Newbie Why aren't portal mills more popular? - Page 2


Page 2 of 3 FirstFirst 123 LastLast
Results 21 to 40 of 42

Thread: Why aren't portal mills more popular?

  1. #21
    Member JoanTheSpark's Avatar
    Join Date
    Jun 2016
    Location
    Australia
    Posts
    26
    Downloads
    0
    Uploads
    0

    Default Re: Why aren't portal mills more popular?

    Quote Originally Posted by mactec54 View Post
    ...mostly used to straighten the rail, plus align the rail parallel, the rails are not straight so you need some way to straighten them.
    OK, thanks for that info.
    So if I get me some sort of long, very straight edge (bolt on) I can potentially omit the milled reference edges/pockets, which should make milling the flats easier/cheaper.
    Once the rail bolts are tightened, the reference (theoretically) is not needed anymore, right?
    Parallelism should be achievable by copying the 'master' to the 'slave' via the gantry/sled.

    On another note, I stumbled across this machine today - Ingersoll High Speed Eagle V9
    Their gantry is very tall, to counter bending I guess, a very goo idea IMHO - I was coming from a SHS 150x150 section, so didn't see that in my design.
    And they got the rails+drives for the z-ram mounted on the y-sled, so they won't get down into the swarf area.
    The only downside to this is, that the moments caused by the tool on the z-ram will be permanently at the worst for the z-carriages (rigidity, precision).



    I will redesign mine then to copy that - already increased the gantry box edge dimensions to 200x200 and added pneumatic cylinders to counter the z-ram weight (I estimate they can counter ~100kg at 0.5MPa, MAL40x350).
    This way I can get rid of the triangular plates at the back of the sled and the 3rd rail there as well.



    Attached Thumbnails Attached Thumbnails Why aren't portal mills more popular?-portalmil_y-z-node-2-jpg   Why aren't portal mills more popular?-portalmil_hsev9-jpg  
    Last edited by JoanTheSpark; 11-02-2017 at 08:02 AM.


  2. #22
    Member mactec54's Avatar
    Join Date
    Jan 2005
    Location
    USA
    Posts
    15362
    Downloads
    0
    Uploads
    0

    Default Re: Why aren't portal mills more popular?

    Quote Originally Posted by JoanTheSpark View Post
    OK, thanks for that info.
    So if I get me some sort of long, very straight edge (bolt on) I can potentially omit the milled reference edges/pockets, which should make milling the flats easier/cheaper.
    Once the rail bolts are tightened, the reference (theoretically) is not needed anymore, right?
    Parallelism should be achievable by copying the 'master' to the 'slave' via the gantry/sled.
    And if you ever crash the machine, the rails will move out of position, without the rail guide in place, and Doweled in place, bolting by it's self would not be good enough

    Quote Originally Posted by JoanTheSpark View Post
    The only downside to this is, that the moments caused by the tool on the z-ram will be permanently at the worst for the z-carriages (rigidity, precision)..
    That is correct, your design is better, but harder to build, note they are using 2 Ballscrews on the Z Axes, main carriage

    Here is a PDF that may help with some parts of your build, it's quite old information, but plenty to be gained from it

    Attached Thumbnails Attached Thumbnails Why aren't portal mills more popular?-principles-rapid-machine-design-pdf  
    Mactec54


  3. #23
    Member JoanTheSpark's Avatar
    Join Date
    Jun 2016
    Location
    Australia
    Posts
    26
    Downloads
    0
    Uploads
    0

    Default Re: Why aren't portal mills more popular?

    Quote Originally Posted by mactec54 View Post
    And if you ever crash the machine, the rails will move out of position, without the rail guide in place, and Doweled in place, bolting by it's self would not be good enough
    Hm (naively asking) - if a crash like this happened, wouldn't the alignment and condition of the rails/carriages be questionable anyway?
    If the bolts are able to 'give' a little it might just save the rails vs. if the rails stay in place the balls will indent in them/be permanently deformed.

    While we're on the subject of crashing - has anyone used/done anything with pre-job/real-time 3D scanning of the working envelope, that feeds absolute maximums into the controller, so it is able to create (software) no-go-zones/limits it will not enter?
    So, before one puts the workpiece into the machine, the machine runs along a safe 2D plane and scans the envelope to create that 'maximum limits data'. This would account for changes to the table/vice setup etc.
    There is software out there that does 3D scanning from photos, called photogrammetry.
    Most modern (but might still be to expensive right now) would be LIDAR, but standard cameras should work.

