584,842 active members*
4,621 visitors online*
Register for free
Login
Page 1 of 2 12
Results 1 to 20 of 35
  1. #1
    Join Date
    Feb 2016
    Posts
    58

    Steel Gantry Router

    Hi all,

    After a few years with my Shapeoko 3 XXL, it is time to build myself a more serious machine. I have done considerable reading on here and other sites over the last year or so and though I know what I want to achieve, I would like a bit of input into my design from those more experienced members.

    Mechanical:
    - Work area minimum 1250x1250mm. More is ok, but must fit a half sheet of plywood
    - Z-axis travel 150-250mm – I’d like the clearance to fit taller jobs in
    - Able to comfortably mill aluminium. This is always a contentious comment, my Shapeoko *can* mill aluminium, but not comfortably. I don’t expect it to cut like a tormach.
    - Steel frame, happy to use Aluminum for brackets & smaller parts.
    - Steel parts will all be welded. At this stage Heat treating is not feasible, So will weld 20mm sections evenly around the frame to keep temperatures and hopefully warping to a minimum
    - Hiwin linear rails all round - 20mm seems well suited.
    - 10mm strips of steel welded to beams on linear rail mounting points. Machined flat.

    Steel sizes:
    Gantry - 200x100x6mm RHS - 1500mm long - Now 200x200x6mm
    Side beams - 150x100x5mm RHS - 1500mm long
    Cross supports - 50x50x5mm SHS - 1350mm long
    Gantry/bearing plates - 10mm steel plate
    Z-Axis(when I get there) will be 16-20mm plate


    Electrical:
    - DMM 750w servos direct drive to (originally 2010) Now 2525 ballscrews. - Not settled on this setup yet - I’d like to avoid belts or any reduction if possible. Now 3:1 reduction. Not sure what rapid rates I should be targeting, maybe 6000-8000mm/min Now 25m/min with screw @ 1000rpm
    - There are substantially cheaper 750w AC servos on Aliexpress, at close to half the cost. Not sure if anyone has any experience with these, obviously DMM are better but I don’t want to blow the budget out
    - Centroid Acorn control board. Also now considering Mesa 7I77 and LinuxCNC
    - Power supplies yet to be decided
    - 3-4kw spindle – yet to be decided


    So I started drawing up the below design (first two pictures), I got the basic bed and gantry set up, I’m quite happy with this design so far. It seems quite sturdy and has the added benefit of keeping the rails and ballscrews well away from the bed and the majority of dust.



    I then started googling for router designs for some z axis ideas and realised that there are not a huge amount of machines designed like this, and most commercial designs have the linear rails at bed height or below, with a ‘gantry riser’ style of design. This doesn’t feel as rigid as what I have designed, but I whipped up a quick draft of how this might look. Don’t be too harsh it’s just a concept at this stage.

    Any thoughts on what might be the better option? I haven’t put a lot of time into the gantry riser design, It needs a lot more support on the ends of the gantry and on the riser-bearing joint. The first design is my preferred option, it just feels a lot more rigid, and the joint from the gantry to the Y rails is very solid.

    Thankyou for any input, much appreciated!
    Attached Thumbnails Attached Thumbnails R1.jpg   R2.jpg   flat 1.jpg   flat 2.jpg  

    flat 3.jpg  

  2. #2
    Join Date
    Mar 2003
    Posts
    35538

    Re: Steel Gantry Router

    Not sure what rapid rates I should be targeting, maybe 6000-8000mm/min
    That machine should be able to cut plywood at 10m/min. I'd be looking for 15m/min or more rapids.

    3000rpm motors would give you 30m/min, but you don't want to spin the screws that fast. A better choice would be 2525 screws with a 3:1 reduction. That gives you 25m/min rapids with the screw spinning at 1000 rpm.
    Gerry

    UCCNC 2017 Screenset
    http://www.thecncwoodworker.com/2017.html

    Mach3 2010 Screenset
    http://www.thecncwoodworker.com/2010.html

    JointCAM - CNC Dovetails & Box Joints
    http://www.g-forcecnc.com/jointcam.html

    (Note: The opinions expressed in this post are my own and are not necessarily those of CNCzone and its management)

  3. #3
    Join Date
    Feb 2016
    Posts
    58

    Re: Steel Gantry Router

    3000rpm motors would give you 30m/min, but you don't want to spin the screws that fast. A better choice would be 2525 screws with a 3:1 reduction. That gives you 25m/min rapids with the screw spinning at 1000 rpm
    Thats awesome info, thank you. I didn't expect to need reduction as I didn't think those rapids would be possible. What sized pulleys and belts should I look at for that sort of torque/rpm? I'm digging into research now but would something like a 5mm pitch HTD belt of 20mm width work? and 20/60 tooth pulleys? or are GTx belts a better idea?

