Page 2 of 3 FirstFirst 123 LastLast
  1. #11
    Silyavsksi,

    Thanks for your feedback. I've learnt much from your posts (don't know how you keep up with them!). I'll attempt to address each of your points in the order you have raised them.

    1. Z axis travel, bed depth ,4th axis, gantry

    -Z 200mm is ok and going more will lead to another problems, which would need reinforcement, that would l lead to more weight and that to servo motors.
    I am doing a similar build so 200mm is ok for all, even for plasma, as i intend to do also.

    The Z travel I have is 220mm, which is a standard length by THK Actuators, so presume you are saying this okay. The THK actuators at this length are very accurate, so I need to concentrate on the gantry to ensure it is robust enough/will not flex. I'm not sure what it will cost me, but I still feel it will be more accurate than what I will construct, with less effort!

    -that singular gantry beam contradicts with the 200mm Z travel. 2 beams, say 100x100x4mm welded together will be better. Vibration wise.


    I wander if this is over engineering it? My gantry (pictured above) is RHS 250 (height) x 150mm (wide) x 5mm (thick), reinforced partially by two welded 10mm steel plates over carriages for X axis. I have utilised RouterCNC's extremely useful stiffness calculator posted, and using my dimensions and cutting aluminum, came up with deflection of Z = 6.3um and Z - 1.6um, which I find more than acceptable. Using the new Stand-alone Z axis calculator, at the extreme Z Axis extension of 220mm (which I would only use for wood turns), I get a deflection of 4.185um. Unless I am using this tool incorrectly, I would think this accuracy is more than enough for most builds?

    Perhaps there is good reason to make it more robust. You mention vibration. Is this a similar consideration to deflection, or something else that needs to be factored?

    I've attached the spreadsheet below.

    -that removable bed you have drawn is mistake. Either make it proper removable bed like some designs here make it, not with angles but bolted directly side of the beams, or calculate the correct bed depth for 200mm axis and make it fixed. For a fixed bed, depending on design you need more or less 300mm from gantry beam when all mounted.
    Thanks. I inserted brackets for greater accuracy so that the removable bed could physically sit on something solid (I am concerned that side bolts will have play and the removable bed would shift considerably when removed and then re-attached (without resting on something solid (i.e. bracket) - bolts move)). I could also bolt the removable shelf to the main frame for greater sturdiness. If there are sturdier and easier designs can you point me in the right direction?

    The fixed bed is already there for the maximum Z axis / 4th axis (approx 373mm distance from tool).

    -4rth axis is no problem with say 300mm deep bed/from gantry beam/. The most rigid 4rth axis worth for that type of machine, could be made from mini lathe, as i intend to, and it enters in 300mm bed from gantry, or you could make the axis inside the bed if you intend to spin things bigger than 300mm OD
    Thanks. I think we have similar designs in mind, but I do need to give more thought on what the 4th axis will be (phase 2 of the build which needs more thought now!). I've discovered your dialogue on your most recent build which I'm absorbing!

    Again, many thanks for your interest and input!
    Attached Files Attached Files

  2. #12
    Nice drawings.
    I can't compete with silyavski on the frame design but I did notice the spindle is cantilevered well out, have a look at this diagram where it shows the spindle should ideally be inside the footprint of the bearings at each end of the gantry. http://www.cncroutersource.com/do-it...NC-router.html
    Last edited by EddyCurrent; 01-10-2014 at 05:04 PM.
    Spelling mistakes are not intentional, I only seem to see them some time after I've posted

  3. The Following User Says Thank You to EddyCurrent For This Useful Post:


  4. #13
    Thanks Eddy. I thought my design was too simple. Now I know why! Thanks for the link. It was extremely helpful in explaining COG and design considerations. Conceptually, I will look to draw the shape of a RHS section at right angles to my gantry to move COG back towards Y axis rails.

    I've noticed that a number of designs have placed Y axis screw on top of gantry (my design has it on side of gantry between rails). Placing it on top would help marginally with moving COG back. Is there an inherent preference for placing screw on top? Is my design inferior or not practical?

  5. #14
    Quote Originally Posted by Tinkerer View Post
    The Z travel I have is 220mm, which is a standard length by THK Actuators, so presume you are saying this okay. The THK actuators at this length are very accurate, so I need to concentrate on the gantry to ensure it is robust enough/will not flex. I'm not sure what it will cost me, but I still feel it will be more accurate than what I will construct, with less effort!
    Its very difficult to find the correct sized actuator for that job. I have been that way. And actually constructing it yourself you will reap all the benefits over ready actuator. You will waste more time and money searching for it. And most of the commercial designs have the rails mounted on the fixed plate, the fixed plate is too thing, all weights too much, travel is too short, the ball screw is not the correct pitch and so on. I spend months looking for a correct one. from the THK ones only the biggest ones are good.

    Quote Originally Posted by Tinkerer View Post

    I wander if this is over engineering it? My gantry (pictured above) is RHS 250 (height) x 150mm (wide) x 5mm (thick), reinforced partially by two welded 10mm steel plates over carriages for X axis. I have utilised RouterCNC's extremely useful stiffness calculator posted, and using my dimensions and cutting aluminum, came up with deflection of Z = 6.3um and Z - 1.6um, which I find more than acceptable. Using the new Stand-alone Z axis calculator, at the extreme Z Axis extension of 220mm (which I would only use for wood turns), I get a deflection of 4.185um. Unless I am using this tool incorrectly, I would think this accuracy is more than enough for most builds?

    Perhaps there is good reason to make it more robust. You mention vibration. Is this a similar consideration to deflection, or something else that needs to be factored?

    I've attached the spreadsheet below.
    Yeah but you have 250mm side that can vibrate and i suggest you 100mm side that can vibrate. Which drum will make louder sound 100mm one or 250mm one? Think of it like aditional internal bracing.
    Also i like working with one and only profile. And buy by the 6m. So as i see it i would do a machine from 80x80x3 to 100x100x4 and construct all i need from these ones. As they also cut perfectly with small saws at home.

    Quote Originally Posted by EddyCurrent View Post
    Nice drawings.
    I can't compete with silyavski on the frame design ...
    Thanks Eddy but if you remember we all came together with that design on my build, i just calculated it the cheapest and lightest possible for acceptable rigidity for ali jobs. Its very funny though, instead of making steel instruments sound, i started to learn how to make steel structures not make sound. Its no much different than Romans did it- Divide and Conquer

  6. #15
    Quote Originally Posted by Tinkerer View Post
    I've noticed that a number of designs have placed Y axis screw on top of gantry (my design has it on side of gantry between rails). Placing it on top would help marginally with moving COG back. Is there an inherent preference for placing screw on top? Is my design inferior or not practical?
    It makes sense to have the screw central to avoid twisting forces, for example mine is central but it's on the back. Sometimes people make a gap between two gantry beams and mount the screw in there.
    You have the rails mounted on your gantry front face, moving these to top and bottom would also move the COG back. Things look great on a drawing but when it comes to building it you have to consider how to align everything so it's accurate accros the whole working area. If you go through that setting up stage in your mind, such as how to get the two rails on your gantry ligned up in all planes, then that might help steer the design.
    Spelling mistakes are not intentional, I only seem to see them some time after I've posted

  7. #16
    Hi everyone, I've been off the air, school holidays, etc. I have taken on board comments from Eddy and Silyavski and embarked on a substantial redesign after spending literally days going through numerous design iterations, stiffness calculations, etc. The resulting redesign now has the following inherent features:

    1. Frame

    Largely unaltered, however, removing cantilever from the Z Axis (thanks Eddy!) required Y axis lengthening one side and reducing on the other.

    I have shown inserted bolts for the removable bed. I have used a combination of bolts as well as angle to attempt to remove and play movement when removing and re-installing it. I've review some other designs on the forum which appear to just use bolts to the side of the frame. I feel these designs would result in a significant shift in the bed from the original leveling. If I can avoid needing to re-level the bed, it would be a major advantage.


    As regards the Frame, most material is SHS 89mm, 3.5mm thick. I have gone to 6mm thick on the top where I need to weld brackets for motors, bearings, bolt Rails on X Axis. Can anyone advise whether 6mm is thick enough to directly tap (thread) and bolt bearing supports and rails to it?

    2. Gantry

    I was initially wedded to the simplicity and robustness of my RHS Beam - 250 x 150mm. After comments from Silyavski, I recognised the weakness of the extensive cantilever in my design. After several designs (utilising welded steel plates in triangular shapes) to reinforce the Gantry (to remove cantilever), I then had a further look at the RHS beam I was using.

    Although the beam was certainly robust enough, I grappled with the issue of the square face, necessitating me installing the ball screw shaft on the same front face (per my original design), thereby increasing the Z Axis overhang / cantilever by about 25mm. I then looked at placing the Y Axis screwball shaft on the top of the Gantry, thereby eradicating the increased Z Axis overhang. I had considered this in an earlier design, and for all purposes it is a better design (for Z axis), however placing the screwball behind the Y axis rails adds another leverage point, this time, on the Y Rails. I am sure this wold have been fine, but I knew that the ballscrew was better placed between and in line with the Y axis rails to avoid any leverage.