    Quote Originally Posted by mactec54 View Post
    That is correct, your design is better, but harder to build, note they are using 2 Ballscrews on the Z Axes, main carriage
    Hm, you might be right - this counters the moments in x-direction (along gantry) better than a single screw.
    I'm seriously considering that now. Can use two smaller ballscrews (compared to the others) driven by the same servo even.
    Good catch.

    Making the gantry taller will probably also allow me to put some concrete blocks in there more easily post milling vs. the 200x200 gantry concept, if I'm to add ribs, etc. to it.

    Quote Originally Posted by mactec54 View Post
    Here is a PDF that may help with some parts of your build, it's quite old information, but plenty to be gained from it
    Thanks for the pdf, very informative.

    Seems I'm on the right track then.. I was planning to glue pre-fab concrete blocks (~10kg each, 290x190x90 mm3, 15 per panel) into the structure with PU sealant, might even try to get my hands on some that is expanding a little before curing. PU glue does that when one adds water, don't think sealant does. I glued concrete blocks to each other with PU glue before - this connection is like a healed broken bone - if it ever cracks, it cracks somewhere else ;-)
    Anyhow, touching all steel parts that might resonate via dampening PU sealant and the concrete mass attached to that, I should get similar results to the ones in the paper.



    As per CAD tool the wight of the substructure so far is at ~1,200 kgs (still needs crossing struts in the boxed area under the table).



    It consists out of 4 identical 'grates', made from 100x100x3 SHS - weight per CAD tool ~76 kg each.
    Welding those together carefully should result in pretty straight/flat/rectangular pieces to assemble the machine out of.
    I've done this with table frames before and the unfortunate bending only started once I begun to attach 'out-of-plane' elements like legs to this. Up unto that point they were flat/straight.



    Oh, and the pdf also has this (page 83):
    3.4.1 Reference Edge
    Using a straight edge it is possible to replicate the reference edge used to align machine elements such as linear rails. This eliminates the need for machining this particular feature, cutting down fabrication time and the need for machine tools big enough to hold the structure whose reference edge is to be created.
    So if I can avoid crashing (or live with the messed up alignment when it happens) I should be good to do it that way.

    Attached Thumbnails Attached Thumbnails Why aren't portal mills more popular?-portalmil_concreteblock-jpg   Why aren't portal mills more popular?-portalmill_baseframe-jpg   Why aren't portal mills more popular?-portalmill_grate-jpg  
    Last edited by JoanTheSpark; 11-03-2017 at 12:43 AM.


  4. #24
    Member mactec54's Avatar
    Join Date
    Jan 2005
    Location
    USA
    Posts
    15362
    Downloads
    0
    Uploads
    0

    Default Re: Why aren't portal mills more popular?

    Quote Originally Posted by JoanTheSpark View Post
    Hm (naively asking) - if a crash like this happened, wouldn't the alignment and condition of the rails/carriages be questionable anyway?
    If the bolts are able to 'give' a little it might just save the rails vs. if the rails stay in place the balls will indent in them/be permanently deformed.

    While we're on the subject of crashing - has anyone used/done anything with pre-job/real-time 3D scanning of the working envelope, that feeds absolute maximums into the controller, so it is able to create (software) no-go-zones/limits it will not enter?
    So, before one puts the workpiece into the machine, the machine runs along a safe 2D plane and scans the envelope to create that 'maximum limits data'. This would account for changes to the table/vice setup .
    Most good Machine control software, have Soft Limits, this is where you can control the maximum machine movements, so when you start up your machine, each axes is Homed, and then what is called soft limits will be activate, a crash does not always mean hitting the max travel of an axes, Just breaking a cutter or running the spindle into the work piece can cause misalignment

    Quote Originally Posted by JoanTheSpark View Post
    Oh, and the pdf also has this (page 83):

    So if I can avoid crashing (or live with the messed up alignment when it happens) I should be good to do it that way.
    The PDF is dated but most of this information still applies, Yes you can do it like you planed, the whole exercise in the PDF was to do it as cheap as possible, normally the Rails and Bearings are not damaged when you crash a machine, I pocket the Bearings and the Rails, so no chance of any misalignment, with that part

    Mactec54


  5. #25
    Member
    Join Date
    Jul 2009
    Location
    NL
    Posts
    419
    Downloads
    0
    Uploads
    0

    Default Re: Why aren't portal mills more popular?

    Quote Originally Posted by JoanTheSpark View Post
    With this design you introduce some overcomplication.

