Australia Designing new Router called Maximus - Page 2


Page 2 of 23 FirstFirst 1234512 ... LastLast
Results 21 to 40 of 442

Thread: Designing new Router called Maximus

  1. #21
    Member peteeng's Avatar
    Join Date
    Jul 2018
    Location
    dum dum
    Posts
    6248
    Downloads
    0
    Uploads
    0

    Default Re: Designing new Router called Maximus

    Hi All - I increased the Z Axis plate thickness from 6mm to 8mm and reran the numbers. The deflection dropped from 0.47mm to 0.26mm which is worthwhile. See attached. I'll review the corrugated design properly next design round. Round 1 is too determine the main structural features of the machine and that the machine has the DNA or skeleton so to speak to achieve the desired performance.

    Since there is a lot of talk about vibration in the forum I ran the eigenvalue solver and determined the first three vibration modes. Would like to put up videos but it's a hassle editing them and converting them to fit here so here's some stills. The first mode is the gantry lozenging evenly all along itself. I intend to put some shear webs in so this will solve that vib. The next vib is a combined lozenge and twist, again the webs will push these up the scale. The model so far is stainless steel so if I did the parts in aluminium the deflection would be 3x the figures quoted. So far its the Z Axis plate that is the main contributor to deflection and since I'm going to make it a bit longer I need to make it stiffer!! I've looked around at a few routers here and 25mm aluminium plate is a common size for such a part. So I've done some comparison calcs against this plate. The 207mm wide SS corrugated plate is the same rigidity as a 25mm thick aluminium plate so I'm happy with that. It's also 76% the weight of the 25mm thick Al plate so I'm very happy with that. So the 8mm will be even stiffer and lighter but less margin. This is worthwhile to reduce the stack of the system so far. I'd like to stay with 6mm plate so maybe I can bend it deeper to make it stiffer. This points at using 20mm bearings so I have more depth to play with. Better get the X Axis going then can back to these decisions.

    I'm also thinking about a metal bench vs the timber bench. Pros and cons....Peter

    Attached Thumbnails Attached Thumbnails Designing new Router called Maximus-026mm-jpg   Designing new Router called Maximus-047mm-jpg   Designing new Router called Maximus-211htz-jpg   Designing new Router called Maximus-166htz-jpg  

    Designing new Router called Maximus-z-plate-jpg  


  2. #22
    Member peteeng's Avatar
    Join Date
    Jul 2018
    Location
    dum dum
    Posts
    6248
    Downloads
    0
    Uploads
    0

    Default Re: Designing new Router called Maximus

    last one for today better get on with my real work. I added the X bearings and bits and ran the simulation. Looking good. So now to think it through again and detail the parts a bit better. Add some internal shear webs to the gantry and make the Z plate much stiffer. I used the mini z plates for the X axis as they were there. But they do need to be longer and better braced in Round 2.Probably drop the nominal design load to 50kgf as well. Cheers Peter S

    Attached Thumbnails Attached Thumbnails Designing new Router called Maximus-loz-023mm-jpg   Designing new Router called Maximus-cant-047mm-jpg  


  3. #23
    Member peteeng's Avatar
    Join Date
    Jul 2018
    Location
    dum dum
    Posts
    6248
    Downloads
    0
    Uploads
    0

    Default Re: Designing new Router called Maximus

    Decided to bend the Z Axis plate outward vs inward and it doubled the stiffness ie halved the deflection and it's still the 6mm plate and 100kgf load. Deflection dropped from 0.47mm to 0.21mm and cost and weight will be the same. Good direction now. If I use 8mm it will be even stiffer so maybe that's the answer. Excellent...

    I think the basic DNA of the machine has been established. Onto Round 2...Peter

    Attached Thumbnails Attached Thumbnails Designing new Router called Maximus-assm-021mm-jpg   Designing new Router called Maximus-max-1-jpg   Designing new Router called Maximus-max-2-jpg   Designing new Router called Maximus-max-3-jpg  

    Designing new Router called Maximus-max-4-jpg  


  4. #24
    Member peteeng's Avatar
    Join Date
    Jul 2018
    Location
    dum dum
    Posts
    6248
    Downloads
    0
    Uploads
    0

    Default Re: Designing new Router called Maximus

    Back onto the Gantry Assembly performance. I made the outside 25mm flanges 90deg and re-ran the solver. This improved the stiffness. But the deflection was still largish. I also extended the Z Axis plate 50mm which didn't help. This gives me 300mm travel. 250mm is fine but I'll look at that closer in Round 2. Now the flanges are 90deg I can make them any height I want so 50mm it is. I then ran the solver so the Z Axis plate was "rigid" and "steel". This allows me to look at the assembly and decide how sensitive the plate is in the structure. Turns out the assembly deflects 0.24mm when the plate is rigid and 0.32mm when steel. So I think they are close enough to stop the chase for now. Maybe 8mm is the go and were done or maybe 75mm flanges. That can be in Round 2. Very happy so far.

    I added shear webs at the ends and the deflection dropped from 0.49mm to 0.18mm so that's it I think. Now to Round 2 and motion elements.

    Review
    1) A 50mm deep 6mm thick corrugated Z Axis plate shall be used. This is stiffer than a 25mm solid aluminium plate
    2) Shear webs have improved global deflection heaps
    3) The gantry and Z axis model weighs 48kg so far
    4) 15mm bearings seem to be OK so far
    5) Can convert larger parts to Al to save some weight if it does not impact on stiffness

    Actions
    1) Decide on working envelope and adjust gantry width to suit
    2) Start from spindle clamp and refine parts. Add bolt, lightening and access holes as needed
    3) Try to make it a little smaller
    4) Decide metal or timber bench
    5) Start adding motion elements...

    Stay tuned for Round 2 - Peter

    Attached Thumbnails Attached Thumbnails Designing new Router called Maximus-50mm-flange-rigid-jpg   Designing new Router called Maximus-6mm-steel-jpg   Designing new Router called Maximus-shear-web-jpg  
    Last edited by peteeng; 04-02-2019 at 06:34 AM.


  5. #25
    Member
    Join Date
    Jan 2008
    Location
    Australia
    Posts
    1523
    Downloads
    2
    Uploads
    0

    Default Re: Designing new Router called Maximus

    Much better. I actually wrote part of a post earlier today suggesting you use perpendicular stiffeners (but on the go and didn't have time to finish it).

    I recommend you use an 80mm spindle. Besides being superior to a standard wood router, out makes it much easier to design a stiff Z carriage. You can make the whole thing into a box section pretty much.

    I think your actual gantry shape is overcomplicated. How are you going to fabricate it? Cost to do so? Why not just use a big bit of standard square hollow section (with bracing webs at regular intervals inside and end caps)?



  6. #26
    Member routalot's Avatar
    Join Date
    Dec 2003
    Posts
    1206
    Downloads
    0
    Uploads
    0

    Default Re: Designing new Router called Maximus

    I can see an area that might cause issues in the future and it often occurs with machines that are intended to have a fairly high gantry for Z axis clearance.You may struggle to get all jobs done that will fit under the gantry as the corners of the spindle mounting plate may foul the job-it really spoils your day when that happens.Could I suggest working out the angle of the sides of the "valley" that you will be able to work within? Its worth doing this in both directions as they are likely to be different angles. I hope you don't mind that I'm from the other side of the world and wandering into your area!It is an area of interest for me as I have spent lots of time with a plane and spokeshave shaping rudders and now I know that I can get a machine to move under it's own power a suitably sized machine beckons.



  7. #27
    Member peteeng's Avatar
    Join Date
    Jul 2018
    Location
    dum dum
    Posts
    6248
    Downloads
    0
    Uploads
    0

    Default Re: Designing new Router called Maximus

    hi Pippin - Having it laser cut and bent is straightforward. My machines are designed for production kits to sell to Makers. If you use SHS you have to drill and tap many holes and generally you have to weld it and it distorts. If you do it the maximus way it's assembled easily with nuts. A nut is much cheaper then a threaded hole. You say use bracing webs inside. This is impossible to do in an SHS and being thick walled it is also unnecessary. For this machine a 250SHS 6mm thick would be needed and that would be way too heavy for this machine as intended.

    A note on design. The gantry rail spacing is a "found" spacing. It is the distance required via the vertical stack height of the Z Axis assembly. It has not been chosen or driven by a SHS size. So by bending metal sheet it is the "correct" size for the job. Also I can vary the front and back sheet thickness saving weight yet acheiving the performance.

    An 80mm spindle has always been the best solution for this size machine if you can afford it. I think I'll bite the bullet and design it as a full sheet machine. But the people that I talk to and aim these routers/lasers/plasma/food extruder etc at, are always looking for a cheaper solution and generally start with a Mikata or a Ryobi etc. Scoot my 1/2 sheet machine has a Makita barrel router that I have thrashed for 2 years and it's still cutting fine, just have to wear muffs. Thanks for participating Peter

    At the end of the Maximus design process (maybe Round 4 end) this machine will be costed completely and interests in buying determined. Typically I have to get three or 4 sets of parts made to get to minimum order requirements for a production run. So if anyone wants a full sheet cnc router kit its on its way, stay tuned.

