[Sapele]

\r\n \r\n Thanks everyone! Many of the articles and resources that I had bookmarked now make more sense to me.
\n
\nI am going to pursue two options at the same time.
\n
\n1- Ballscrews for the x-axis. Especially now that I understand pitch. I had been picturing TPI. Now I understand that a higher number is a coarser pitch. I understand that this approach will hinge on finding the right screws at an affordable price.
\n
\n2- Rack and pinion for the x-axis. I had fallen into the upgrade trap and thought that gearboxes were an upgrade. Now I understand that I can pursue belt reduction which will reduce the width of my machine (I am picturing these kits PRO Rack and Pinion Drive, Nema 34 | CNCRouterParts)
\n
\nI am still hoping to reduce the protrusion of the stepper/servos. I also realize that ending up with an oddball solution that saves an inch or two of width is not worth it.
\n
\nspeaking of which here is a close up of my space saving idea (I don\'t think that I had described this well in an earlier post)
\nAttachment 220212
\n
\nI will work up Wizard\'s suggestion of top mounting the drive assembly. I dragged a few components onto a 1530 extrusion
\nAttachment 220214
\n
\nI don\'t think I can implement the in gantry solution that Louieatienza posted. I really enjoyed seeing it. Thanks!\r\n \r\n

[wizard]

\r\n \r\n

\r\n

\r\n

\r\n

\r\n \r\n \r\n

Thanks everyone! Many of the articles and resources that I had bookmarked now make more sense to me.
\r\n
\r\nI am going to pursue two options at the same time.
\r\n
\r\n1- Ballscrews for the x-axis. Especially now that I understand pitch. I had been picturing TPI. Now I understand that a higher number is a coarser pitch. I understand that this approach will hinge on finding the right screws at an affordable price.

\r\n \r\n

\r\n

\r\n

Or just accept slower rapids.
\r\n

\r\n

\r\n

\r\n

\r\n \r\n 2- Rack and pinion for the x-axis. I had fallen into the upgrade trap and thought that gearboxes were an upgrade. \r\n \r\n

\r\n

\r\n

It depends upon what you need and how compact you need to make it. The real problem with gear boxes is the expense of the gear box, sometimes it is worth the expense, sometimes not. Personally I\'d try to work out a suitable solution using belt drives first. If that does not give you the right gear ratios in the allotted space then consider a gear box and a ballooning budget.
\r\n

\r\n

\r\n

\r\n

\r\n \r\n Now I understand that I can pursue belt reduction which will reduce the width of my machine (I am picturing these kits PRO Rack and Pinion Drive, Nema 34 | CNCRouterParts)\r\n \r\n

\r\n

\r\n

If you go with leadscrews and stepper motors it probably will make sense to direct couple those steppers to the leadscrews and avoid any gearing. At least on some of your axis.
\r\n

\r\n

\r\n

\r\n

\r\n \r\n I am still hoping to reduce the protrusion of the stepper/servos. I also realize that ending up with an oddball solution that saves an inch or two of width is not worth it.\r\n \r\n

\r\n

\r\n

Honestly you need a clear space around the machine anyways unless you goals are to have one side up against a wall. Even then with your designs you need clearance. The best arraignment would be for one end of the machine against a wall. Yes it sticks out but you get access to both sides of the machine for load/unloading and system maintenance.
\r\n

\r\n

\r\n

\r\n

\r\n \r\n speaking of which here is a close up of my space saving idea (I don\'t think that I had described this well in an earlier post)
\r\nAttachment 220212\r\n \r\n

\r\n

\r\n

Now I understand! Question though, how is that floating beam supported? There is one advantage here in that the rack is protected from stuff falling directly on it.
\r\n

\r\n

\r\n

\r\n

\r\n \r\n I will work up Wizard\'s suggestion of top mounting the drive assembly. I dragged a few components onto a 1530 extrusion
\r\nAttachment 220214\r\n \r\n

\r\n

\r\n

That is what I was thinking though on steel tubing. The nice thing here is that getting the rack parallel to the X axis linear bearing is easy.
\r\n

\r\n

\r\n

\r\n

\r\n \r\n I don\'t think I can implement the in gantry solution that Louieatienza posted. I really enjoyed seeing it. Thanks!\r\n \r\n

\r\n

\r\n

All of us run into tooling limitations. That is why so much effort goes into making simple solutions that are good enough for the task at hand. Which brings us back to those gearboxes, sometimes a gearbox is the simple solution. This especially the case if you have limited ability to machine parts for this router.\r\n \r\n

[louieatienza]

\r\n \r\n

\r\n

\r\n

\r\n

\r\n \r\n 1- Ballscrews for the x-axis. Especially now that I understand pitch. I had been picturing TPI. Now I understand that a higher number is a coarser pitch. I understand that this approach will hinge on finding the right screws at an affordable price.\r\n \r\n

\r\n

\r\n

If you can find precision ground ballscrews, they typically have higher RPM ratings than a rolled screw of the same lead and diameter. The good thing is that many people gravitate toward the finer pitched ground ballscrews, making the coarser pitched screws more available (and cheaper) if and when you can find them on eBay.
\r\n
\r\n

\r\n

\r\n

\r\n

\r\n \r\n 2- Rack and pinion for the x-axis. I had fallen into the upgrade trap and thought that gearboxes were an upgrade. Now I understand that I can pursue belt reduction which will reduce the width of my machine (I am picturing these kits\r\n \r\n

\r\n

\r\n

Some gearboxes are definitely an upgrade, if you can afford them or are fortunate enough to find good surplus or NOS ones. If you familiarize yourself with some of the major brands of planetary reducers (Parker, Wittenstein/Alpha, Thomson, Apex) and their part numbers it becomes easy to identify the base size, reduction ratio, the accuracy and whether they are stepper or servo grade. Sometimes the accuracy is implied in the model name (i.e. DuraTrue, UltraTrue). A servo grade reducer won\'t necessarily work with steppers since the bores on the coupler are usually larger metric sizes and the pilots (the round protrusion at the motor base) are usually larger as well. Also some reducers are meant to be oriented in one configuration only and it helps to know that as well (though this usually is the case with right-angle gearboxes.) 10:1 may be fine for stepper use, though it would depend on your pinion size. I would say the smaller the pinion, the lower the gear reduction you could use but going too low would give you decreased resolution. A larger pinion might offer more tooth engagement and quiter operation.
\r\n
\r\nAtlanta Drives also makes integral tracks that work with major linear bearing manufacturers. This can also help save space as well as eliminate the alignment of rack to linear rail, as they are both integral. It would complicate things a bit, but I can\'t see why you couldn\'t have the drive motors mounted vertically in the gantry tube complete with whatever reduction you use, and drive the pinion via a jack shaft, though this would require being able to move the entire drive assembly as one unit for adjsutment. Or have the motors mounted horizontally within the gantry tube and use timing belts to drive the pinions, with the integral rack/linear rail mounted on the sides.
\r\n
\r\n=======================
\r\n
\r\nThe nice thing with belt drives, aside from ease of use, is you\'re not limited to the stock reduction ratios that gearboxes offer, so it\'s easier to fine-tune the ratio that gives you the best performance just by changing pulleys and belts.\r\n \r\n

[Sapele]

\r\n \r\n I have emailed Linearmotionbearings2008 asking if he has any options with a coarser pitch.
\r\n
\r\nWizard- The "floating" beam would be supported with legs welded at the end and a middle leg that is bolted on the inside. Here is a top view and a side view. My concern is that the rack is 6" away from the linear bearings. I have read that closer is better. No?
\r\n
\r\nAttachment 220262
\r\n
\r\nAttachment 220264
\r\n
\r\nThe router will be placed in the middle of my wood shop. With partial access from all sides. Reducing the protrusion will allow for much better flow in the aisles and a slightly wider machine.
\r\n
\r\nAttachment 220266\r\n \r\n

[wizard]

\r\n \r\n

\r\n

\r\n

\r\n

\r\n \r\n \r\n

I have emailed Linearmotionbearings2008 asking if he has any options with a coarser pitch.
\r\n
\r\nWizard- The "floating" beam would be supported with legs welded at the end and a middle leg that is bolted on the inside. Here is a top view and a side view. My concern is that the rack is 6" away from the linear bearings. I have read that closer is better. No?

\r\n \r\n

\r\n

\r\n

I\'d be more worried about the unsupported length of the beam than anything else. That is easy to deal with though. If I remember correctly this is a long machine which is probably why you are considering rack drive. In any event a long beam will be subject to vibration and deflection if not supported properly.
\r\n

\r\n

\r\n

\r\n

\r\n \r\n
\r\nThe router will be placed in the middle of my wood shop. With partial access from all sides. Reducing the protrusion will allow for much better flow in the aisles and a slightly wider machine.
\r\n \r\n

\r\n

\r\n

If that is the case I would think a leadscrew drive would save you the most space. However creative engineering can do wonders.\r\n \r\n

[pippin88]

\r\n \r\n With your idea of hiding the rack on the bottom of the x axis beams: be aware that you will have to have a rigid and beefy mounting between the gantry and the drive components. If this connection is not rigid you\'ll get problems. It looks like you\'ve modeled a C channel?\r\n \r\n

[Sapele]

\r\n \r\n

\r\n

\r\n

\r\n

\r\n \r\n It looks like you\'ve modeled a C channel?\r\n \r\n

\r\n

\r\n

Sort of. I was just beginning to think of materials, thickness and how to reinforce that part of the x-carriage when I started to doubt the whole design because of the distance between the linear bearings and the rack. If the consensus is that this design is feasible I will continue to develop it by trying to make that part of the carriage as stiff as possible.\r\n \r\n

[mbronkalla]

\r\n \r\n David
\r\nThank you. You insights are most helpful..
\r\nThe distortion covers one of my fears. I did not want to spend the time welding up the tube and ending up with something resembling an overgrown cheeto.
\r\nEpoxy I know is too soft. Once you add fillers for rigidity, it will not flow for self leveling. I went through gallons of it building my boat.
\r\nMilling the steel rail flat is a problem. The length is probably too much for a bridgeport with my skills..
\r\nHowever what about a face layer applied to the tube (e.g. MDF) and then machined? Epoxy underneath would provide a stable substrate match the tube undulations. The MDF or other outer layer would be easily machined on a large jointer . I have been able to do 8\' lengths for table top glue ups with wide planks on my ancient 16" joiner (<.005 gap between planks over before clamping or is the mdf just too soft as well and will deform like epoxy this would provide a reasonable alternative to metal machining most cities have large cabinet shops that could service for very>\r\n

[wizard]

\r\n \r\n

\r\n

\r\n

\r\n

\r\n \r\n \r\n

David
\r\nThank you. You insights are most helpful..
\r\nThe distortion covers one of my fears.

\r\n \r\n

\r\n

\r\n

Nothing to fear here, if you weld on the beam there will be distortion how much depends upon many factors. Ideally if you weld on the beam you would also stress relieve.
\r\n

\r\n

\r\n

\r\n

\r\n \r\n I did not want to spend the time welding up the tube and ending up with something resembling an overgrown cheeto.
\r\nEpoxy I know is too soft. \r\n \r\n

\r\n

\r\n

Possibly but it depends upon many factors.
\r\n

\r\n

\r\n

\r\n

\r\n \r\n Once you add fillers for rigidity, it will not flow for self leveling. I went through gallons of it building my boat.\r\n \r\n

\r\n

\r\n

One common use for self leveling epoxies is to make surface plates. If you make such a plate you can use it to replicate a flat surface from. The approach would be like what David M suggested to use with a Bridgeport table. {as an aside don\'t assume that old Bridgeport tables are flat themselves}. In other words create your surface plate, treat it with release compound and then sandwich epoxy between it and the beam. Of course the epoxy that you adhere to the bean has to be a harder more suitable resin than the self leveling stuff. You also need a resin that has an open time that is long enough to give you a chance in hell of pulling off this. Epoxy resins have a long history of being used in the machine tool industry, you may want to try the Moglice web site for more information about stuff marketed specifically for such uses.
\r\n

\r\n

\r\n

\r\n

\r\n \r\n Milling the steel rail flat is a problem. The length is probably too much for a bridgeport with my skills.. \r\n \r\n

\r\n

\r\n

That is what job shops are for! Seriously, if you weld up a beam with pads for the rails and have it milled flat afterwards you probably won\'t be out that much more money than screwing around for an epoxy solution. Obviously this depends upon the shops the area, how busy they are and how far behind they are on boat payments😂😂😂.
\r\n

\r\n

\r\n

\r\n

\r\n \r\n However what about a face layer applied to the tube (e.g. MDF) and then machined?\r\n \r\n

\r\n

\r\n

I\'m highly biased against MDF so take this with a grain of salt but I\'d never recommend using MDF on a machine like this!!! Not only is it soft, it breaks down over time and is adversely affected by moisture. I\'ve never have had good experiences with the stuff, if you go this route I\'d look into other composite sheet goods that would be more reliable.
\r\n

\r\n

\r\n

\r\n

\r\n \r\n Epoxy underneath would provide a stable substrate match the tube undulations. The MDF or other outer layer would be easily machined on a large jointer .\r\n \r\n

\r\n

\r\n

You bring up an interesting question in my mind here, would a jointer give you a flatter surface than the beam as is from the factory? I really don\'t know the answer to that question.
\r\n

\r\n

\r\n

\r\n

\r\n \r\n I have been able to do 8\' lengths for table top glue ups with wide planks on my ancient 16" joiner (<.005 gap between planks over before clamping or is the mdf just too soft as well and will deform like epoxy>\r\n

\r\n

I would have to say that the MDF is far worst than epoxy. At least with epoxy there are options for much harder resins.
\r\n

\r\n

\r\n

\r\n

\r\n \r\n This would provide a reasonable alternative to metal machining and most cities have large cabinet shops that could provide this service for very reasonable $$$.\r\n \r\n

\r\n

\r\n

Possibly if you can get the shops owner to expose is precious jointer to such a beam. I still wouldn\'t use MDF but that doesn\'t mean phenolics or other materials wouldn\'t work. I can just see the look on a cabinet makers face when you tell him you want to run a steel backed phenolic plate over his fine jointer. Beyond all of that there is the issue of getting the face square to whatever mounting pads you expect to use with the beam.
\r\n
\r\nAll in all it would probably be easier to replicate off a surface plate. If you take the beam to a good machine shop they should be able to keep the face for the rails square to the mounting pads you will have on the beam. This would make assembly far easier.\r\n \r\n

\r\n\r\n \r\n\r\n\r\n

[Sapele]

\r\n \r\n Linearmotionbearings2008 does not have an option coarser than 10mm pitch. I did learn that he can provide 32mm diameter screws. I did not ask the price. The nook calculator and adjusting the length closer to my need produces 1 X 1.000 Right Hand Recirculating Lead Screws & Nuts for Power Transmission - Roton Products, Inc.
\n
\n
\nThanks for the link to Atlanta Drives. That integrated rail/rack is very elegant!\r\n \r\n

[wizard]

\r\n \r\n

\r\n

\r\n

\r\n

\r\n \r\n \r\n

Linearmotionbearings2008 does not have an option coarser than 10mm pitch. I did learn that he can provide 32mm diameter screws. I did not ask the price. The nook calculator and adjusting the length closer to my need produces \r\n \r\n

\r\n

\r\n

That is true but varies with the wood and other parameters. Frankly you can see this in most wood machining processes that are done manually. For example running Cherry through a table saw or other machine to slow will lead to burning. You can even see what happens with a handheld router if you go to slow in the material you are working on.
\r\n

\r\n

\r\n

\r\n

\r\n \r\n This does not seem fast compared to other peoples machines. \r\n \r\n

\r\n

\r\n

What you need to determine is the common feed rates for the materials you expect to work on. Note I said feed rates not rapid movement rates. Just be careful that the suggestions aren\'t for high performance industrial machines.
\r\n

\r\n

\r\n

\r\n

\r\n \r\n Are there any other suppliers out there that can provide a complete package with blocks and end machining like Linearmotionbearings2008? The only other option I have found is Roton 1 X 1.000 Right Hand Recirculating Lead Screws & Nuts for Power Transmission - Roton Products, Inc.\r\n \r\n

\r\n

\r\n

I was under the impression that most ball screw suppliers can build to print - for a price.
\r\n
\r\nBy the way, once you start talking about large diameter screws you really should be looking at the inertia to see if that will be a problem. Many of the solutions you see in the ezcam/forums here are in a sense canned, that is the mechanical performance is well understood, once you start looking at solutions outside the norm you really need to consider doing the engineering to see if you have a workable solution.
\r\n

\r\n

\r\n

\r\n

\r\n \r\n
\r\nThanks for the link to Atlanta Drives. That integrated rail/rack is very elegant!\r\n \r\n

\r\n

\r\n

Interesting, I have to wonder about the cost though. You will also be looking at a servo for this axis but that might be the case anyway.\r\n \r\n

\r\n\r\n \r\n\r\n\r\n

[mbronkalla]

\r\n \r\n So far so good. I went with 6x6x.25 steel tube 66" long. 3 diagonal rods were welded in. Holes cut with plasma cutter after cutting to length
\nIt has mounting rails 1x.25 stee. I ran these through my performax trum sander with several light passes of 60 grit paper to provide a nice finish and tooth for the epoxy. They were then glued on the straightest face with thickened epoxy. I used west system 105 with colloidal silica filler for a peanut butter consistency. The silica will probably make drilling and tapping no fun but is much less of an issue than the high density silica based on past experience
\nMy jointer provided the reference surface. I used a dial indicator to get the tables coplanar (same method I use for setting the knives and outfeed height). Wax paper and packing tape kept everything from sticking togetherAttachment 264852Attachment 264854\r\n \r\n

[louieatienza]

\r\n \r\n

\r\n

\r\n

\r\n

\r\n \r\n \r\n

Linearmotionbearings2008 does not have an option coarser than 10mm pitch. I did learn that he can provide 32mm diameter screws. I did not ask the price. The nook calculator and adjusting the length closer to my need produces 1 X 1.000 Right Hand Recirculating Lead Screws & Nuts for Power Transmission - Roton Products, Inc.
\r\n
\r\n
\r\nThanks for the link to Atlanta Drives. That integrated rail/rack is very elegant!

\r\n \r\n

\r\n

\r\n

The screws are reasonable. The prices for the ballnuts are breathtaking!
\r\n
\r\nI\'m still not convinced 25mm is large enough for a ballscrew spanning over 8". Unless, however, you conceive a spinning nut design, which would eliminate problems due to whipping. I think Techno-Isel uses spinning nuts for their long axis. Phife has a build thread where he uses a fixed-end bearing block for the ballnut. I believe the Grumblau plans utilize a spinning nut. A surplus hollow-shaft direct drive motor is another option, albeit expensive.\r\n \r\n

[mbronkalla]

\r\n \r\n Checking for flatness is hard. Sliding the baem on the joiner or the bridgeport shows very nice flatness but neither are not a perfect reference surface. A precision straightedge / camelback is out of my price range. Fortunately the jointer tops are not very worn. You can still see the original planer marks from the factory (~125 years ago) over much of the tops. I see what you are saying about the end flex. I put the tube on the mill and went to trim the ends. A fair amount flex, even with light cuts and a roughing cutter. I will need to add 2 more diagonals near the ends as you suggest. I want to leave the ends open for bolting it down to the x axis carriages. I only destroyed one drill bit and did not break a tap in over 50 holes for the rails. So the idea of using the colloidal silica (cabosil) seemed to work well.\r\n \r\n

[tahustvedt]

\r\n \r\n If you search for "2525 ball screw" on ebay you find sellers who sell 25mm dia, 25 mm pitch screws. They can probably make them to your specs. Example: 2525 L1100MM Ball Screw Dia 25mm Lead 25mm High Lead ballscrew with Ballnut | eBay\r\n \r\n

[pippin88]

\r\n \r\n

\r\n

\r\n

\r\n

\r\n \r\n \r\n

If you search for "2525 ball screw" on ebay you find sellers who sell 25mm dia, 25 mm pitch screws. They can probably make them to your specs. Example: 2525 L1100MM Ball Screw Dia 25mm Lead 25mm High Lead ballscrew with Ballnut | eBay

\r\n \r\n

\r\n

\r\n

Nice work.
\nLooks like two sellers carry a range of high lead ball screws:
\n2525
\n2020
\n1616
\nAre all listed\r\n \r\n

[Sapele]

\r\n \r\n I am looking at them now. Thanks so much for this!\r\n \r\n

[stern-69]

\r\n \r\n Thanks to all for posting in this thread, as I know it has helped me a lot in my design phase, and Im sure many other CNC newbe\'s have found it invaluable.\r\n \r\n

[Sapele]

\r\n \r\n Wizard, you really got me thinking with this comment.
\r\n
\r\n

\r\n

\r\n

\r\n

\r\n \r\n By the way, once you start talking about large diameter screws you really should be looking at the inertia to see if that will be a problem. Many of the solutions you see in the ezcam/forums here are in a sense canned, that is the mechanical performance is well understood, once you start looking at solutions outside the norm you really need to consider doing the engineering to see if you have a workable solution.\r\n \r\n

\r\n

\r\n

Louieatienza, great point. I think this concern goes to the point that Wizard was making.
\r\n
\r\n

\r\n

\r\n

\r\n

\r\n \r\n I\'m still not convinced 25mm is large enough for a ballscrew spanning over 8\'.\r\n \r\n

\r\n

\r\n

I wish that I had the ability to figure out the answers to these questions. My strength is as an artist and woodworker. That said, I am going to continue to work on the rotating screw design because it makes a significant difference with my need for a narrower machine. A preliminary drawing has revealed that the actual distance between the bearings is 2650mm. 25mm or 50mm pitch will allow the screw to rotate quite slowly but I am concerned about such a long unsupported screw. Once I am a little farther along I will reach out the ebay sellers and see if they can provide larger diameter screws. I am very aware that I am beyond my limits, but I can\'t let the spinning screw solution go just yet.
\r\n
\r\nI have realized that my other design needs to focus on the "Canned" solutions. I take that to mean 20° pressure rack with CNC Router parts PRO Nema 34 reducers with the spring tensioner. The machine will have to become narrower but there should be fewer unknowns as this is a well used solution.
\r\n
\r\nI feel that I need to take this part of the design slowly so that I make the right decision for my particular situation.
\r\n
\r\nThanks for all of the help\r\n \r\n

[LeeWay]

\r\n \r\n Before I designed anything for a large long ball screw, I would first buy a small one to get the design concept and functionality down. No sense buying a big truck to see if you can rotate the transmission to turn the wheels when a junk yard Volkswagon would do. \r\n \r\n

[dmalicky]

\r\n \r\n Nook\'s metric critical speed calculator says a 21.7mm root dia (25mm ballscrew), 2650mm length, end fixity B, gives critical rpm of 374rpm. If the lead is 25mm, that\'s about 370 ipm -' + '- not bad. And that calculator assumes no stability contribution of the ballnut, which is normally not very near one of the ends, so most of the time the unsupported screw length is much less than 2650mm.
\n
\nThe other issue is droop due to gravity, which might exacerbate whipping. The max deflection (for simply supported at each end) is (5*w*L^4) / (384*E*I).
\n- w is the weight per unit length. The screw\'s cross-sectional area is probably about (Pi/4)*(0.92")^2 = 0.66 sqin. Density of steel is about 0.29 lb/in3. So w = 0.66 sqin * (1 in unit length) * 0.29 lb/in3 = 0.19 lb per unit inch length.
\n- L is (2650mm - 70mm ball nut length?) = 2580/25.4 = 102" worst case.
\n- E is 29E6 psi
\n- I is (Pi/4)*(root radius)^4 = 0.026 in4. [EDIT, error the first time, propagating below but now corrected]
\nSo max deflection is (5*0.19*102^4) / (384*29E6*0.026) = 0.37". That\'s a fair amount, might be a problem. I\'d use a bigger screw.
\n
\nAlternately, IIRC there was a nice large format machine (maybe 6\'x12\' or so?) from the UK on cnczone last year (?) that had the racks near the rails at or just above the table, and the motors mounted vertical above the gantry feet, shaft facing down, behind the gantry tube. I think he used compact gearboxes. I don\'t think anything stuck out to the side. But this is a vague memory. I do remember it was a nice machine in a old large room-' + '-an interesting contrast. Anyone know the link?\r\n \r\n

[MI370]

\r\n \r\n

\r\n

\r\n

\r\n

\r\n \r\n \r\n

The max deflection (for simply supported at each end) is (5*w*L^4) / (384*E*I).
\n- w w =

\r\n \r\n

\r\n

\r\n

A Ball screw is not a simply supported beam at both ends so I don\'t think it\'s the right equation.
\n> I think it\'s [W*L^4/(384 EI)]: 5 times less than the above prediction.
\n
\nA ball screw is more clamped at both ends or maybe in some cases clamped at one end (fixed bearing) and a hybrid at the other end (floating bearing). Nonetheless there are implied moments due to the bearing blocks which will reduce the deflection vs a simply supported case.\r\n \r\n

[Sapele]

\r\n \r\n Thanks David,
\n
\nI was concerned about the screw sagging! Thanks for the heads up on the 6\' x 12\' build. I am searching for it\r\n \r\n

[mbronkalla]

\r\n \r\n Dave, this is great. Real analysis and moving beyond opinion. Your previous posts helped me greatly and surprised me in the gantry beam analysis , which was borne out in my machinig. Now I really have to reconsider just how stiff my Y& Z carriages are and add likely some bracing.\r\n \r\n

[tahustvedt]

\r\n \r\n My 2500mm long 25x10mm rolled ball screw can spin up to 550 rpm before it starts whipping, when the gantry is at the ends. I kind of wish I had designed a rotating nut when I had the chance, to reduce the inertia and increase the speed.\r\n \r\n

[LeeWay]

\r\n \r\n You guys are aware that they make precision rack and pinion, right? How much precision do you really need over 12\' and doesn\'t speed and budget matter?\r\n \r\n

[louieatienza]

\r\n \r\n I think the big problem with using a ballscrew on the long axis would be racking or skewing of the axis. A lot would depend on how far apart you spread the bearing blocks, or even which bearing blocks/rails you choose to use. I don\'t know if FK/FF supports would work easily because they would have to rely on your two end plates being perfectly coplanar to each other, othewise when you tighten the bolts you\'ll pull the block out of alignment; and getting them aligned over about 10" distance is no easy task. A stiff transverse plate or beam connecting both gantry uprights could help keep the bearing blocks in the same relationship to each other which could minimize skewing. the other way to prevent skewing would be to use two screws, run by two motors, one slaved to the other. Or you can run two screws with one motor via timing belt. This of course increases the cost and complicates the build slightly. I feel if you want to do any serious work with your machine (and hardwoods require as ridgid a machine as possible) you might want to seriously consider driving both sides of the gantry, rather than driving the gantry from the center.
\r\n
\r\nThe reality is, it all depends on the timeframe you have for your machine. Sometimes, a "canned" or other proven solution is better than spending time on "discovery." You can use Nook\'s or Thomson\'s calculators all you want, but it\'s only a guideline as you may not be using their screws. The China ballscrews are generally fine, but I\'ve seen mixed reviews on their support units. I bought a set of three that I wouldn\'t use on a machine, and don\'t want to spend the time to fix.\r\n \r\n

[dmalicky]

\r\n \r\n Found it! http://www.cnczone.com/ezcam/forums/diy_cn...g_machine.html
\nSuitable gearboxes can be found on ebay for reasonable prices, but it takes a lot of searching and research on part numbers and brands. Louie gave some great advice on these back in post #77. We looked at them for our 4x8 machine but we already had Nema34 steppers and at the time we couldn\'t find 2 input-compatible boxes with the right ratio, size, backlash, price, etc. Most of the reasonably priced ones are for Nema23 or smaller. I would suggest around a 5:1 ratio. 4:1 is about the steepest I\'d want for good acceleration and it would be plenty fast.
\n
\nOr a vertical motor can certain be done with a belt drive or a DIY gearbox. We went DIY gears for ours, pretty economical. Pics and part numbers available if of interest.
\n
\ntahustvedt, thanks, that\'s a good point of reference. The calculators have a small safety factor built in, and often the end fixity \'B\' is better in reality, especially when the ballnut is near the \'free\' bearing mount end, as it effectively makes that end \'fixed\' (double-bearing), turning the system into end fixity \'C\'.\r\n \r\n