    To get the z-axis go up and down straight you need to get the gantry square to the bed and before that you need to get the cariage vor the z-axis square to the gantry.
    On top of that you "need" the third rail.
    Similarly you put rails on the side of the z-axis instead of behind it.

    In the thread in my signature you can see how much simpler it can be.
    That design was handed to me by a gentleman who took me under his wing and got me set up with a router that according to him should be able to mill steel even though it is made of aluminium.

    And you might be able to build it with just a drill press and a table saw...
    Well, except for the sides of your "box" that is.

    Sven http://www.cnczone.com/forums/diy-cnc-router-table-machines/320812-aluminium-1250x1250x250-router.html


  6. #26
    Member JoanTheSpark's Avatar
    Join Date
    Jun 2016
    Location
    Australia
    Posts
    26
    Downloads
    0
    Uploads
    0

    Default Re: Why aren't portal mills more popular?

    Quote Originally Posted by CaptainVee View Post
    With this design you introduce some overcomplication.

    To get the z-axis go up and down straight you need to get the gantry square to the bed and before that you need to get the cariage vor the z-axis square to the gantry.
    On top of that you "need" the third rail.
    Similarly you put rails on the side of the z-axis instead of behind it.

    In the thread in my signature you can see how much simpler it can be.
    That design was handed to me by a gentleman who took me under his wing and got me set up with a router that according to him should be able to mill steel even though it is made of aluminium.

    And you might be able to build it with just a drill press and a table saw...
    Well, except for the sides of your "box" that is.
    Ah yes, bonding aluminium to avoid welding while getting the same rigidness.
    You might enjoy a series on youtube:



    I already moved towards the Ingersoll design, got rid of the 3rd x-rail and might even put the x-rails on the gantry on the front of it, not sure yet as this means the gantry needs to be milled perpendicular for the y-rail carriages vs the x-rails - sounds more complicated to me and prone to error.
    If I keep the x-rails in their current position the gantry has to be milled parallel flat on the bottom and top only, which is easier to achieve IMHO.

    The y-rail flats on the walls I have to grind flat myself - no access to such a big machine here afaik that I can afford to have this done on - also the reason I won't have machined reference edges there. Parallelism will be achieved during assembly via adjustable legs.

    As for the bed - this will be milled flat by the CNC itself. I know about the problems with mild steel and internal stresses that come out once you asymmetrically remove material, I hope it's not that bad when I do it to one of three surfaces that are sandwiched together by welding and not touching the bed again afterwards (it stays bolted down, last step of making the machine).



    Attached Thumbnails Attached Thumbnails Why aren't portal mills more popular?-portalmill_towardsingersoll-jpg  


  7. #27
    Member
    Join Date
    May 2011
    Location
    Canada
    Posts
    790
    Downloads
    0
    Uploads
    0

    Default Re: Why aren't portal mills more popular?

    Quote Originally Posted by JoanTheSpark View Post
    I'm still torn between building a single machine or making two - space available vs money vs reality.
    Make one. You will undoubtedly want to change something (fix mistakes and make improvements) in the way you do things after the 1st one, so the 2nd one is better.

    Quote Originally Posted by JoanTheSpark View Post
    I suppose there is no spindle for hobby kind of money that can cover both, otherwise I'd have come across one by now on this forum I guess.
    I don't think you've grouped those together correctly. A spindle that is good at cutting aluminum should be good for wood too. For steel, it's high torque at lower RPM's you want obviously. Most (99%) of the machines we make wouldn't be rigid enough to cut steel regardless of what spindle was used. Builds that can do it are rare. Perhaps that is also a reason you haven't come across it.
    Quote Originally Posted by JoanTheSpark View Post

    This is the example I'm mostly hanging onto for the size/power/capabilities of the machine:
    http://www.cnczone.com/forums/uncate...aluminium.html
    That's a fantastic build, one of the best ever. I'm super jealous of it. I've read some of his comments on his YouTube channel.

    This one I find very informative

    "EG is under table, steel frame is filled with EG, and bridge is also filled with EG. Machine handles steel, I even made video of milling stainless 304 steel, which I will post in short time.
    Mild steel can be machined without any problems , but I didnt use tool more than 12mm in diameter. For aluminium, I tried 50mm face mill, and it works without any problem.
    Additional spindle is BT30. I use it for steel. However this spindle will be removed, actually both of them, because I already ordered ATC BT30 spindle water cooled for up to 12k rpm, and I will use only ATC, so no more 2 spindles.
    About aluminium gantry.. It wont handle steel if its not filled with EG. Even when milling aluminium, machine was much more sensitive to parameters. Without EG, it happned that sometimes I hit parameters (mainly tool rpm) which came in resonance with machine structure, and it made ugly high pitch noises.
    Aluminium structure, steel structure.. Those are all very very bad materials to build machine bed out out of. Grey cast or EG, or UHCP are way to go.

