballscrews lower down, but that is not practical (!) so you do still need to space the Y bearings out even with double Y ballscrews. Rotations about the Y axis are dealt with by having the double gantry beams no worries there.
I've gone with something a bit different here which does not use tensioning idlers. The steppers are on plates which are slotted, and the steppers are also in slots. Between them I should be able to tension the short belt up to the ballscrew, and the syncronising belt across to the other stepper at the same time by pushing the motor off into one corner. Everything is 5 HTD with 15mm belts.
I've decided that there are a couple of options to place the radiator, plus the option just to go for the 'big metal bucket' out of sight. So I'm going to build it and then just see which one takes my fancy. Sometimes you can CAD things too much and get tied up in the last details.
Luckily the new design retains the parts I'd already made so nothing lost. Thank you for the comments made so far, stirring up the doubts I had about some aspects, and making me revisit the previous designs. I think it is all the better for it.
Right, time to start CAMing up some of the parts ready for whenever the next workshop session is . . .
Last edited by routercnc; 13-01-2016 at 10:34 PM. Reason: added more pictures
The Following User Says Thank You to routercnc For This Useful Post:
This lot arrived recently.
2off - RHS for the raised X axis sides - seems pretty square and reasonably straight
2off - 80x40x4 RHS for the gantry beams
1off 30x5 flat bar to reinforce the gantry beams where the rails will sit
2off ballnut mounting blocks (~£9 each)
And 4off 16mm open bearings to double up on the Y axis to make 4 per side (no photo)
Also, got a bit of time this week end to nearly finish the X axis motor mount / belt tensioning plate:
Laying it out:
More cutouts and slots:
Holes spotted, top skimmed, and profile roughed and finished:
Bought a new 6mm 2 flute 45deg helix (for aluminium) carbide from 'cncpoorboy' on ebay (for about £9) recently and just tried it out.
Very impressed with the finish:
Some work on the drill press:
Need to skim one side and tap the M5 holes for the stepper, then its another one done.
In case you were wondering the slot will hold the head of 2off M8 bolts to stop them rotating whilst allowing it to slide in the slot. On the other side will be a spacer and a bearing guide to tension the belt.
Last edited by routercnc; 25-01-2016 at 09:21 PM. Reason: correction
The Following User Says Thank You to routercnc For This Useful Post: