Thoughts on a 5 axis router/welder?


Page 1 of 2 12 LastLast
Results 1 to 20 of 35

Thread: Thoughts on a 5 axis router/welder?

  1. #1
    Member
    Join Date
    Apr 2017
    Location
    United States
    Posts
    24
    Downloads
    0
    Uploads
    0

    Default Thoughts on a 5 axis router/welder?

    Hello all,

    I'm not sure if this is exactly the right place to post this but I wanted to get some ideas out there and get some feedback on them.

    I've seen lots of CNC routers designed and built by folks on this forum, and they're all very impressive. Eventually I'd like to build one myself too, but if I do build something myself, I'd want it to be unique. Maybe the part that makes it unique makes it more useful, maybe less, but I think that doing something new or interesting would make building something custom more worthwhile. So, I have a few things that I'd want to do with this project. These are just goals to play around with in CAD, who really knows if they will ever become reality:

    1. It should be 5 axis. There are so many neat things to do with a 5 axis machine, but it is complicated and seems to add another point to stack up tolerances. I hope this goal will become more clear with the next goal.
    2. It should have both a spindle and a MIG gun. Not sure if MIG gun is the right word for it in this context, but some way to do MIG weld deposition. Having an additive side to the machine would be fun to play around with and I can see a few areas where it would be useful. For one, you could build up areas on existing parts rather than making replacements or new versions. You could also (if you were good with software) make the machine into a sort of metal 3D printer. I think this point in particular would be interesting for the sake of accessibility of metal parts. If you can work sequentially like an FDM printer does, maybe a program can do more of the work for you rather than requiring extensive CAM training?
    3. I have a unique drive idea in mind that I haven't seen before. This could be because nobody's done it, or because it doesn't work, I have yet to see. Either way, I think I can incorporate this into the machine. If it doesn't work, anywhere this mechanism would have been used can be replaced with a belt drive. More on this later.
    4. The whole thing should be made cheaply. It's not like this is a new goal, I know that everyone who builds their own machine doesn't go looking to spend top dollar on everything. But there are some ideas I'd like to try. For example, could linear rails be made from PVC pipe filled with concrete? Or what about 3D printed parts with a gyroid infill that can be filled with concrete as a stiffener/damping element? These obviously aren't even close to the accuracy of hardened and ground rods and fully machined parts, but they may be good enough for some parts of the machine.

    With these goals in mind, I sketched out a rough idea of what the machine might look like as far as the structure is concerned. The drive mechanisms aren't shown, but my idea was to drive the pink main body piece with a sort of CoreXY belt drive setup like you see on some 3D printers. Depending on how the belt is routed, the center Z axis column will be resistant to twisting about the Z axis, which I think is something a 5 axis machine has to deal with that a regular 3 axis machine does not. The pink section slides on the red colored rods, and the Z axis column is composed of the blue rods with orange ends. The light pink L piece shows how the 4th and 5th axes might be setup. The black part represents the combination of spindle on one side and MIG gun on the other. My thought was that the MIG gun would be on the other side of the spindle so you could switch from cutting to welding by rotating the 5th axis 180 degrees.

    This is just a rough concept at the moment, but the people on this forum seem to have a good intuition for designing these sorts of things. I'm curious as to what you guys think. If it's stupid, let me know! I haven't seen this style of setup before outside of a 3D printer, so I'm thinking it's probably stupid...

    Similar Threads:
    Attached Thumbnails Attached Thumbnails Thoughts on a 5 axis router/welder?-sample-png  


  2. #2
    Member
    Join Date
    Dec 2003
    Location
    East Anglia,England
    Posts
    721
    Downloads
    0
    Uploads
    0

    Default Re: Thoughts on a 5 axis router/welder?

    Before you get too far,I suggest you find a way to generate the Gcode to run the machine.Five axis post processors need a good deal of customisation in any event and if your chosen software doesn't have a good base-you don't have a working machine.In other respects it looks like you have designed a portal machine with a few additional rails.



  3. #3
    Member
    Join Date
    Apr 2017
    Location
    United States
    Posts
    24
    Downloads
    0
    Uploads
    0

    Default Re: Thoughts on a 5 axis router/welder?