    Sure there were some lessons learned.. Like use ballscrew with high pitch - 10mm or more. It is best if ballscrew rotates slow - less vibration, quiet and smooth operation.
    None of commercial made machines use 5mm pitch, like us hobbyst do.This is why I will change both X and Y ballscrews to 10mm pitch, and make higher ratio with pulleys.

    Then there is so important Z axis.. Aluminium or steel plate is again very bad solution. Any even half serius machine should have gray cast square struture for Z axis. Or somekind Of EG filled Z stage. Vibration dampening of Z axis is super important."

    Here's another one that has great performance:

    https://www.usinages.com/threads/cnc....63531/page-54

    You can see what kind of spindle he has. Many design iterations, so look at the end and check out his youtube channel to see videos of cutting steel.

    For my spindle, I'm going to be using a spindle that came out of a Biesse Rover. Not sure yet if it was a good idea or a mistake. It weighs near 100 lbs. Also, I still don't know what some of the pins are for and can't find any documentation. As a bonus it has an ISO 30 ATC (which I plan to use as a quick change, not auto change) and lots of power.

    First off, let me say, that I am a fan of your design and thought process. But, there are things about your current design iteration that I really do not like.

    1. The base is overly complex. Also, fewer, but much larger, square tubes would give IMO, better performance, and be less prone to weld distortion.

    2. Instead of adding weight with prefab concrete blocks, I think it would be better to fill the frame with something, epoxy granite perhaps.

    3. I don't believe that gantry will give you the best stiffness to weight ratio. I believe the piece where the rails are bolted may distort like crazy when welded in place. And there's no way you could decide to fill it with any vibration dampening material later on (because it's open).

    4. It all depends on what kind of a spindle you have as to whether the spindle body helps with the rigidity of your Z axis. For a long Z axis like you have, I think a square tube concept is better.

    What I do like

    1. You're looking at designs that work and have a great design iteration process and methodology.

    2. Ribs in the gantry.

    Of course, these are just my opinions, and I don't claim to be anyone special. I think at the end of the day you may come up with something great, perhaps better than any of us could, I just think you should explore some different ideas to have something really awesome.

    Handlewanker,

    Nice to see you here mate. Good to see you can dish it out as well as you can take it!



  8. #28
    Member JoanTheSpark's Avatar
    Join Date
    Jun 2016
    Location
    Australia
    Posts
    26
    Downloads
    0
    Uploads
    0

    Default Re: Why aren't portal mills more popular?

    Quote Originally Posted by NIC 77 View Post
    The base is overly complex. Also, fewer, but much larger, square tubes would give IMO, better performance, and be less prone to weld distortion.
    I haven't seriously stress tested this yet, but here is a simple one - 50kg centrally at the top tube with the bottom face the fixed bearing. Gives 0.12 mm lateral displacement. Now imagine two of those and the fixed area being at least half up vertically and the force being more realistic - it's good enough for what I will be doing - especially if other people manage with aluminium.



    I also don't really want to go any bigger. The largest tubes I have cut & welded were 75x75x4 mm ones for a welding table. That stuff already looks & feels sturdy as hell.

    And on complexity I wholeheartedly disagree. Anything I've seen made for steel out of tubing is way more complex and prone to welding distortions - I will stay within the 2D plane(s), as are the welding stresses. These grates will come out fine.

    Quote Originally Posted by NIC 77 View Post
    Instead of adding weight with prefab concrete blocks, I think it would be better to fill the frame with something, epoxy granite perhaps.
    1) I would need to cut holes into each and every one of them or redesign with compromises to get the EG in there
    2) a block cost me $3, for 10kg, another $5 for the PU glue
    If I ever need to dismantle it - the blocks are knocked out easily and then the thing cut up.

    From my POV it doesn't matter if you attach the dampening mass on the outside with some rubber/glue or from the inside.

    Quote Originally Posted by NIC 77 View Post
    I don't believe that gantry will give you the best stiffness to weight ratio. I believe the piece where the rails are bolted may distort like crazy when welded in place. And there's no way you could decide to fill it with any vibration dampening material later on (because it's open).
    Not finished. The image was just to show were I'm headed vs. were I came from.
    It will be enclosed and possible to put heavy mass in/on there and welding wont distort it, at least not worse than a certiflat welding table, who's principle I will adapt and make use of..