    Is there any guide of how far apart the pulleys should be to minimise belt length?

  4. #4
    Join Date
    Jan 2008
    Posts
    1523

    Re: Steel Gantry Router

    Gantry beam should be 200x200.

    The Z axis induces a torsional force on the gantry. The forces are not just in X, Y and Z.

    Add diagonal bracing within gantry beam. Close the gantry beam ends also.
    7xCNC.com - CNC info for the minilathe (7x10, 7x12, 7x14, 7x16)

  5. #5
    Join Date
    Feb 2016
    Posts
    58

    Re: Steel Gantry Router

    Add diagonal bracing within gantry beam
    Ok Cool, I'll bump it up to 200x200 and see how that looks. should I increase the thickness of the beam or is there not so much benefit in that?

    The design that I think I will go with (first 2 pics in OP) does have end plates on the gantry, that attach to the Y rail bearing blocks

    As far as diagonal bracing goes, I saw a build recently (that I can't find now) where the guy drilled holes in the gantry beam at opposing corners, he then welded in 20mm round bar and ground it flat. I'm thinking of doing that every 200mm and alternating the direction (as shown in attached primitive drawing).

    Would that suffice?

    *edit* I came across this post and am taking a few tips from it too. https://www.cnczone.com/forums/diy-c...ml#post1858440

    Still not sure if I leave the Y rails as the 2x 150x100 beams or build risers, I can't find anything definitive or any reason to go one way or the other

  6. #6
    Join Date
    Feb 2016
    Posts
    58

    Re: Steel Gantry Router

    So I've continued on with the gantry on raised sides design, It looks the most rigid so I will run with it.

    There is 290mm from the bottom of the gantry beam to the cross beams on the bed, I'm thinking of putting a 10mm steel plate as the base then a sheet of phenolic as the wasteboard which will bring it up to around 260mm - the Z-travel will allow a 1/8" endmill to hit the wasteboard, and a 50mm endmill to clear the bottom of the gantry. - the exact locations of the linear bearing carriages will be adjusted when I confirm spindle choice/dimensions

    Still working on the ballscrew locations, Z-axis lwill likely change a little to accomodate the ballscrew. I'm thinking I will just weld plate under the Z rails then machine flat to increase the space in there to fit the ballscrew & nut.

    The two Y-axis ballscrew mounts look a bit flimsy to me so I will likely make them more rigid by increasing plate thickness or putting a perpendicular brace in there.

    I'm not commited to electronics yet, I like the versatility of MESA card + LinuxCNC, but UCCNC and Centroid both seem pretty turnkey and a bit easier to set up

    Still a ways off ordering anything but I'm comfortable with how the basic frame is looking.

    Questions:

    1. Any comments on how rigid the Z-axis is? I plan to make both plates out of 10 or 12mm steel plate, thicker is no problem.
    2. Would it be better to weld the braces on the XZ carriage or drill, tap and bolt them on?
    3. Is there any major benefit to placing the X Ballscrew between the rails, or would sitting it on top of the gantry close to the carriage be acceptable?

  7. #7
    Join Date
    Jul 2018
    Posts
    6249

    Re: Steel Gantry Router

    Hi Stuart- You won't need the internal bracing in the gantry tube, anything inside your gantry tube is wasted effort . So far your designing a very rock solid machine. To match the stiffness of your machine you will need a very thick Z axis plate something like 1" to 1.25" or use a heavy channel or use a 1" plate but screw webs either side of it say 50-100mm high. 12mm steel plate for the Z axis is not near stiff enough to match the rest. Peter

    https://buildbotics.com/ consider this controller/driver/ all in one box. Seems to be very good.