[tahustvedt]

\r\n \r\n I would definitely go for rack and pinion if I was making a 12\' long machine.\r\n \r\n

[Sapele]

\r\n \r\n Always interested!
\n
\n

\r\n

\r\n

\r\n

\r\n \r\n Or a vertical motor can certain be done with a belt drive or a DIY gearbox. We went DIY gears for ours, pretty economical. Pics and part numbers available if of interest.\r\n \r\n

\r\n

\r\n

Thanks for the link to the post. I am looking forward to reading it tonight.
\n
\nLeeWay, I started out with rack and pinion and have come full circle. There seem to be too many concerns with a long ballscrew for my particular application. A spinning nut design would solve those issues but I might also loose the advantage of a narrower stance. And then there is the reality of the budget!\r\n \r\n

[dmalicky]

\r\n \r\n Yes, rack is a good solution for a long axis. If only cutting wood, I\'d probably use a CNCRP\'s Pro drive, since it\'s easy and reasonable $: PRO Rack and Pinion Parts | CNCRouterParts
\r\n
\r\nIf cutting aluminum, a gear drive would be far stiffer, and offers more reduction for more acceleration and resolution. Ours uses 84T:20T for the reduction (4.2:1) and another 20T for the pinion. All gears are 20 pitch, 20 deg pressure angle. The big gear and pinion are on a 5/8" hardened dowel shaft, located by precision needle bearings. Bronze thrust bearings locate the big gear between the alum plates. The motor mount holes are a bit oversize to adjust gear-gear backlash. As you can see, some nut and bolt holes were counterbored to clear the big gear and rack. The build is a little rough, but they work great. Parts cost per drive is about $130, plus aluminum and the stepper.
\r\n
\r\nAttachment 221452 Attachment 221454 Attachment 221458 Attachment 221456
\r\n
\r\nParts
\r\n
\r\nMotor gear: Boston 20T, 1/2" bore for our NEMA34s.
\r\nBoston Gear YA20 Spur Gear, Steel, Inch, 20 Pitch, 0.500" Bore, 0.500" Face Width, 20 Teeth
\r\nBoston Gear YA20-1/2 Spur Gear, Steel, Inch, 20 Pitch, 0.500" Bore, 0.500" Face Width, 20 Teeth
\r\nThe less $ one is plain bore, so needs 2 setscrews tapped.
\r\n
\r\nBig gear: Boston 84T, 5/8" bore.
\r\n Boston Gear YA84 Spur Gear, Cast Iron, Inch, 20 Pitch, 0.625" Bore, 4.300" OD, 0.500" Face Width, 84 Teeth
\r\nCast iron helps reduce noise.
\r\n
\r\nAxle bearings: precision needle roller, 5/8" bore, 1 1/8" OD, 3/4" long.
\r\nNew TORRINGTON HJ 101812 Bearing | eBay
\r\nNeedles are compact and extremely stiff. If placed above the table, probably use a sealed bearing: McM 8258K21.
\r\n
\r\nAxle thrust bearings and washers: McM 5906K515 & 97022A261
\r\n
\r\nAxle: 5/8" hardened dowel pin, McM 98381A812
\r\nNeedle bearings required a hardened axle. We ground flats for the 84T set screws, and chamfered the case away for welding to the pinion.
\r\n
\r\nRack pinion: Martin 20T, 1/2" Bore.
\r\nMartin TS2020 Spur Gear, 20° Pressure Angle, High Carbon Steel, Inch, 20 Pitch, 1/2" Bore, 0.500" Face Width, 20 Teeth
\r\nWe reamed it on a lathe to 0.625" to fit the axle; the gear must be plain bore to do this concentrically. Martin gears are a little harder than Boston, so a good choice against the rack.
\r\n
\r\nRack: Moore Gear R2020x6
\r\n
\r\nDeciding between 1/2" and 5/8" axle size is a little complex. At 1/2", FEA models showed that bending of the pinion shaft was a big source of flex (we were aiming for a ~50k lb/in machine). So we upsized to 5/8", which also fits the 84T. But a 5/8" axle requires reaming the pinion, which didn\'t leave much steel for the setscrews, so we light press-fit the axle in the pinion and welded them at the end (welding elsewhere would cause runout).
\r\n
\r\nWith a less stiff machine, a 1/2" axle is likely fine. Then the only question is how to get a big gear with a 1/2" bore.
\r\na) Adapt the Boston YA84 to the 1/2" axle. McM 2867T57 may work if concentric enough.
\r\nb) Use Martin\'s 72T steel gear with 1/2" bore: Martin TS2072 Spur Gear, 20° Pressure Angle, High Carbon Steel, Inch, 20 Pitch, 1/2" Bore, 3.7" OD, 0.500" Face Width, 72 Teeth
\r\nSteel is noisier than cast iron. Ratio with a 20T motor gear is 3.6:1. Easy.
\r\nc) Switch to 14.5 deg pressure angle gears, which have both 80-90T and 20T in 1/2" bore.
\r\nMartin Spur Gear, 14.5° Pressure Angle, High Carbon Steel, Inch, 20 Pitch
\r\nThese have more backlash than 20 deg gears, but probably not significant compared to the rack backlash. Steel. Better ratio than the 72T. Easy.
\r\nd) Switch to 24 pitch brass gears: Brass Spur Gear, Brass, Inch, 24 Pitch
\r\nI\'d need to analyze to see if strong enough, but probably fine. Probably a better solution if using a Nema23 motor with 1/4" shaft.\r\n \r\n