    Quote Originally Posted by routalot View Post
    Before you get too far,I suggest you find a way to generate the Gcode to run the machine.Five axis post processors need a good deal of customisation in any event and if your chosen software doesn't have a good base-you don't have a working machine.In other respects it looks like you have designed a portal machine with a few additional rails.
    Absolutely. I'm a university student at the moment, so I think part of the use of this machine might be in developing that software side. That's not my focus, but I know of people who would like to work on that sort of thing. Even a commercial program would not likely be able to drive the MIG side of things. That's one reason I think 5 axis would be useful: in a research setting, 5 axis let's you change a lot of variables with respect to how the MIG deposition is done.

    Sent from my FRD-L14 using Tapatalk



  4. #4
    Member peteeng's Avatar
    Join Date
    Jul 2018
    Location
    dum dum
    Posts
    1891
    Downloads
    0
    Uploads
    0

    Default Re: Thoughts on a 5 axis router/welder?

    Hi Malach - One of my jobs some time ago was programming industrial robotic MIG welders, currently I design machinery in the medical and mining industries & build cncs on the side....But lets start at the beginning. 1) Your design is a little over complicated for the purpose. The crossed gantry is redundant and this creates synchronization problems eg if the cross axis does not operate perfectly it will lock up or chatter. It may work in printers as there are no contact loads. plus unsupported round rails are not stiff enough in a router application.

    2) The Z axis will not be stiff enough. Its a bad plan to combine the structure with the motion elements. The structure needs to be substantial and the motion elements are relatively small Plus there is the issue of a cantilvered load. You need to look up stick slip behavior. Round bearings or bushes have stick slip and you need much higher aspect ratios of support so they work then you have drawn. The multi bearing assembly in the middle is going to be huge. Don't fret the napkin sketches rarely work...you need to get into CAD and do the mechanics math 3) MIG welding requires quite a large stiff wire feeder attached to it. Programming these always revolved around how that large stiff hose would need to be managed in space. Your design has limited headroom and the ability to revolve to manage that issue. Industrial anthropomorphic type 6 or 7 axis robots are much better suited to the task then your configuration 4) Cost you use the word "cheap" and this config uses many more parts to achieve an end then it needs to. ie it won't be cheap and it is complex (again read not cheap) 5) 5 axis software is expensive. Although freeCAD and linuxCNC may have CAM that does your job for free. But if you pursue this idea as mentioned by routealot you need to investigate and nail down the software first. The SW is the crux of this proposition the mechanicals are nearly trivial in comparison. I suggest you look at anthropomorphic configurations for your combined concept. It will serve you much better and the market needs this sort of device much more than more gantry machines. 6) The major machine makers all have metal printers combined with mills these days so your on the right track. Keep thinking...cheers Peter



  5. #5
    Member
    Join Date
    Apr 2017
    Location
    United States
    Posts
    24
    Downloads
    0
    Uploads
    0

    Default Re: Thoughts on a 5 axis router/welder?

    Quote Originally Posted by peteeng View Post
    Hi Malach - One of my jobs some time ago was programming industrial robotic MIG welders, currently I design machinery in the medical and mining industries & build cncs on the side....But lets start at the beginning. 1) Your design is a little over complicated for the purpose. The crossed gantry is redundant and this creates synchronization problems eg if the cross axis does not operate perfectly it will lock up or chatter. It may work in printers as there are no contact loads. plus unsupported round rails are not stiff enough in a router application.

    2) The Z axis will not be stiff enough. Its a bad plan to combine the structure with the motion elements. The structure needs to be substantial and the motion elements are relatively small Plus there is the issue of a cantilvered load. You need to look up stick slip behavior. Round bearings or bushes have stick slip and you need much higher aspect ratios of support so they work then you have drawn. The multi bearing assembly in the middle is going to be huge. Don't fret the napkin sketches rarely work...you need to get into CAD and do the mechanics math 3) MIG welding requires quite a large stiff wire feeder attached to it. Programming these always revolved around how that large stiff hose would need to be managed in space. Your design has limited headroom and the ability to revolve to manage that issue. Industrial anthropomorphic type 6 or 7 axis robots are much better suited to the task then your configuration 4) Cost you use the word "cheap" and this config uses many more parts to achieve an end then it needs to. ie it won't be cheap and it is complex (again read not cheap) 5) 5 axis software is expensive. Although freeCAD and linuxCNC may have CAM that does your job for free. But if you pursue this idea as mentioned by routealot you need to investigate and nail down the software first. The SW is the crux of this proposition the mechanicals are nearly trivial in comparison. I suggest you look at anthropomorphic configurations for your combined concept. It will serve you much better and the market needs this sort of device much more than more gantry machines. 6) The major machine makers all have metal printers combined with mills these days so your on the right track. Keep thinking...cheers Peter

    Thanks for the help Peter.

    1) My thought behind the crossed gantry was that it would be a different way to handle the loads of the highly cantilevered Z axis. A regular CNC router has a relatively tall gantry which can take the load that would otherwise twist the gantry. With the crossed gantry, each set of 2 rods can take the twist in either direction, if that makes sense. And as far as synchronization goes, if the belts are run in a modified CoreXY setup, I don't think that there would be any issue. My idea would be to modify a CoreXY style belt drive so that the trucks on either end of the crossed gantries would be fully constrained by the position of the motors. I can draw a picture later if you think it would help, CoreXY is a bit weird.

    2) Is there something inherently wrong with combining the structure with the motion elements? I saw an NYC CNC video where he toured Major Tool in Indianapolis, and at the 20 minute mark they look at a vertical turning machine. I know it's a lot bigger than what I'm describing, but their Z axis (I think that's the right axis, it's the one with the tool head) seems to be a square column acting as both the structure and the motion element. Link is here: I agree with your point about the aspect ratio of support. Could you give a rough number for a common ratio? The diameter of the rods shown here is 2", and I'm basing that on using PVC pipe filled with concrete. The multi bearing assembly will be 3D printed. I've already 3D printed a bearing that works on 1.5" PVC pipe, so I think I can 3D print a lot of parts together as one piece to shrink down assemblies. Are these bearings as good as linear bearings? Obviously not, but I'd like to see what kind of results I get.

    3) Yeah, MIG is an unknown for me at the moment. Given that I want to machine the results of the MIG welding, I was thinking I'd be limited to aluminum wire, which I understand requires a spool gun to MIG normally. So, the spool would have to be mounted to the spindle area I guess. Still not sure how this part would work.

    4) One of the reasons to use CoreXY is that it requires 2 motors to drive 2 axes. That means that this setup uses 5 motors to drive 5 axes. So, a 6 or 7 axis anthropomorphic robot needs 1 to 2 more motors, and by extension drives. I haven't designed anthropomorphic style robots so I don't know, but I'm not sure that one of those would really have a lower cost. Could you break down what would make them cheaper?

    5) I agree about the need for 5 axis software, but I'm not worried about that at the moment. What do you mean by "the market needs [anthropomorphic configurations] much more than gantry machines?

    6) Thanks for the encouragement. I have seen DMG Mori's Lasertec setup with the blown powder deposition. I guess this machine is trying to approach the same end goal.



  6. #6
    Member peteeng's Avatar
    Join Date
    Jul 2018
    Location
    dum dum
    Posts
    1891
    Downloads
    0
    Uploads
    0

    Default Re: Thoughts on a 5 axis router/welder?

    Hi Malach
    1) The crossed drive as drawn has many problems from experience, binding, chatter lack of stiffness. I think if you built it you would be bitterly disappointed. The coreXY is too complex for my liking for a mill. You are actually talking about a mill not a router as you want to weld then machine. You need to look at some mills and see the very significant difference in build and stiffness to what you propose
    2) Yes economically and functionally. If you combine them you put too many $$$ into the motion system or not enough $$$ into the structure. As individuals they can be optimized, as a unit they can't. The Major tool Z has two very large linear rails down each side so they have not combined the two functions.
    3) No you will not mount the spool at the head as it will be too heavy and you can't change the spool easily. It will be mounted remotely maybe on the gantry or best on the floor with a long wire feeder as I mentioned prior. Welders have to move very fast to be economical. Inertial loads will be big so you need to get into dynamics...
    4) Cost is an illusion. Custom machines are designed to do a job. If you generalise a machine you actually get a more expensive machine to do a job. Industry spends 10xBillions of dollars on single machines to get a job done. If you chase cost you will fail, you have to chase performance. The machine will cost something and you have to compare it to others that do the job you want. One of your problems I think is that you are trying to solve a perceived cost problem vs a functional problem. Plus you need to think BIGGER
    5) There are probably millions of gantry machines out there. If you are going to put the next few years into learning about gantries and welders you may as well learn about industrial robots. They have been very big in the past but they are now getting bigger and have more uses. Cobots, machine handling, routing, trimming 3D printing, AI personal robots, they are all going to be robotised vs using a gantry. A gantry is not flexible enough for the applications that are appearing. So I encourage you to learn about robots.... If you go to an employer in the near future and say you know all about gantry machines they will say ho hum. Alternatively if you say you have built and know all about IRBs they will say when can you start...Peter