    Last edited by peteeng; 04-02-2019 at 04:14 PM.


  8. #28
    Member peteeng's Avatar
    Join Date
    Jul 2018
    Location
    dum dum
    Posts
    6248
    Downloads
    0
    Uploads
    0

    Default Re: Designing new Router called Maximus

    Hi Routalot. Welcome down under !! I suppose we are -Z. Yes the square bottom does hit things occasionally. My 1/2 sheet Scoot has a similar blunt end and it does catch stuff (including lots of dust) if you don't think ahead. Maximus is being developed as a kit, so stay tuned for final costing. I have intended to a) have the spindle stick down quite a bit or b) angle the edges of the Z Axis plate as much as possible. We are in Round 1 of the design and this round gets the concepts, geometry and limitations understood. So your comment is noted and will be addressed in time. Also to solve this issue, you can use longer tooling. But I have not found a supplier yet. There are lots of videos of robots and 5 axis machines with long tooling. I may have to make tool extensions for deep cavities. I had a client some years ago that had a Z500mm 3 axis mill and they made 400mm extensions for their tooling but did not do it quite right. I was there when a tool decided to whip and break not pretty!!

    Just over 2.5 years ago I was hand shaping a daggerboard for a trimaran and said, that's the last one I'm going to do!! If you search for Scoot cnc carbon-works you will find its blog.

    Peter

    3-04-19 added a chamfered bottom just for you Routalot. Its 50degs. Its not quite doable on a laser as built here but the feature is now captured and next time around I'll do it so it is laserable (if that's a word)

    Attached Thumbnails Attached Thumbnails Designing new Router called Maximus-plate-1-jpg   Designing new Router called Maximus-plate-2-jpg  
    Last edited by peteeng; 04-03-2019 at 04:32 AM.


  9. #29
    Member peteeng's Avatar
    Join Date
    Jul 2018
    Location
    dum dum
    Posts
    6248
    Downloads
    0
    Uploads
    0

    Default Re: Designing new Router called Maximus

    Good Day out there - Been looking at the parts as they stand and looking for any improvement of stiffness or cost part by part. Made a tweak here or there and some mental notes for Round 2. I also did some weight reduction efforts. The current all steel assm weighs 48.6kg and deflects 0.18mm under the nominal 100kgf load. I converted the front plate of the gantry then the rear plate to aluminium. These are the biggest parts in the assm. This brought the weight down to 30.6kg, an 18kg saving. The deflection grew to 0.32mm which I'll live with at present. I can make the back plate thicker and add the internal shear webs to improve this. 18kg extra is a lot to push around so we'll go with the aluminium at the mo.

    The current Z travel is 205mm which is the same as Scoot. Wanted 250mm but this will depend on the final stiffness of the machine. Haven't had to need more Z on Scoot in 2 years so 200mm maybe OK. The only time I had trouble was after a Z Axis rebuild and I put the router bracket on upside down. After that my gas strut that I use to hold the Z up kept bottoming and this never happened before. A quick look found the problem and it was fixed quick sticks.

    Peter

    The front plate in Al (5mmt) weighs 4.41kg and the rear in Al weighs 4.7kg (3mm t). In steel they would be 7800/2700= 2.89x heavier

    Attached Thumbnails Attached Thumbnails Designing new Router called Maximus-maxi-jpg  


  10. #30
    Member hanermo's Avatar
    Join Date
    Aug 2004
    Location
    barcelona
    Posts
    780
    Downloads
    0
    Uploads
    0

    Default Re: Designing new Router called Maximus

    Fwiw..
    sorry to be so negative.

    For cutting alu efficiently you need about 10x more stiffness.
    Look at the haas router for inspiration, it is the same size, grx series iirc, now discontinued.

    Belts wont work.
    Example, experience:
    I used 30 mm wide HTD-8 belts, and they are useless for rigidity.
    My spindle motor mount is 60 kg in mass in steel, 30 mm thick steel posts. 20 mm thick tool steel plates.
    The belts are simply much too bendy.

    Forget loads/capacity from linear guides.
    You want about 50x the capacity vs your working load.
    Every lathe/mill in the world is like this.

    My mill has 50 tons capacity in z, 35 mm linears, for a working load about 1 ton.