    2. If you weld the flanges to the bearing plate it will change shape and need to be machined. If you can machine it then all good. If not then edge bolting is your next best option. Or find a large SHS or RHS and cut it down. Maybe these some large channel out there?
    3) either is fine. In between is slightly better but makes the stack higher. So the spindle will be outboard further.

    Re: welding if you have oxy welding equipment and are competent with it then bronze welding or brazing is a better plan. It normalises as you go because of the larger heat zone and because you don't melt parent, distortion and stresses are minimised.

  8. #8
    Join Date
    Feb 2016
    Posts
    58

    Re: Steel Gantry Router

    That’s good info thank you Peter, I will see what plate I can source for the Z, I remember seeing some guys use thick steel tooling plate that’s been ground or machined on both sides, I will look into that. I intend to buy a manual bench top mill (meted DM45) for this project so will be able to square up the ends of beams and mill plate. I’ll also look into heavy channel. Even something huge that I can mill down a little may work.

    Do you mean webbing as I have in the last post? The X-plate has 100mm and the Z plate 75mm, is that what you are referring to? Just a thicker gauge?

    The Buildbotics looks like a good controller, but it doesn’t look like it supports running Servos the way I’d like to. I intend to run the DMM servos in Analog, with encoder feedback to both the driver and to the control (I know a lot of guys do this with LinuxCNC) part of this is to have very accurate motion control, part is because I want to learn how to do it.


    At the moment I am planning to go with standard Chinese rolled ballscrews, i understand this will be a weak point but ballscrews are an easy upgrade, and I will blow out the budget going with ground screws straight up

  9. #9
    Join Date
    Jul 2018
    Posts
    6249

    Re: Steel Gantry Router

    Hi Stuart there is a large discussion here about belts.

    https://www.cnczone.com/forums/linea...84496-cnc.html

    and I've just spent a few months working thru a similar machine. Look at

    https://www.cnczone.com/forums/austr...ml#post2312728

    cheers Peter

    also as a thought, its advantageous to be able to slide the entire Z axis assembly off an end or both ends if possible. So look at the ends and Z plates to see if you can do this. Saves a lot of work sometimes vs pulling the Z apart piece by piece to get deep into it somewhere.

  10. #10
    Join Date
    Jul 2018
    Posts
    6249

    Re: Steel Gantry Router

    Send buildbotics a note, they respond very fast and being the designer and builder they know their stuff really well. But didn't think to look if it will run servos....Peter

    Re Z axis - The z axis is a really good place to use aluminium. A 25mm thick steel plate say 200mm wide weighs 39kg/m yet a 36mm thick aluminium plate is the same stiffness yet weighs 19kg/m which is a huge advantage in moving it around. Since your getting a mill maybe a large chunk of aluminium is the go for the Z axis plate. I'm near Byron Bay so were about as far away as we can get in OZ. Peter

  11. #11
    Join Date
    Feb 2016
    Posts
    58

    Re: Steel Gantry Router

    The buildbotics looks interesting, Though I can't find anywhere what type the controller is: Mach/LinuxCNC/GRBL/proprietary etc.Also I don't like the web based interface, it's an interesting concept nonetheless.

    You make an Interesting point about using aluminium, I may cost some aluminium just to get an idea and go from there. 450x200x36mm (or closest size) is not likely to be cheap, but I will check it out. As far as I can tell weight savings on the XZ-carriage are a bonus, though cost will be the deciding factor.

    That is some excellent info on the belts too, thankyou - I had searched on here for any belt-related posts but didn't come up with much. There is plenty of reading for me to do there.

    I am actually based in Armidale NSW, I moved from Perth around a year ago. So not too far away! Can you recommend any good suppliers for metal, linear motion or electronics that I can get some quotes from? at this stage I'm using local guys for steel and ebay/aliexpress for linear motion & electronics, but happy to travel to Sydney, Brisbane or between to save a few bucks or acquire the right gear.

  12. #12
    Join Date
    Mar 2003
    Posts
    35538

    Re: Steel Gantry Router

    I think it might be based on the TinyG, as that's the only hobby controller I know of with S Curve acceleration.

    If you want to run analog servos, that limits your options to LinuxCNC, KMotion, or Mach4 with one of the few Analog controllers that work with it.
    Gerry

    UCCNC 2017 Screenset
    http://www.thecncwoodworker.com/2017.html

    Mach3 2010 Screenset
    http://www.thecncwoodworker.com/2010.html

    JointCAM - CNC Dovetails & Box Joints
    http://www.g-forcecnc.com/jointcam.html

    (Note: The opinions expressed in this post are my own and are not necessarily those of CNCzone and its management)

  13. #13
    Join Date
    Mar 2003
    Posts
    35538

    Re: Steel Gantry Router

    Looks like it did indeed start with the TinyG, but now uses their own code.

    https://news.ycombinator.com/item?id=17288432
    Gerry

    UCCNC 2017 Screenset
    http://www.thecncwoodworker.com/2017.html

    Mach3 2010 Screenset
    http://www.thecncwoodworker.com/2010.html

    JointCAM - CNC Dovetails & Box Joints
    http://www.g-forcecnc.com/jointcam.html

    (Note: The opinions expressed in this post are my own and are not necessarily those of CNCzone and its management)

  14. #14
    Join Date
    Feb 2016
    Posts
    58

    Re: Steel Gantry Router

    Quote Originally Posted by ger21 View Post
    If you want to run analog servos, that limits your options to LinuxCNC, KMotion, or Mach4 with one of the few Analog controllers that work with it.
    Ok, I thought that might be the case. What's your view on how to run the servos? I can only really go off what I've read, and the general consensus I have seen is that Step/Dir is OK, but the preference is to run analogue with encoders back to the drives, then also back to the motion control(for full closed loop) as it's smoother and more accurate..

    If you were setting up servos on this machine what controller would you run and how would you set it up? I'm very flexible at this point as I haven't ordered anything, and very inexperienced - the only motion control I've set up is 4 Nema23 steppers off a Mesa 5i25 > Gecko540


    *edit* Also, does anyone know good suppliers for 5mm GT closed belts and pulleys that would suit my servo reduction? It seems like 60/20T or 72/24T at a 15mm width would work. Still digging through the material above, but no specific suppliers

  15. #15
    Join Date
    Mar 2003
    Posts
    35538

    Re: Steel Gantry Router

    I'm using DMM's on the machine I'm building, but running them step/dir.

    Not sure there would be any noticeable difference in smoothness, and accuracy should be exactly the same, as both methods use the same encoder, and the motor is always moving to a specific encoder position.

    Scratch Mach4 from that list, as it's can't close the loop in the control.
    Gerry

    UCCNC 2017 Screenset
    http://www.thecncwoodworker.com/2017.html

    Mach3 2010 Screenset
    http://www.thecncwoodworker.com/2010.html

    JointCAM - CNC Dovetails & Box Joints
    http://www.g-forcecnc.com/jointcam.html

    (Note: The opinions expressed in this post are my own and are not necessarily those of CNCzone and its management)

  16. #16
    Join Date
    Jun 2009
    Posts
    251
    Quote Originally Posted by ger21 View Post
    That machine should be able to cut plywood at 10m/min. I'd be looking for 15m/min or more rapids.

    3000rpm motors would give you 30m/min, but you don't want to spin the screws that fast. A better choice would be 2525 screws with a 3:1 reduction. That gives you 25m/min rapids with the screw spinning at 1000 rpm.
    At 1.5 meters length spinning the ballscrew at 1000rpm is looking for trouble, big trouble. At 800 they started vibrating, 2505 that is. And it is 1.5m for 1.25m cut area or thereabouts.
    There are calculators online that can give you a rough idea of what you can spin them.
    You should check that as it has been several years since i did that, but am pretty sure that was it. The Y axis was 2.5m with ballscrews on both sides, had to setle for 2m/m. What a pitty.

  17. #17
    Join Date
    Jul 2018
    Posts
    6249

    Re: Steel Gantry Router

    Hi Stuart,
    1) Gerry recommended https://bstmotion.aliexpress.com/store/314742 and I have emailed Fred there a few times with questions and he has been very good
    2) For pulleys etc start at https://www.naismith.com.au/list_products.php?m=42 they have been good to deal with. They currently can't supply ATL belts but I can get these if you want to go that way
    3) Yes the web based thing is my hurdle with BB at the moment. I have a machine in a shed with no web access. BB have sent me 3 options to counter that. I'm working thru them now. I'd like to use it on my next machine.