    Quote Originally Posted by NIC 77 View Post
    It all depends on what kind of a spindle you have as to whether the spindle body helps with the rigidity of your Z axis. For a long Z axis like you have, I think a square tube concept is better.
    The 200x200 SHS 'tube' for a gantry has a couple of problems:
    1) twisting moments are better taken care of by a round tube, the ribs will transfer those to it and then to the rails on the outer walls (imagine the z-ram fully down and apply a force in y-direction)
    2) I need more vertical distance of the rails for the sled and can afford some overhang at the rear - shape becomes triangular
    I would love to put the full triangle on the y-rails, but I'm stubbornly set on 1,600mm long y-rails and a 1,200mm y-envelope, so that's that.

    The spindle in the image up there is a 3kW water cooled ER20 spindle with up to 20k rpm or thereabouts.
    As I wrote earlier, I didn't come across a ATC 'ready' one, that has got a BT30/NT30/etc. adapter and is able to mill steel in that size class and in a single unit.
    There is all sorts of spindle powerheads, which are then driven by a low rpm/high torque system on a parallel axle and the ATC is bolted on (all very big and clumsy and not in one single axis). Also I don't like the pneumatic ATC, I'd rather have something electric.

    I can imagine a swappable spindle assembly. One for high rpm with ER20 collet for routing and another one for high torque with a BT30 cone and ATC for milling.

    This video on youtube was pretty interesting (imaging an outrunner BLCD on a BT30 power head), but doesn't look like he made any progress as he got himself an ATC spindle with high rpm since then :-(



    Quote Originally Posted by NIC 77 View Post
    Make one. You will undoubtedly want to change something (fix mistakes and make improvements) in the way you do things after the 1st one, so the 2nd one is better.
    I don't think I want to build a 2nd one in short succession.
    Till I'm ready for one 3D metal printing will nearly be economical for home users ;-)

    Quote Originally Posted by NIC 77 View Post
    " ..Additional spindle is BT30. I use it for steel. However this spindle will be removed, actually both of them, because I already ordered ATC BT30 spindle water cooled for up to 12k rpm, and I will use only ATC, so no more 2 spindles..."
    The question for me then is - what is the new ATC spindle with BT30 running at for steel milling if he wants to use the 50mm face mill?
    What spindle is that?

    Quote Originally Posted by NIC 77 View Post
    " ..Sure there were some lessons learned.. Like use ballscrew with high pitch - 10mm or more. It is best if ballscrew rotates slow - less vibration, quiet and smooth operation.
    None of commercial made machines use 5mm pitch, like us hobbyst do.This is why I will change both X and Y ballscrews to 10mm pitch, and make higher ratio with pulleys..."
    Very good catch. Will take that to heart.

    Quote Originally Posted by NIC 77 View Post
    " ..Then there is so important Z axis.. Aluminium or steel plate is again very bad solution. Any even half serius machine should have gray cast square struture for Z axis. Or somekind Of EG filled Z stage. Vibration dampening of Z axis is super important..."
    Not there yet. Will keep it in mind though.

    Quote Originally Posted by NIC 77 View Post
    Here's another one that has great performance:
    https://www.usinages.com/threads/cnc....63531/page-54
    You can see what kind of spindle he has. Many design iterations, so look at the end and check out his youtube channel to see videos of cutting steel.
    Ha, came across that, but after 6-7 pages it was still a complex tubular mess, so didn't bother to go deeper in that thread.
    His final build has a lot of welds in all directions = lot's of distortion, I try my best to avoid that.

    Quote Originally Posted by NIC 77 View Post
    For my spindle, I'm going to be using a spindle that came out of a Biesse Rover. Not sure yet if it was a good idea or a mistake. It weighs near 100 lbs. Also, I still don't know what some of the pins are for and can't find any documentation. As a bonus it has an ISO 30 ATC (which I plan to use as a quick change, not auto change) and lots of power.
    50kgs for the power head alone, that's hefty. Can you drive it?
    I won't make it past 3kW where I am, unless I somehow manage to get a solar feed battery system with a custom inverter going, then only the sky is the limit (or my pockets) ;-)

    Attached Thumbnails Attached Thumbnails Why aren't portal mills more popular?-portalmill_grateanalysis-jpg   Why aren't portal mills more popular?-portalmill_jigsawprinciple-jpg  


  9. #29
    Member JoanTheSpark's Avatar
    Join Date
    Jun 2016
    Location
    Australia
    Posts
    26
    Downloads
    0
    Uploads
    0

    Default Re: Why aren't portal mills more popular?