  11. #31
    Member peteeng's Avatar
    Join Date
    Jul 2018
    Location
    dum dum
    Posts
    6248
    Downloads
    0
    Uploads
    0

    Default Re: Designing new Router called Maximus

    Hi Hanermo - negative input is as good as positive here - I am designing a small kit router not a mill. If I wanted to build a mill I would buy a second hand manual mill CNC convert it or I would contract the parts that I wanted to a machinist with a big mill like I do now. I totally agree a Mill is a very different animal. BUT cutting aluminium is possible with a Maker grade router but not in the production oriented utility sense like you are in. That's why mills exist. For a hobbyist, Maker, DIYer cutting the occasional aluminium part on their router is a plus. I get asked all the time can this be done and the answer is yes with some provisions, limitations and understanding. I agree belts are not as rigid as screws or racks but the price point I'm aiming at can't bear the cost of these unless I find a cheaper supplier. This machine is not intended as a commercial production oriented machine like a HAAS or your several tonne machine, it's a <$4500AUD build so comparing it to these is a little unfair. I'd prefer to use helical racks but they are 4x the price of the belts. Everything in this design has to be honed down to the best price/performance point ratioed to the linear bearings cost. Otherwise we are in construction extrusions, belts and wheels territory so may as well buy a shapeoko or Ox kit or similar. Hope to keep you entertained... Peter



  12. #32
    Member peteeng's Avatar
    Join Date
    Jul 2018
    Location
    dum dum
    Posts
    6248
    Downloads
    0
    Uploads
    0

    Default Re: Designing new Router called Maximus

    I'm back at the start deciding if it's time to rebuild the CAD model. One initial design premise was to gain 20mm in space by making an odd shaped Z Axis plate. I'm thinking is this worth it? I'd like to have the spindle line up with the gantry bearings and this helps. But It also creates 60mm in extra width with the bends and limits the real estate to bolt on spindles and things to the gutter width. I think the other elements are pretty good at this stage. So the mantra "stiffness wins" needs to be applied to the gutter plate. So I need to model a straight one to compare it to. So perhaps a flat aluminium plate, a simple bent plate and the gutter plate to make the decision. Peter

    Attached Thumbnails Attached Thumbnails Designing new Router called Maximus-bearing-jpg   Designing new Router called Maximus-gutter-jpg  


  13. #33
    Member peteeng's Avatar
    Join Date
    Jul 2018
    Location
    dum dum
    Posts
    6248
    Downloads
    0
    Uploads
    0

    Default Re: Designing new Router called Maximus

    Hi All - I modelled the three options. 1) the gutter design weighs 7.7kg in 8mm SS and deflects 0.07mm under 100kgf 2) I started with a 25mm thick al plate but it deflected heaps so went to a 30mm plate. It deflects 0.11mm and weighs 6.2kg 3) The straight 8mm SS plate of same depth as the gutter plate 70mm deflects the same as the gutter plate 0.07mm and weighs 8.9kg. So now I have to weigh up the pros and cons of the gutter vs the plain design. More thinking. I think I'll use the plain design and move onto Round 2. Can come back to the gutter design if needed. Which is detailing the parts and adding the motion parts. Peter

    Attached Thumbnails Attached Thumbnails Designing new Router called Maximus-comparison-jpg   Designing new Router called Maximus-capture-jpg  
    Last edited by peteeng; 04-07-2019 at 07:49 AM.


  14. #34
    Member peteeng's Avatar
    Join Date
    Jul 2018
    Location
    dum dum
    Posts
    6248
    Downloads
    0
    Uploads
    0

    Default Re: Designing new Router called Maximus

    For some time I have been wondering about how to model the compliance of the linear bearings. The thesis link has given me the correct path. Usually a bearing is modelled in steel and the connection stiffness is therefore overestimated. By modeling the bearing as a steel block and a compliant block at the contact we can get this right if we know the stiffness of the bearings. Turns out that the compliant bit needs to be about 6GPa. So in Round 2 I'll model the bearings as a two piece unit so I can include this. Manufacturers of bearings publish their stiffness so the model can be tuned to the correct compliance. That thesis is a good read.... Peter

    Attached Thumbnails Attached Thumbnails Designing new Router called Maximus-bearing-jpg  


  15. #35
    Member
    Join Date
    Aug 2014
    Posts
    232
    Downloads
    4
    Uploads
    0

    Default Re: Designing new Router called Maximus

    Interesting work, thanks. Are your deflection figures an average of the deflection in all directions ? Also, can your software simulate resonance in the Z plate ? I wonder if the solid ally plate might damp it better.