    PS you will need to work up designs in parametric CAD so download fusion360 or Freecad. happy to help. Keep at it. Peter

    Regarding bearings - home made bearings obviously cannot meet the accuracy of mass produced ones. Your PVC concrete creation has so many negatives I'll just say don't go there. Trying to save $$$ by making bearings is false hope. Been there done that. The only bearing that I would make these days myself is air bearings and this is something you may look into and would be helpful in your current holey grail. Economically and Performance wise choose a well spec'ed industrial bearing that does your job. Put your intellectual effort into other areas. Since this exercise is on paper forget about cost, design something that works and can be well reviewed by experts then your on a good path.

    Last edited by peteeng; 09-12-2020 at 10:22 PM.


  7. #7
    Member
    Join Date
    Apr 2017
    Location
    United States
    Posts
    24
    Downloads
    0
    Uploads
    0

    Default Re: Thoughts on a 5 axis router/welder?

    Peter,

    That image from earlier was just a screenshot of something I made in Inventor, so I'm familiar with parametric CAD.

    Isn't there something to be said for designing a tool to be as low cost as possible? The Ender3 certainly, and I would argue the Prusa i3, have both been very popular 3D printers due to their low cost and good performance. And if I'm designing something for myself, I want there to at least be a chance that I could buy the parts for it. My limit would basically be the same cost as buying an OMIO CNC router, to give you an idea.

    Let's say that this concept is too flimsy for routing aluminum. What about machining foam? That would still be something 5 axis. Even slashing routing entirely, weld deposition would have no cutting forces (I assume), so parts of this could still be useful. I had considered air bearings, but the tolerance of PVC pipe is too low. My main point is that cost is not low on the totem pole, it stands near the top for me.



  8. #8
    Member peteeng's Avatar
    Join Date
    Jul 2018
    Location
    dum dum
    Posts
    1891
    Downloads
    0
    Uploads
    0

    Default Re: Thoughts on a 5 axis router/welder?

    Hi Malach - Yes a commercial machine should be made at lowest cost but not at the expense of performance. If there is a commercial machine mass produced that does your job its unlikely you can make it cheaper. You just buy it and get on with what you want to do. Welding has a very heavy head and it has to move quite fast. It has to be statically and dynamically stiff even if there is no cutting forces. Just use a MIG for an hour and you will see how heavy it is and how you have to keep getting the feeder out of the way etc etc.

    Yes cost is high priority as you are a student and you are trying to do something for nothing, Everyone wants to do something for nothing I've had heaps of bosses that want that. But it costs something to do anything. Mass produced machines are good value and are indicative of their actual cost. A person who buys a Rolls Royce would say a VW is low cost yet they get you to the same place at the same legal speed. Perhaps a VW gets you down a narrow laneway to the café faster. Your expectations of milling/welding and your costs and machine reality are just disconnected as you are inexperienced. You are now adjusting your expectations. You need to develop a specification of what you want to do otherwise you have a fast moving target. Yes cutting foam is like cutting cake very easy, decide on what you want or need to do. Its a paper exercise so just "costs" your time. Time well spent on your education.... Peter

    You have inventor so you have FEA and generative design at your fingertips. All the tools to design a good machine. I'm learning Inventor at the moment. I use Alibre, mechsoft, rhino3D, Strand7 and Simsolid

    Thinking about this the best way to get away from the "low cost" is to define the budget. Low cost is a relative non concrete description. So tell us is this a $500USD, $5000USD or $500,000 whatever budget then the $$$ is defined. Then a bullet point spec of what you want to achieve and we have a great starting point.

    Last edited by peteeng; 09-13-2020 at 12:23 AM.


  9. #9
    Member
    Join Date
    Apr 2017
    Location
    United States
    Posts
    24
    Downloads
    0
    Uploads
    0

    Default Re: Thoughts on a 5 axis router/welder?

    Peter,

    I think the budget for this thing would fall in the $500 to $5000 range. It seems like it would be impossible to acquire anything more than the stepper drivers in the $500 range haha.

    The reason that the initial design was sort of square and relatively small is that I wanted to sort of emulate the Ultimaker 3D printer in a way. The volume the Ultimaker takes up is not much larger than the build volume. I wanted my machine to be a sort of workstation that allows you to go from nothing to a machined part, similar to a 3D printer for plastics. This seems to be a different sort of approach than a robotic arm, where you want the arm to be compact at times yet have it still reach a broad work area.

    I used a Monarch VMC 75b when I worked at a machine shop over the summer. That machine was extremely rigid and you could take big cuts with it in steel. However, if my machine is cutting aluminum at best, with likely a 1/4" cutter most of the time, surely I don't need as much rigidity. Are there some standard cutting forces people use to design CNC machining centers? For example, is it reasonable to expect 20 pounds at the end of the tool? 200? The closest comparison I have a real feel for would be using a hand router, and the force required to move it through material seemed very low. Having a good set of numbers to design for would help as far as the FEA goes.

    Also, a question about the MIG side. You mention that speed is critical for MIG robotic arms, which makes sense to me in an assembly sort of context. The arm needs to reach many locations quickly to assemble a larger part. But in my context, the welding would be done on a small work area, say 12" by 12" square. Would speed still be critical? I'm not sure what sort of "feed rates" are common when the actual welding is being performed, rather than during the "rapid" movements.

    Thanks again for letting me ask you a lot of questions!



  10. #10
    Member
    Join Date
    Dec 2003
    Location
    East Anglia,England
    Posts
    721
    Downloads
    0
    Uploads
    0

    Default Re: Thoughts on a 5 axis router/welder?

    This is an interesting thread.One of the points that peteeng hasn't really emphasised is that a robotic arm can reach into places that a gantry machine can't get to and isn't hindered by parts of a job that would be too large to fit inside the machine's working envelope.It may also be easier to write a program for a repetitive task manually if you have a robotic arm as the background calculations for a five axis gantry machine can be quite complex.Obviously the leverage has to be considered when using more than one arm attached to another and has a rotating joint or two in the system.



  11. #11
    Member peteeng's Avatar
    Join Date
    Jul 2018
    Location
    dum dum
    Posts
    1891
    Downloads
    0
    Uploads
    0

    Default Re: Thoughts on a 5 axis router/welder?

    Hi Malach - So $5000USD is the budget?
    Some numbers for you:
    1) A modern VMC has a tool stiffness of at least 100N/um ie put 100N on the tool and it only moves less then 0.001mm
    2) Modern large VMC's have the capacity to cut with up to 3000kgf at the tool
    3) Cutting aluminium by hand with a router is deceptive. You usually are cutting along an edge not plunging or full cut widths. Your arms are damp and you are usually using a guide. You need about 15kgf of force minimum to cut aluminium. You can do it with less with small depth of cuts, This is how shapeOk and small routers do it but they usually have the wrong speed to feed ratios and wear tools fast. This has to do with the "chip load" look that up you'll need to know about it. High sped machining uses full tool depth cuts. This maximises tool life. If you use the bottom 0.5mm of the tool to cut you wear that zone and throw the tool away. You need correct power and the correct chip load to use modern HSM techniques
    4) For your machine I would aim at 20N/um static stiffness and you'll get about 10N/um in reality. Connections are about 50% efficient unless you model them correctly with preloads and friction
    5) you can look up welding speeds for the type of thing you want to do and you can neglect rapids as you say
    6) I looked up the Monarch and it weighs 18000lbs. Can buy a used one for $14000USD only 3x your budget you should save up and buy one. https://www.machmarket.com/sale/mach...vmc-75/as17905 why do people think that you can design a machine that you can pick up and put in your boot can do the same job as a 9T machine? The manufactures would make a 1T machine if they could....
    7) You need the same rigidity to cut metal small machine or big. If the tool flexs the cutter will slide over the surface and it won't cut (it will just wear out). If you look at your hand router cut its inconsistant, uneven etc. That's because the teeth are bouncing around and maybe cutting every 3x around or 5x around or none at all. A sloppy machine will chatter and skid and slide, it has to be stiff
    8) with aluminium if the tool slides (rather then chipping) it picks up some metal then some more and suddenly you gum up your tool and its cactus (the tool and the job) . So it has to be stiff for different reasons then steel. Cutting aluminium has to be clean and sharp with lots of lubricant to stop pick up.

    9) Oh did I mention your machine has to be very very stiff? some benchtop mills have been measured at 2N/um but they complain about chatter.