    So, I came to the conclusion many here already know, and that is, my Y Axis Beam needed to be formed (welded) to provision for the Y screwball shaft behind / under the Y rails. Silyavski, I looked at your design and thought there was way too much welding in it (like you I am a relative novice!). I experimented with all manner of available steel materials and eventually came to a close solution to yours. That is, after eventually discounting the following designs largely because of their added weight, with flow on impacts on motors, etc:

    Considered Gantry Designs (subsequently discounted)

    a. RHS 200 x 150, with 2 RHS 65 x 35 welded to the face to lift Y Rails away from beam (preferred solution as less fiddly/welding/cutting but added another 15kg+)

    b. Silyavski design - Two SHS 100 stacked high with smaller welded tube (75 SHS) to create the void for the Y axis. This would (and has) worked well for Silyavski, however I preferred utlising lighter material and therefore reinforced the X direction by utilising RHS (see final solution)

    c. 2 x RHS 150 x 100 separated by SHS 75 along the whole length (less cutting/welding). Again too heavy with too much unnecessary continuous length.


    My final solution - Utilising 2 x RHS 125 x 100 x 4mm, separated by 7 cut SHS 100 x 4mm
    (I have used 9mm for the end SHS separators to give more meat to tap into to bolt the Bearing Supports for screwball shaft).



    So Silyavski, I looked at your design and thought I would be able to improve on it/ create less work for myself. Ultimately, I have no doubt gone through the same process as you, and although I have changed some of the materials, I ultimately came to the same conclusions as you and intend to build a similar design (just not as bomb proof as yours!). Where you have opted for SHS 100 10mm thick, I have gone for 4mm, but increasing thickness to 9mm where bolts are to be fastened. I have used some RHS 125 x 100 x 4mm to strengthen the beam in the X direction due to use of lighter thickness.

    You have used 20mm and 12mm plate I think. I have opted for 10 and 8mm plate believing it to be more than adequate. I have checked most against the stiffness calculator and believe there will be relatively little movement below 10um in all cases whilst cutting aluminum.

    Z Axis

    I was looking for a simplistic solution, researched and found actuators which I could largely bolt onto plates, etc. There are a number of Actuators in the market from THK and others and I resolved that they would be able to manufacture a far more accurate product than I could. Feedback from Silyavski caused me to delve into designs a little more thoroughly and yes Silyavski, they are all quite flimsy, using thin walls and I believe would never have withstood cutting forces for a router.

    I spent considerable time playing with the stiffness calculator mentioned in my first upload. To get reasonable accuracy with up to a 100 - 200mm Z axis travel, I have needed to utilise Steel plate with perpendicular reinforcing ranging between thicknesses of 10, 20 and 30mm. I got myself a little challenged with the material stiffness calculator, as its design is really for gantries, but I think I was able to interpret it to apply to the Z Axis.

    So with my easy purchase solution ended in tatters (bug.er!), I embarked on more research to understand how Z axis work (I confess to initially glossing over what pieces were attached to each other - I thought I'd get to that once I purchased one!), looking at numerous designs and then designing my own.

    So, you will all be somewhat relieved to know that I now fully understand what parts move ( are driven) and are to be fixed to create a working Z axis! That is unless anyone advises me that I have got it all wrong!

    Here are views of my design. All steel construction. Yes, more welding and I think I will have to get at least one plate machined.




    Okay, after many many hours of toil, I lay myself open to criticism/comments from everyone. As I have found out, there is no easy way getting into this build. For me it has all been a big, but enjoyable, learning curve, and all the shortcuts I had hoped to take are not available for good reason. So, in time I expect top be able to provide you all with a few laughs - welding has me a little anxious, as is putting together the electronics. All in the fun I suppose.

    As always, I welcome any comments, but I will slash my wrist if I have overlooked something fundamental. This is a time consuming process which has taxed the brain and resulted in many hours of drawing and getting to understand Rhino software (which incidentally is great)!

    I'm anxious to start actually building, but thankfully did not leap into things ordering material, as my design has evolved considerably, thanks to your input and my own quest to optimise use of materials and the accuracy of the machine.

    I have made some progress:

    1. Bought my Everlast Welder and purchased (rented) my first bottle of Argon and consumables.

    2. Designed my steel workbench, which will be built first to get into a bit of welding practice and to assist in having a good surface to square things etc. I had in mid to utilise this to also act as a stand/ frame for my Router (to minimise workshop space), but eventually dispelled that notion when I realised my 'portable benchtop' router was a little heavier than expected!