    Another analysis, this time the whole base frame, with simulated crossing struts under the table by fixing the bottom faces and the bed (lower half rigid).
    Total force is 50kgs, split half between each wall.
    Displacement comes in at 9 micrometers



    Attached Thumbnails Attached Thumbnails Why aren't portal mills more popular?-portalmill_baseanalysis-jpg  


  10. #30
    Member mactec54's Avatar
    Join Date
    Jan 2005
    Location
    USA
    Posts
    15362
    Downloads
    0
    Uploads
    0

    Default Re: Why aren't portal mills more popular?

    Quote Originally Posted by JoanTheSpark View Post
    Another analysis, this time the whole base frame, with simulated crossing struts under the table by fixing the bottom faces and the bed (lower half rigid).
    Total force is 50kgs, split half between each wall.
    Displacement comes in at 9 micrometers

    Looks good, I doubt that you would ever see 50Kg working forces in those directions, so some good number to use and worth while tests

    I like what you said in your other post 28, (I would love to put the full triangle on the y-rails, but I'm stubbornly set on 1,600mm long y-rails and a 1,200mm y-envelope, so that's that. ) most would not have a clue what you where talking about, so its good to see, that you know how and what you are designing

    I know mass is better, I have been looking at aircreate I may give it a try for a Gantry Beam fill, and add some other additives to it

    Mactec54


  11. #31
    Member JoanTheSpark's Avatar
    Join Date
    Jun 2016
    Location
    Australia
    Posts
    26
    Downloads
    0
    Uploads
    0

    Default Re: Why aren't portal mills more popular?

    Quote Originally Posted by mactec54 View Post
    I like what you said in your other post 28, (I would love to put the full triangle on the y-rails, but I'm stubbornly set on 1,600mm long y-rails and a 1,200mm y-envelope, so that's that. ) most would not have a clue what you where talking about, so its good to see, that you know how and what you are designing


    Purple - (some sort of plate) vertical stiffness for gravity bending by weight of gantry, sled, z-ram + spindle
    Yellow - (some sort of plate) horizontal stiffness for reaction forces
    Tube/Pipe - takes rotational moments via ribs and 'hands' them over to the ones riding on the y-carriages
    Red block - y-carriages applicable length for moments
    Green triangle - ~converts moments from x-rails to y-rail

    Lot's of tinkering to do, I don't like it yet ;-)

    Attached Thumbnails Attached Thumbnails Why aren't portal mills more popular?-portalmill_triangularpipegantry-jpg  


  12. #32
    Member mactec54's Avatar
    Join Date
    Jan 2005
    Location
    USA
    Posts
    15362
    Downloads
    0
    Uploads
    0

    Default Re: Why aren't portal mills more popular?

    Quote Originally Posted by JoanTheSpark View Post


    Purple - (some sort of plate) vertical stiffness for gravity bending by weight of gantry, sled, z-ram + spindle
    Yellow - (some sort of plate) horizontal stiffness for reaction forces
    Tube/Pipe - takes rotational moments via ribs and 'hands' them over to the ones riding on the y-carriages
    Red block - y-carriages applicable length for moments
    Green triangle - ~converts moments from x-rails to y-rail

    Lot's of tinkering to do, I don't like it yet ;-)
    It would all depend on where you place the Z axes Bearing carriages, if in the Green Zone I would not worry to much about it, adding out to the yellow would mean moving the Bearing Carriage as well for it to have the best affect, this would then compromise your travel, 354mm is very good but you can not always have what you need, If you have to compromise to make one part work, it will suffer some where else

    The added plate ribs, could also be cut at 45 deg from the orange green point intersection

    Mactec54


  13. #33
    Member handlewanker's Avatar
    Join Date
    Sep 2006
    Location
    Australia
    Posts
    6463
    Downloads
    0
    Uploads
    0

    Default Re: Why aren't portal mills more popular?

    All I can say is, if you fitted a pair of wagon wheels to the sides of it and a disselboom out front it would be the makings a nice farm cart.....LOL.

    Actually, I very much doubt ANYBODY would go to the lengths of this build just to mill a bit of plastic or occasional aluminium.......it seems the sledgehammer to crack a walnut is still alive and kicking.