[Sapele]

\r\n \r\n Thanks for posting this Dmalicky. The detailed explanation is so useful! If I remember correctly you used a balls crew for the Y axis. If this is true I would be curious to know why? Price is my guess. I would also be curious to know if you used reduction with that axis? Did you try to match the inches per minute of the x-axis.
\n
\nthanks again for the great post!\r\n \r\n

[LeeWay]

\r\n \r\n I find nothing wrong with the affordability or usefulness of ball screws on a shorter axis. 4\' or less. The differences become dramatic when you start getting longer.
\nMy router uses a rolled ball screw that is 48" long. I was able to use it initially at lower speeds. Say 150 IPM. Cutting speed. When I installed a double nut, I was able to increase it by double the speed. 300 IPM. The entire machine became more stable with a double nut on X and Y. That tells me more math needs doing in order to get an accurate idea of what is actually needed for a long ball screw. You can do what others have done, but question them on satisfaction of rapid speed.\r\n \r\n

[tahustvedt]

\r\n \r\n Maybe my machine would have been better off with R&P, but it definitely wouldn\'t have been more affordable, or easier to design/build, for me as I can get by with a single ball screw. A single 2510 ball screw and motor costs $350 including shipping. I am not that satisfied with my rapids now at 520 rpm (reduced slightly for headroom). Maybe I\'ll figure something out eventually, but I\'ll manage somehow. \r\n \r\n

[ger21]

\r\n \r\n I think that once you get longer than about 5-6 ft, you\'re better off going with servos when using ballscrews. Servos have a couple advantages when using high lead ballscrews.
\nOne, their high rpm allows you to use a belt reduction, which minimizes the affect that the much higher inertia of the screw will have on acceleration.
\nAnd you also get much higher resolutions, which you lose when using steppers with high leads screws.
\nI would think that a 2525 ballscrew would be fine up to about 6ft and 1000rpm (1000ipm), but I\'d probably want to use either a spinning nut, or a larger screw for an 8ft machine. Remember, for an 8ft machine, you\'d typically need at least as 9\'-10\' screw.
\nUnfortunately, 2525 seems to be the largest commonly available chinese screws. Anything larger, and the price tends to go up by about 5x.\r\n \r\n

[dmalicky]

\r\n \r\n Glad it\'s useful! Yes, a ballscrew on the Y has a lot of advantages over a rack: linearity, repeatability, packaging, easy of installation (unless the rack is spring loaded), longevity (pinions wear, esp with spring loading), backlash (unless rack is very accurately placed or spring loaded).
\n
\nYes, we did direct drive with a 2020. Ours is from the ebay site of Bob LaLonde\'s (he\'s on cnczone)-' + '-he had a bunch of Taiwanese Gten ~zero-backlash 2020s for a reasonable price; he may still have them. Depending on accuracy needs, Gerry\'s point on stepper resolution is important for high lead screws. 20mm lead results in a ~0.002" half-step size. 50mm lead would be 0.005". (Higher microstepping is usually done, but the steps aren\'t as evenly spaced). For ease of use, we tried to ~match the X and Y step sizes. With 4.2:1 and a 20T pinion, a half-step moves the X ~0.0019".
\n
\nMI370, yes, that\'s correct, take away the \'5\' for perfectly fixed-fixed end conditions. My calc was a worst-case, since I wasn\'t sure how rigid the BK bearings actually are (and I was sleepy). From what I have seen, the usual config is BK-BF (fixed-simple), but BK-BK is also done. For fixed-simple, the 5/384 or 1/384 changes to 2/384. But even a preloaded double back-to-back AC BK is not \'rigid\'. Nook shows End Fixity \'D\' is for "both ends mounted with a double preloaded angular contact bearing spaced apart by least 1.5 times the diameter of the mounting journal", or the catalog shows a quad AC on each end. So \'D\' is probably as close to \'fixed\' as is possible, and would result in acceptable deflection.
\n
\nI wonder how rigid the low $ BK and BF bearings are. The few BKs I\'ve seen are just 2 600Xs next to each other: a big step down from a Nook BK. A BF with a single 600X normal clearance bearing would be simply supported initially, but would offer an increasing moment if angular deflection were high enough (seems unlikely a ballscrew would get there).
\n
\nRereading the pros of ballscrews, I took another look at if or how to make a 2525 work at 8\'. Nook\'s \'length\' is between the supports, so for 8\' travel and a nut, about 100". (The ballnut-to-farthest-support distance is probably more accurate, and you might subtract off whatever distance is needed to accelerate to speed.) If you used preloaded back-back ACs on each end, you get end fixity C. Plugging in 21.7mm, 2540mm, and \'C\' gives 600 rpm, 600 ipm. Or if you separate the bearings by at least 30mm, that\'s end fixity D, giving 900rpm. That\'s about as fast as the Nema34 will go anyway.
\n
\nAnother issue with a long screw is the column strength; Nook has that calculator, too, and plugging in 21.7, 2440, and \'C\'.... Nook says "You have exceded the Slender Ratio for this Diameter." Their max safe length for \'C\' is about 1550mm (61"). For \'D\' it\'s about 2150mm (85"). Then I checked our 2020 screw-' + '-it fails the calculator but runs fine. So their slenderness check is likely pretty conservative; it\'s hard to say how the risk ramps up.
\n
\nIf the column buckling issue can be resolved, and some custom bearing blocks were made to reproduce fixity \'D\', a 2525 for 8\' may be reasonable.\r\n \r\n

[MeAgain]