    On that note, if anyone has got a good design for attaching raising/lowering wheels so I can move both router and bench around, I would be most appreciative. I have designed wheels utilising an inner and outer SHS, with a bolt, Contemplated inserting springs which can carry the load when bolt released, with springs compressed (and wheels up) when bolts tightened? Not sure how practical it will be?


    My learnings along the way

    I thought I'd share my experience for any new builders as I enter my Build logs. To date:

    1. There are no shortcuts. Be prepared to invest considerable time reading through this and other forums to get a good handle on many of the issues you will encouter/ must consider.

    2. Investment in design is key. I have found a CAD program invaluable in testing my ideas, redesigning, redrawing. Doing simple things such as determining the size of the frame to accommodate designed working bed areas is so much easier when using a CAD package. To view the extents of my working axes, to date, I have simply copied and dragged critical drawn components off to the other extreme of the axis and see if it fits, or if I have overlooked a component, such as provision for limit switches, ballscrew nut hits bearing support, etc. This should be backed up by math, but CAD is an extremely useful tool. I am using Rhino and although a big earning curve (as any CAD program), it has been worth my investment /perseverance, particularly for future capability in kinetic design needs.

    3. Where possible, incorporate manufacturers drawn CAD components into your CAD drawing. They are often readily available, they save work and the dimensions are there for incorporation into your design.

    4. Don't order materials before your design is near complete. If I ordered materials before now, I would have incurred much cost and great waste. I am about to order steel, but still wait for the valued input of the forum members to see what may need to be tweaked.

    5. Don't start cutting materials until you have many of the drive components to practically consider against your concept design. I have read on this and other forums how manufacturers' components vary from the written material. I am yet to fully research and order the drive components. Until this is done and I have the components, I know I will probably need to tweak final dimensions of the steel work.

    6. Set realistic parameters (cutting area, accuracy, material to be cut, speed, max floor dimensions etc) regarding use of your machine as they will dictate components purchased and your ultimate design).

    7. Useful Design Tools

    Apart for a CAD package, essential to getting your design right, is an understanding of:

    1. Motor torque requirements
    2. Material (steel, aluminum) Stiffness Calculators
    3. Material Optimisation software (freeware or trial software) - assists in optimising cuts for ordering materials (plate, RHS/SHS lengths etc. I started this process manually as then Googled to learn the best approach and found useful tools which save much time and effort
    4. Spreadsheets (e.g. Excel) - I have found it invaluable to input all parts/materials and their weights to assist in calcuating mass (kgs) to determine motor torque requirements
    5. Material weights - generally readily available form steel suppliers and manufacturers of other components
    6. Cutting speeds - apart from mass, acceleration and inertia, motor torque also needs to take into account your designed cutting speed and whether under load (router) or little load (e.g. plasma). Having an understanding of the speeds generally recommended for cutting different materials will help in setting realistic parameters.

    I will ultimately package these items together to create a useful toolkit for others, rather than as I and I am sure many others have done, stumble across threads where these tools are provided or are mentioned. It would be a good idea if this forum had a central repository for useful tools, which are then voted on by the members to gauge their effectiveness. If it already exists, then my apologies, but it is not easy to find?


    Lastly, this forum is very useful, and individuals are very happy to lend a hand, but you need to do your homework first and read and search the threads.

    Although a recent member, I have already saved myself from design errors through the simple guidance of individuals such as Silyavski and Eddy and tools provided by the likes RouterCNC. Thanks guys!

    Assistance Please

    If anyone can assist with the following then I would be most grateful:


    1. Frame

    Whether 6.0mm thick SHS is sufficient to directly tap (thread) and bolt rails and bearing supports to (no nuts).


    2. Gantry


    Comments on the design


    3. Z Axis


    Comments on the design

    4. Transmission Components

    I have read much but not enough about suitability of components. Current design parameters are:

    Working Area
    X - 1,500mm, Shaft circa 1,787mm
    Y - 1,240mm, Shaft circa 1,662
    Z - up to 200mm (generally 100mm working), Shaft circa 355mm

    Weight carried by axis
    X - 158 kg
    Y - 53 kg
    Z - 49 kg

    Are these weights unusual for the size / type machine?