    Oh well, the proof will no doubt be in the pudding if/when the real build begins.........then, if the final figures for rigidity and accuracy are not HYPER met in the real world.......it might as well have just been a 6040 china buy in model lookalike for all the trouble.

    I wonder where the phrase KISS went to.......that's the first rule of engineering I learned at my father's knee.
    Ian.



  14. #34
    Member
    Join Date
    Jun 2007
    Location
    USA
    Posts
    47
    Downloads
    0
    Uploads
    0

    Default Re: Why aren't portal mills more popular?

    I'll say it - I don't like your base design.

    I don't see the point of anything that is below the table. Half of the design is wasted time, material and ineffective mass. You can see this clearly enough in your deflection plots - everything under the table is blue and useless because you're already limited by anything red.

    You'd be better off doubling the thickness of the walls and table and putting the whole thing on the ground. Same material cost, same weight, much stiffer design.

    Ian.

    Sent from my SAMSUNG-SM-G900A using Tapatalk



  15. #35
    Member hanermo's Avatar
    Join Date
    Aug 2004
    Location
    barcelona
    Posts
    780
    Downloads
    0
    Uploads
    0

    Default Re: Why aren't portal mills more popular?

    Lemme ad some stuff and opinions.

    I have built a double-column portal milling machine for steel.
    Really a very large work-envelope VMC.
    The version 5 is now nearing movement .. again.

    I have tested all 5 iterations with a DTI, with the dti apart fom the machine, indicating frame flex alone.
    My load was me standing on the spindle - about 80 kg mass from me,or 800 N force in Z.

    Fwiw..
    cutting forces are appox 100 kgf, or 1000 N, for 2-3 Hp or so at cutting end or tool, in steel.
    So a 80 kg mass is == right.

    My deflections went from ==
    0.3 mm,
    0.2 mm,
    0.15 mm,
    0.09 mm,
    and I now expect 0.05 mm or hopefully much less circa 0.02-0.03 mm.
    All with the similar 80 kg when I stand on the spindle support end.

    My structural members on the current last iteration are about 100x stiffer, 50x stronger, and 5-20x more massive.

    Minimum thickness of stiffening skin is 10 mm in tool steel (on the 2 vertical portals skins).
    Typical is 20-30 mm thick on spindle ram, and the spindle box 20-30 mm thick.
    Machine feet 20+25 thick = 45 mm thick on 250 x 700 mm area.

    The last iteration will be pre-loaded to approx 2000 kgf on the x axis, with 2 x 8 mm bolts pulling the mounting plates, until they are rigidly fixed in place.
    IE the linear guides are fixed to their mount plates rigidly, but the mount plates at one end are loose.
    Then tension via bolts sideways (x), taking out all play in the plates, linear guides, frame etc.
    Then bolt the rigid fixing bolts in place, fixing the position.

    I hope/expect to be able to use 200 kgf force in milling steel, perhaps 300 kgf.
    This would relate to 4 kW or more of usable spindle power (spindle rated for 7 kW).
    I hope to get 1-2 kW out of it in use, while cutting.

    The 32/4 ballscrews are rated to 1400 kgf, +/-, 14.000 N, fwiw.
    Rigidity of 32/4 screws == 54 kg/micron, or 540N/um.

    My point is this.
    Rigidity is everything.
    Oversizing components is vastly easier and cheaper than trying to make things right.
    ALL commercial machines today use similar oversizing.

    I use similar sizes of screws and blocks as commercial machines in the 1.6 m table width class.
    35 mm for me, 45 mm for some machines, 35 for some others.
    But the portal fixed-fixed design is 8x more rigid / length already vs a typical C-frame on a common milling machine.

    Fwiw..
    The industrial machines deliver 25 kW and I hope to do 1-2 kW.



  16. #36
    Member JoanTheSpark's Avatar
    Join Date
    Jun 2016
    Location
    Australia
    Posts
    26
    Downloads
    0
    Uploads
    0

    Default Re: Why aren't portal mills more popular?

    Quote Originally Posted by hanermo View Post
    I have built a double-column portal milling machine for steel.
    Really a very large work-envelope VMC.
    The version 5 is now..
    Sounds very cool.
    Do you have any images/specs/design hints I could look at please to advance my design?
    From what you describe I can only comprehend 1/3rd and probably screw up the other 2/3rds in my head.. :-(
    I tried your nick on the forums, but only found reply posts of you to other topics, nothing that would give me that kind of information.