\r\n \r\n

\r\n

\r\n

\r\n

\r\n \r\n \r\n

Rereading the pros of ballscrews, I took another look at if or how to make a 2525 work at 8\'. Nook\'s \'length\' is between the supports, so for 8\' travel and a nut, about 100". (The ballnut-to-farthest-support distance is probably more accurate, and you might subtract off whatever distance is needed to accelerate to speed.) If you used preloaded back-back ACs on each end, you get end fixity C. Plugging in 21.7mm, 2540mm, and \'C\' gives 600 rpm, 600 ipm. Or if you separate the bearings by at least 30mm, that\'s end fixity D, giving 900rpm. That\'s about as fast as the Nema34 will go anyway.
\n
\nAnother issue with a long screw is the column strength; Nook has that calculator, too, and plugging in 21.7, 2440, and \'C\'.... Nook says "You have exceded the Slender Ratio for this Diameter." Their max safe length for \'C\' is about 1550mm (61"). For \'D\' it\'s about 2150mm (85"). Then I checked our 2020 screw-' + '-it fails the calculator but runs fine. So their slenderness check is likely pretty conservative; it\'s hard to say how the risk ramps up.
\n
\nIf the column buckling issue can be resolved, and some custom bearing blocks were made to reproduce fixity \'D\', a 2525 for 8\' may be reasonable.

\r\n \r\n

\r\n

\r\n

\nI have seen this or similar methods used with great success. This addresses both whip and slender ratio concerns.
\n
\nCNC whipping ball screw - lead screw remedy - YouTube
\n
\nI agree with “The ballnut-to-farthest-support distance is probably more accurate”
\n
\nWorst case with this method, ballnut at either end limit while simply supported at screw midpoint.
\n
\nDisadvantages – 1) Introduces drag due to friction. 2) Introduces heat to screw and as a result thermal expansion. 3) Wear due to friction.
\n
\n1) As long as motors have been properly sized, Load to Motor Inertia Ratio of around 10, then drag would be negligible.
\n2) Should be insignificant, given the intended use. I don’t think he needs or will get aerospace tolerances with an 8’ router.
\n3) Provided the cradle in made of delrin or some other self-lubricating plastic, shouldn’t be an issue. I would be more concerned with repeatedly fastening material for smaller sized jobs to the same location on a large table. This will result in the same small length of the screw being used repeatedly. Better to move the jobs around so that the screw does wear more evenly.
\n
\nGood Luck with your build\r\n \r\n

[dmalicky]

\r\n \r\n More info on column buckling. Nook\'s slenderness limit appears to 100:1, as it\'s consistent with this article: Fixed-end, supported-end, free-end ball-screw support conditions | Machine Design "A length-to-diameter ratio (L/D) exceeding 100:1 requires special design consideration; consult the manufacturer regardless of anticipated loads or fixity."
\r\n
\r\nBut, Thomson\'s own ballscrew engineering pdf shows no 100:1 limit, nor does THK mention it.
\r\npage 86: http://www.thomsonbsa.com/pdf/bsa_en...og_section.pdf
\r\nhttps://tech.thk.com/en/products/pdf/en_a15_030.pdf
\r\nThomson\'s chart for a 100" 2525 screw (fixed-fixed) shows a limit of about 1000 lb; they don\'t mention what safety factor is accounted for. THK\'s equation calcs to 7kN (1500 lb), including a safety factor of 2. Either are plenty for our needs.
\r\n
\r\nAs for the 100:1 limit, I\'m not sure what to make of it. A 2525 x 100" would be about 117, so only a little more, and "100" was almost certainly an arbitrary choice. My guess is they had some bad experiences and want engineering to look at the longer ones.
\r\n
\r\nOn end conditions, Thomson has a nice figure indicating the rigidity of BF, BK, and the spaced-BK types:
\r\nAttachment 221722
\r\n
\r\nI can think of three potential disads of fixed-fixed:
\r\n- Mounts are harder to make or more $.
\r\n- A tensioned screw is usually needed, so it doesn\'t buckle when it heats up (or if mounted to an aluminum frame, differential thermal expansion could make it buckle). But, no tension is needed if the far mount used a pair of needle bearings to do the angular fixing (or make a sleeve mount for ball bearings that slides axially). Tension only has a minor benefit to critical speed, so I\'d rather use needles and avoid the whole issue. Or, if the 117:1 ratio is a concern, a tensioned screw should alleviate that.
\r\n- The alignment of that 2nd fixed mount is much more important, as it needs to be very co-linear with the screw. An easy way to do that is to make it a face-mount (like FK) on a plate; then drive the gantry car over to it and tighten the face bolts.
\r\n
\r\nSo, with 2 mounts using spaced bearings, an 8\' 2525 is looking pretty reasonable to me.\r\n \r\n

[dmalicky]

\r\n \r\n Yes, definitely a ballscrew on each side-' + '-sorry that was not clear before. That\'s a good point on the FK blocks needing parallel surfaces at each ends of the X. One idea is to use a ~4.5\' long square or rect tube mounted transverse across the lower frame: one at X=0\', and another at X=8\'. If the tubes are straight, faces are flat, and the corner diagonals are measured equal, the mounting faces should be parallel. Then there is the JBW conforming shim method.
\n
\nAnd a good point on canned vs new/developed. I lean to the latter for the fun, performance, and often less $, but it does take more time, often a lot more.\r\n \r\n

[ger21]

\r\n \r\n Our router with 14\' screw and spinning nut uses something similar, but more to prevent sag than whipping.\r\n \r\n

[Sapele]

Linearmotionbearings2008 does not have an option coarser than 10mm pitch. I did learn that he can provide 32mm diameter screws. I did not ask the price. The nook calculator and adjusting the length closer to my need produces 1 X 1.000 Right Hand Recirculating Lead Screws & Nuts for Power Transmission - Roton Products, Inc.


Thanks for the link to Atlanta Drives. That integrated rail/rack is very elegant!