  16. #36
    Member peteeng's Avatar
    Join Date
    Jul 2018
    Location
    dum dum
    Posts
    6248
    Downloads
    0
    Uploads
    0

    Default Re: Designing new Router called Maximus

    Hi Zorbit - The deflection figure is the maximum distance from its starting point. I can image it in X,Y or Z if required. I can figure the modal frequencies of the assemblies. I just tried to solve the 3 assemblies in one model but there are too many directions of freedom for the solver to handle so have had to solve it in two goes.

    Summary
    The Gutter design first vibrates at 319htz, this is a torsional vib. Then it vibs at 553htz in a pendulum sort of way. The aluminium vibes as a cantilever at 541htz (same mode shape as gutter at 553htz). The plain design vibs at 290htz then 435htz. I may stiffen the plain design with a web or use the gutter design not sure yet. Next design round is to add more detail to the parts like bolt holes and start adding the motion components.

    There is a misconception out there that steel increases vibration somehow. This would fall into the "free lunch" category of which there is none. Steel is damp just not as damp as other materials. From the chart you can see that steel is 60% the dampness of cast iron and about 1/10 as damp as granite. The modal analysis calculates the free vibration frequency. If the material had no dampness it would vibrate at this freq forever once excited. But since materials have dampness this vibration settles out fast unless it is reinforced by some force input such as a tool load or inertial load. In another thread damping is discussed and I have put forth the idea to use laminated metals with thin rubber interlayers (or epoxy them together) to create a damp material. Search for damping to follow that up if interested. Laminated metals are used in aircraft for various reasons. Look up GLARE an aluminum laminate that Airbus makes its fuselages out of. Peter

    Al vs steel damping - From the chart you can see that Al is less damp then steel. This is because Al is locked up internally more than steel is. Steel has a cubic grain structure which has lots of slip planes for movement. Look at a brick wall and you can see how all the horizontal joints could slip easily. This is why steel forms so well. Al has a hexagonal grain structure so has very few slip planes. For it to slip it has to climb up and down the hexs so it's quite stiff internally. So steel is better from a material damping point of view.

    Attached Thumbnails Attached Thumbnails Designing new Router called Maximus-plain-290htz-jpg   Designing new Router called Maximus-aluminium-541htz-jpg   Designing new Router called Maximus-gutter-torsion-jpg   Designing new Router called Maximus-damping-jpg  

    Designing new Router called Maximus-plain-435htz-jpg  
    Last edited by peteeng; 04-09-2019 at 07:44 AM.


  17. #37
    Member davida1234's Avatar
    Join Date
    Oct 2009
    Location
    United States
    Posts
    733
    Downloads
    0
    Uploads
    0

    Default Re: Designing new Router called Maximus

    Hi Peter,

    IMO, this is going to be your biggest challenge regardless of the Z-slide shape.

    Designing new Router called Maximus-capture-jpg



  18. #38
    Member
    Join Date
    Aug 2014
    Posts
    232
    Downloads
    4
    Uploads
    0

    Default Re: Designing new Router called Maximus

    Excellent post, I can see I will learn a lot from this, thanks.



  19. #39
    Member peteeng's Avatar
    Join Date
    Jul 2018
    Location
    dum dum
    Posts
    6248
    Downloads
    0
    Uploads
    0

    Default Re: Designing new Router called Maximus

    Quote Originally Posted by davida1234 View Post
    Hi Peter,

    IMO, this is going to be your biggest challenge regardless of the Z-slide shape.

    Designing new Router called Maximus-capture-jpg
    I presume you mean the small distance between the bearings? If so there is no recommended spacing for these. All you can do is stay within the manufacturers recommended loads. Obviously the further apart the better but there are limits on this to the size of the machine being designed. So this distance is considered in two ways 1) That the bearings have a satisfactory static load and life loading 2) it provides maximum stability in the geometry available. ie they are as far apart as possible. In routers and mills it is recommended to use heavy preloaded bearings. My experience with std light preloaded bearings is that some time down the track they turn into clearance bearings. This is with medium quality and priced units. Peter



  20. #40
    Member routalot's Avatar
    Join Date
    Dec 2003
    Posts
    1206
    Downloads
    0
    Uploads
    0

    Default Re: Designing new Router called Maximus

    Its a very interesting thread so far and the depth of engineering continues to impress me greatly.On the topic of resonance,will the ballscrews and their mountings have a differing effect on the resonance as the travel varies?



Page 2 of 23 FirstFirst 1234512 ... 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

Designing new Router called Maximus

Designing new Router called Maximus