    I want to be able to mill steel if I'm careful and slow (within means), I don't intend to do 24/7 production with this.
    Keep in mind, that I definitely don't want to spend north of 10,000 USD on a machine like this for materials, better less.
    We're talking hobby/amateur prototyping stuff here, with the occasional piece of steel if it is needed.

    Thanks!

    Quote Originally Posted by genixia View Post
    I don't see the point of anything that is below the table. Half of the design is wasted time, material and ineffective mass. You can see this clearly enough in your deflection plots - everything under the table is blue and useless because you're already limited by anything red.

    You'd be better off doubling the thickness of the walls and table and putting the whole thing on the ground. Same material cost, same weight, much stiffer design.
    That everything underneath is blue is mostly because of having the FEM analysis fix the table (the t-slot piece you can see there) and the bottom of the column-grates. For the FEM analysis those pieces are non-moveable. I did this, because I was only interested in the flex of the column-grates above the table to see if 3 mm wall thickness tubing 100x100 SHS would do and the simulation took 45 mins to run already. The box underneath will be (as earlier mentioned) stiffened by crossed bracing to get a similar effect.

    The columns/walls upper pieces are the only stuff left deflecting because of the wasted material & effort underneath it. It's what keeps the columns stiff up to where they extend past the box and have to hold up on their own.
    It's like a clamped beam in this regard.
    You're right, if I was to double up the wall thickness of the column-grates it would be stiffer, but the parts then approach weights I won't be able to handle anymore.

    If the box wasn't there and the whole thing designed like a big-u-shape, the table would start to bend together with the columns and I'd get a more complex reaction of the frame to forces like this.
    It will be worse than a clamped beam then.
    I'll try to make a FEM-analysis that compares your approach and mine with a more simple model if I can find the time.

    Last edited by JoanTheSpark; 11-30-2017 at 12:15 AM.


  17. #37
    Member
    Join Date
    Apr 2009
    Location
    USA
    Posts
    5516
    Downloads
    0
    Uploads
    0

    Default Re: Why aren't portal mills more popular?

    Quote Originally Posted by genixia View Post
    I'll say it - I don't like your base design.

    I don't see the point of anything that is below the table. Half of the design is wasted time, material and ineffective mass. You can see this clearly enough in your deflection plots - everything under the table is blue and useless because you're already limited by anything red.

    You'd be better off doubling the thickness of the walls and table and putting the whole thing on the ground. Same material cost, same weight, much stiffer design.

    Ian.

    Sent from my SAMSUNG-SM-G900A using Tapatalk
    I suppose you like working on the ground?!



  18. #38
    Member
    Join Date
    Jun 2007
    Location
    USA
    Posts
    47
    Downloads
    0
    Uploads
    0

    Default Re: Why aren't portal mills more popular?

    Quote Originally Posted by louieatienza View Post
    I suppose you like working on the ground?!
    I don't, but nothing prevents it from being put on a suitable foundation to raise it up if that is the concern. That doesn't need to be part of the machine itself though.

    I'm saying to make as much of the mass and structure of the machine add directly to total rigidity. Have an extremely rigid table is great, but if your gantry is flopping all over the place in X because it's supported by two comparatively weak cantilever beams then what's the point?

    I'll be interested to see the FEA of the double thick U shape.

    Sent from my SAMSUNG-SM-G900A using Tapatalk



  19. #39
    Member
    Join Date
    Mar 2017
    Location
    United States
    Posts
    926
    Downloads
    0
    Uploads
    0

    Default Re: Why aren't portal mills more popular?

    What makes you think they aren't popular?

    I found loads of portal mills when I was doing research for my build. They aren't always referred to as "portal mills" specifically though. I have seen them called "dual column mills", "gantry mills", "fixed gantry mills", "bridge mills" etc... but... it's all the same thing - I.e. A fixed gantry / moving table machine.


    You can find both routers and mills in any of the commonly used designs. I.e. Both mills and routers can be either fixed column, dual column or moving gantry designs. Either can be too light and flimsy and either can be well made, stiff and weigh many thousands of LB's. If there is any distinction at the premium end of the market, I think it is only what they are designed to cut.



  20. #40
    Member
    Join Date
    Mar 2010
    Location
    switzerland
    Posts
    12
    Downloads
    0
    Uploads
    0

    Default Re: Why aren't portal mills more popular?

    @JoanTheSpark
    I guess it might not be that relevant for you anymore but you might like to look at this d.i.y. portal cnc build: https://www.wadeodesign.com/design-details.html



Page 2 of 3 FirstFirst 123 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

Why aren't portal mills more popular?

Why aren't portal mills more popular?