    Questions
    I still need to thoroughly research manufacturers' specifications (which can be rather daunting with options provided) regarding suitability, but I would appreciate learning from others regarding the suitability of the following:

    A. Ballscrews (supports to be fixed both ends)

    X & Y Axis - 20mm diam (sufficient for length?), Pitch 10mm
    Z - 16mm diam, Pitch 10mm (motor calculator show little torque difference between 5 and 10mm Pitch)I will be using belts and pulleys, but anticipate a 1:1 ratio at this stage.B. Manufacturers / suppliers

    a. Drive components - Members appear to have a preference for Hiwin. Is Hiwin as good as / cheaper than THK? Why the preference?

    b. With so many variables - Bearings/Nuts, Machining ends, etc, can anyone provide me with what they used for a similar machine t mine. It may be a good starting point for me.

    c. Motors

    Weights have changed for the worse with the Z Axis re-design (in lieu of inadequate actuator) and now heavier Y axis. I have used the motor torque calculator which recommends (at a 3 times safety margin) the following motors:

    Motor Type - considering Leadshine Steppers with encoder feedback.

    X Axis - 2 x NEMA 34, 5.5Nm (they only have a 8Nm)

    Y Axis - 1 x NEMA 34, 4Nm

    Z Axis - 1 x NEMA 23, 3.0Nm

    I have looked at cutting speeds, etc and as mentioned previously, speed is less of a consideration. I will be aiming for speeds in the order of 2,000 mm/min

    Does this sound about right? The heavier weights have bumped the motor size up to NEMA 34 for X & Y axis, which if correct, means I have to relook at brackets etc for these axes (currently drawn fro NEMA 23).

    Any alternative motor recommendations appreciated. I am considering encoder feedback (on steppers). Is this deemed very important, or an unnecessary expense if motors are sized appropriately?

    Recommended suppliers

    - Transmission
    - Pulleys and belts
    - Motors and electronics


    5. Portability

    Designs or solutions on wheels for moving workbench and router around. Will post bench drawing separately. At my 10 image limit.


    Bye from Oz for now. I look forward to all/any feedback.

    Regards



    Richard

  8. #17
    Pictures of phase I, steel workbench and welder - it looks too clean to use. Will be having a play with welder this weekend.
    Attached Thumbnails Attached Thumbnails Click image for larger version. 

Name:	Workbench.jpg 
Views:	257 
Size:	92.1 KB 
ID:	13571   Click image for larger version. 

Name:	Welder1.jpg 
Views:	276 
Size:	158.0 KB 
ID:	13570  

  9. #18
    Just going out now but will reply better later, however for the wheels design have a look at this video, at about 4:20 min. into the video it shows how to move it, I have one of those machines and the idea works great. It's two fixed wheels at the back and adjustable height feet at all other points.
    http://www.youtube.com/watch?v=J2ry1DxNFA0
    Last edited by EddyCurrent; 09-10-2014 at 07:56 PM.
    Spelling mistakes are not intentional, I only seem to see them some time after I've posted

  10. re bolting stuff. Generally bolt dia less 1 to 2 x pitch. So for M8x1mm pitch thread, 6 - 7mm is typical, though for light loads 4 - 5mm might be adequate. A simple rule is to look at the thickness of the standard nut : a useful resource is www.roymech.co.UK and specifically for this issue: http://www.roymech.co.uk/Useful_Tabl...Hex_Screws.htm
    Last edited by irving2008; 10-10-2014 at 06:23 AM.

  11. The Following User Says Thank You to irving2008 For This Useful Post:


  12. #20
    To maximise the strength of the bolt in steel you need to go in 1 1/2 times the diameter , m8 = 12mm. The bolt will snap before the thread pulls out
    In ally it's 2x the diameter

    But on a rail your forces are more on the shear , so going in the diameter of the bolt would be ideal

  13. The Following User Says Thank You to Blackrat For This Useful Post:


Page 2 of 3 FirstFirst 123 LastLast

Similar Threads

  1. My plasma build 8 x 4
    By drumsticksplinter in forum DIY Plasma Build Logs
    Replies: 7
    Last Post: 01-02-2014, 10:39 PM
  2. my plasma build mk2
    By oadamo in forum DIY Plasma Build Logs
    Replies: 8
    Last Post: 08-06-2011, 10:25 PM
  3. my first plasma build
    By oadamo in forum DIY Plasma Build Logs
    Replies: 11
    Last Post: 31-12-2010, 07:15 PM
  4. My first Plasma build
    By D-man in forum Plasma Table Machines
    Replies: 4
    Last Post: 10-07-2010, 02:27 AM
  5. plasma build log - sort of
    By mike techserv in forum DIY Plasma Build Logs
    Replies: 0
    Last Post: 23-04-2009, 10:15 AM

Bookmarks

Posting Permissions

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