    What are you studying? cheers Peter

    Hi Routalot - robots are increasingly being used for applications that would usually be a mill or trimmer etc. They can pick up over a tonne these days. Rhino and freecam have free robotic CAM systems so the math is already done for us. We just have to make the machines...



  12. #12
    Member
    Join Date
    Jun 2010
    Location
    Australia
    Posts
    4008
    Downloads
    4
    Uploads
    0

    Default Re: Thoughts on a 5 axis router/welder?

    The combination of low cost (like $500 and PVC pipe) and 5-axis and 'router' is not one can take seriously.
    This is not meant to be rude; rather it is a statement from experience. It is good to learn from experience, and cheaper in the end.

    As for 5-axis CAD/CAM/SW ... I know you can get something fairly good for $50,000. Similar comments.

    Cheers
    Roger



  13. #13
    Member peteeng's Avatar
    Join Date
    Jul 2018
    Location
    dum dum
    Posts
    1891
    Downloads
    0
    Uploads
    0

    Default Re: Thoughts on a 5 axis router/welder?

    Hi Roger - Yes look at the 5 axis shopbot for $44000USD - https://www.shopbottools.com/products/5axis but we are adjusting Malechs expectations at the mo. Now the other thing Malech is you are mixing a wet mill with a MIG that produces heaps of spatter. Spatter is nasty very high hardness stuff. Sticks to surfaces better then epoxy and will kill a bearing in 3 secs. So you have challenges ahead. Don't stop Malech Keep at it, you may find the right rabbit hole... Don't eat the wrong side of that mushroom however... Peter



  14. #14
    Member
    Join Date
    Apr 2017
    Location
    United States
    Posts
    24
    Downloads
    0
    Uploads
    0

    Default Re: Thoughts on a 5 axis router/welder?

    Thanks Peter, RCaffin, and routalot.

    I am currently a junior mechanical engineering student. I've run and programmed a Velox 5050 CNC router when I was on my high school robotics team. I'm currently helping out a team near me and they have an OMIO x8 machine. I'd say I like the OMIO more, it's got ballscrews, linear rails, and a real spindle compared to the Velox which was leadscrews, round rails, and a Bosch router. The OMIO is also a lot smaller, it definitely fits the space for the new team rather than taking up a 7 foot by 7 foot area like the Velox did. The reason I brought up the Monarch VMC was to explain what I don't want out of my machine, if that makes sense. I just don't need all the capabilities it had.

    I do try to use the high speed toolpaths whenever I can. I like the adaptive toolpaths that Inventor HSM has, and I was able to get some pretty good material removal rates with those toolpaths and a single flute bit on the OMIO. For the Monarch, I was programming in MasterCAM, but it didn't matter anyway since the machine memory was only 50kb, so HSM was pretty much unavailable.

    I have considered going the anthropomorphic robot arm route instead. One of the difficult things with those seems to be the joints. It's my understanding that many of the better robotic arms use harmonic drives in the joints in order to get low backlash and a deep reduction. Making one of these drives myself seems to be out of the question since there's a lot of work needed on the materials side, so it would seem to me that building a robotic arm yourself would be mostly buying and integrating high performance parts. If this is a learning exercise, I'd like to be able to make as many parts myself as possible.

    Thanks,

    Malachi



  15. #15
    Member peteeng's Avatar
    Join Date
    Jul 2018
    Location
    dum dum
    Posts
    1891
    Downloads
    0
    Uploads
    0

    Default Re: Thoughts on a 5 axis router/welder?

    Hi Malchi - Its a good exercise to make parts yourself but would you make a roller bearing? Would you make a worm drive gearbox? Would you make a linear rail? No unless that's your business. Harmonic drives are stock parts you buy what you need and integrate them into your design. There is lots more to do then reinvent the wheel. You are going to need lots of wheels where your going. Better to be a master at lego in some ways then try making lego then make your machine. Read up on the engineer that invented the harmonic drive. He was a genius and revolutionized many things. You can use steppers to run an antro there is a kit antro on the market. If a std part or integrated part exists then use it. Put your effort into the bits that don't exist. In our profession we stand on the shoulders of the giants that precede us. You have to take advantage of their efforts to move forward....to become a giant yourself...Peter

    Home Walton Musser

    Last edited by peteeng; 09-13-2020 at 06:09 PM.


  16. #16
    Member
    Join Date
    Jun 2010
    Location
    Australia
    Posts
    4008
    Downloads
    4
    Uploads
    0

    Default Re: Thoughts on a 5 axis router/welder?