    Peter

    Re buildbotic controller/driver- I've just go a reply from Doug. To use locally without internet, get an ethernet switch and plug BB into the switch and a computer into the switch and Bobs your Uncle, it works. Peter

  18. #18
    Join Date
    Jul 2018
    Posts
    6249

    Re: Steel Gantry Router

    Hi Stuart, not so far away... I'd like to play devils advocate on your project.

    1) Your about to buy a manual mill and spend about $8-10k on a half sheet machine. Plus invest 6 months to a year to get it done. You can buy commercial machines for around $10k and your done, on your way?
    2) Your a Maker and good at CAD. You have chosen steel as its convenient and you can probably weld steel yourself. You have started at the usual design space of designing a bench and a gantry but the real effort should be put into the Z axis. That's the crux of the machine and the pointy end that does the work. No value in having an UBER/bench gantry and a wimpy machine holder. Especially when it comes to "aluminium".

    So to discuss the SHS or RHS gantry solution. a) Its convenient but b) it has issues. The corners have large radii so you have to mount the rail a fair way from the corner. The faces of SHS are cupped due to the rolling process, so if you attach to the face, the rails will not be horizontal. Ideally the rail should be mounted to the web (to transfer the loads directly to the SHS) but in this case can't be done. Some people weld a thick plate to the front then machine this to create the mounting surface. But this requires the gantry to be stress relieved as it will definitely change shape when welded/machined. So now you have a rail mounted on a 6mm thin piece of metal that can flex slightly when loaded. This is also the case with other elements using this construction philosophy. All of these flex points add up...

    Moving parts need to be light weight - see the Z axis plate as an example. The same should be considered for the gantry it should be aluminium to allow a very stiff and light structure to be made so it's easily driven. If you make it in Al then you can machine the parts with registers and features that make alignment easier. (using your manual mill or a CNC) Billet machined parts bolted and bonded together is a very successful machine build technique. Use Loctite 290 or 294 to set joints once correctly assembled for instance. (there are many examples in this forum)

    Aligning this large and heavy steel gantry that has no machined features will be a mission and keeping it in line in use as well (cutting aluminium involves large loads that will flex the machine) If you aim at a precise machine to do Al then this will be a bug bear when you put on your machine assembler hat. CAD objects are really easy to move around on the screen!! I move a 360Tonne trailer with my little finger!!

    I suggest the following:
    Since you are interested in servo control and one objective of this machine is to learn about that, make a smaller benchtop machine that can cut Al parts. Use that machine to build the big machine parts. The big machine should be steel bench, aluminium rest. You could also buy a small benchtop CNC that does Al and move along as well I think within my guess at your budget.

    I think you need to write down your objectives and budget and maybe there is a way through this that will result in a better machine then the current "plan".

    I say all of this because I design machinery in the mining, medical and yachting world and see a lot of projects consume time and money that could be better spent. Project inceptions are always enthusiastic and $$$ do not matter. At some point all this changes and reality bites and your well off expectations So do not buy anything until you have fully resolved the design in CAD, sourced everything and costed it. Then at least you have a well understood design and have only spent your time on the project (Hobby time). You may get near to the end and scrap the design due to cost or technical issues. There are a few traps along the way, like designer inertia. This is when you know something has to be changed but you don't because A) you are attached to the sunk time in the design and B) you know it will take more, more time to do the changes. If you don't do the change then you have compromised the work and the result. This approach also will allow the machine to be built in the shortest time as everything will fit and everything is available. The forum is full of projects that replace this and change that and never seem to get to the correct resolution.

    Enjoy the journey, Happy to help... Peter

    Re buildbotic controller/driver- I've just go a reply from Doug. To use locally without internet, get an ethernet switch and plug BB into the switch and a computer into the switch and Bobs your Uncle, it works. Peter

  19. #19
    Join Date
    Feb 2016
    Posts
    58

    Re: Steel Gantry Router

    Of course someone has to play devils advocate!

    You make some valid points, and I have certainly considered most of your thoughts prior to this.