    I looked at the shopbot page.
    The tool is perfect for 5-axis machining of soft materials such as urethane foams, as well as trimming, drilling, and slotting operations in a variety of materials such as vacuum formed ABS. It is capable of light machining in soft woods, but does not have sufficient rigidity for machining of hard-wood or aluminum.
    Well, yes, I guess so.

    Malachi: Making all your own parts? OK, that could be fun. If that is the goal (at first), perhaps 5 ply and tubing are suitable. Ah, but eventually you will have to buy motors and electronics. All very sad, that part!

    Cheers



  17. #17
    Member
    Join Date
    Apr 2017
    Location
    United States
    Posts
    24
    Downloads
    0
    Uploads
    0

    Default Re: Thoughts on a 5 axis router/welder?

    RCaffin,

    Yes, I would like to make many of my own parts. I'm interested mostly in the mechanical side, so I'm fine with purchasing motors and electronics. There's still personal value in making some of your own parts though, no?

    Peter,

    The drive mechanism that I mentioned in my first post is something new that I have yet to prototype, but could be useful. You may know of someone already doing something similar. What I want to try is combining a chinese windlass with a capstan. I have to draw up more of this but the idea would be that the mechanical advantage is based on a difference in diameters of the two spools, just like in the chinese windlass, so you could generate an sort of reduction you need. One of the challenges with this is that you need an infinite loop of cable without an increase in diameter at the splice. Another is the extensibility of the cable itself. Anyway that's a bit of a tangent but I think it addresses your point about putting effort into bits that don't exist yet.

    Malachi



  18. #18
    Member peteeng's Avatar
    Join Date
    Jul 2018
    Location
    dum dum
    Posts
    1891
    Downloads
    0
    Uploads
    0

    Default Re: Thoughts on a 5 axis router/welder?

    Hi Malachi - Does your rope/cable gearbox have an application? Look up 12 strand high modulus rope such as dyneema or DynIce75 | Dyneema Rope with 12 strand you can do very long splices that change the diameter little. I think you may have to braid your own loops eventually to get it right. This shows my age but when I was a kid the valve radios used cable drums to turn the tuners from the knobs same deal I expect. Peter



  19. #19
    Member
    Join Date
    Apr 2017
    Location
    United States
    Posts
    24
    Downloads
    0
    Uploads
    0

    Default Re: Thoughts on a 5 axis router/welder?

    Peter,

    I promise I will eventually 3D print a prototype and make a video about it. I have heard of Dyneema. There is another interesting actuator called a twisted rope actuator being used in research on artificial hands and they seem to make extensive use of Dyneema.

    As far as uses go, I'm not sure yet. One of the advantages I think my mechanism would have if it were used in a linear actuator fashion would be no backlash and a potentially very high reduction. Maybe in places where a long screw is not possible but you still need a deep reduction straight off the motor? Or in the case of a robotic arm, the cable coming off of my mechanism could couple to a cable wrapped around the pivot of the arm in a capstan fashion. Alternative to a harmonic drive? There are a lot of things I haven't thought about yet.

    Malachi



  20. #20
    Member peteeng's Avatar
    Join Date
    Jul 2018
    Location
    dum dum
    Posts
    1891
    Downloads
    0
    Uploads
    0

    Default Re: Thoughts on a 5 axis router/welder?

    Hi Malachi - wave drives or harmonic drives can have ratios in the 100s 1000s & no backlash. A capstan say 10mm dia to get to 500:1 will need a wheel of 5000mm dia to get this. 5000mm is a bit big for a part usually. But keep thinking. Maybe find a problem to solve to focus the bubbles. Cable drives suffer from stretch, the cable walks around on the drum, it needs a tensioning mechanism to maintain friction.... If its used in a positioning application these factors mean you don't know where the effector is. You'll need a feedback system to do this and these are getting cheaper... Peter twisting rope is sort of how actual muscle cells work I think...I prefer vectran it does not creep or relax, look up crystalline vs non crystalline plastics

    sidetrack https://www.taskaprosthetics.com/



Page 1 of 2 12 LastLast

Tags for this Thread

Posting Permissions

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


About CNCzone.com

    We are the largest and most active discussion forum for manufacturing industry. The site is 100% free to join and use, so join today!

Follow us on


Our Brands

Thoughts on a 5 axis router/welder?

Thoughts on a 5 axis router/welder?

Thoughts on a 5 axis router/welder?