    1) yep you're right on the money, I'm budgeting AU$10k. I have considered commercial machines, but could not find anything <$10k that would cut half a sheet and also be rigid enough to cut aluminium. Would you mind posting some links for me to investigate? You also mention buying a benchtop CNC and moving on - to which brands are you referring? I feel like I've been looking in the wrong places

    2) Yes steel is easy for me to cut and weld, I can currently machine aluminium on my little router but not to the accuracy that I'd like. A point I'd like to make here is that I understand the Z axis isn't as rigid as it should be. The whole reason I am posting my design here is to get input from people to help me design something that isn't 'wimpy'. Though I am learning, I don't have the knowledge to just come up with the ideal design, that's why I engage you wonderful people. (let me know if you ever need someone to operate an LNG plant for you - now that I can do!)

    I understand your Issue with using SHS/RHS. However I have seen some pretty good designs come out of here and mycncuk.com that use similar techniques and have achieved good results. I have seen several builds where if welded carefully, the bearing rail mount can be machined without needing to stress relieve the box sections. I've also seen guys successfully epoxy pour the mounting locations for the linear rails - this seems to be more popular in the UK than america, but they have good success with it. Again this is my point, I've not come here to show people how to build a machine, I'm asking for input as it's a project I'd like to do I just need a little guidance.

    I have goals of CNC'ing the DM45 benchtop mill, I know thats likely a more simple project than building a CNC router (depending on how far I go with PDB and ATC), and it'd certainly cut aluminium without issue. I don't know if that will come before or after the router.

    I appreciate your comments on the fact that a lot of projects run out of energy/funding, or become so tied up due to design flaws that they don't ever get finished. I have no intention of rushing this, if it takes me 6 months, a year to get the design sorted then I'm ok with that.

    I will get my design brief down so it is more clear. I have a good idea of what my requirements are and I'm just starting to get some rough ideas of costs together now, This sort of comes alongside the design.

    A final note I'll add is: This project for me is as much about building the machine as it is actually using it to do work. I have found with my current CNC machine that I spend as much time, and get as much enjoyment from modifying it, making new parts, upgrading parts, as I do from designing and cutting jobs on it. That being said I do make an income with it, but being locked down to an 800x800 work area makes working with sheetgoods pretty inefficient, and limits the size of my jobs for signage.

    Thankyou for the input

  20. #20
    Join Date
    Jul 2018
    Posts
    6249

    Re: Steel Gantry Router

    Hi Stuart - I'm not saying your ideas won't work, I'm trying to say in a project like this it is good to think bigger then the convenient solutions. It's a big task and long journey and there maybe opportunities that result in a better machine for no extra effort other than thinking different. There are all manner of solutions and they all end up working if you persist. It comes down to how good a machine do you need and can you realise it. So far your travelling the right road... and you are brave to put this out there and there will be flack but you and your machine will be better for it.

    1) I think the epoxy levelling thing is a poor solution, sure use epoxy to set a machine but levelling is another matter. I have been involved with concrete leveling and epoxy levelling and it's a tough gig. Wouldn't go there, there are better approaches
    2) Sure welding carefully can get you so far but the bottom line is it's a compromise process for the application. I do recommend brazing vs welding. I have worked as a welder, a welder trainer and a robotic welding programmer and distortion is inevitable and permanent so tread lightly
    3) I understand your a Maker and that's the fun bit, Tally Ho!!


    https://www.gregmach.com/product/woo...0-cnc-machine/ this was $12k AUD at a machine show a years ago. I see the price is removed and the smaller version is $12k.

    What CAD are you using? Peter

Page 1 of 2 12

Similar Threads

  1. Granite and steel gantry router
    By multivactubes in forum DIY CNC Router Table Machines
    Replies: 27
    Last Post: 07-07-2019, 07:39 PM
  2. Lessons in overkill! Welded steel, fixed gantry router
    By mcentric in forum DIY CNC Router Table Machines
    Replies: 53
    Last Post: 10-23-2018, 08:53 AM
  3. CNC Drill Gantry/Router Recommended OEM's for Steel Panels
    By BradS. in forum Uncategorised MetalWorking Machines
    Replies: 3
    Last Post: 10-25-2017, 05:34 PM
  4. 80/20 and Steel Gantry Router Build
    By Adamj12b in forum DIY CNC Router Table Machines
    Replies: 14
    Last Post: 07-26-2015, 06:11 PM
  5. Steel Gantry Wood Router
    By gavztheouch in forum DIY CNC Router Table Machines
    Replies: 19
    Last Post: 09-01-2012, 06:37 PM

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
  •