New Machine Build E/G mill for steel and aluminum


Results 1 to 9 of 9

Thread: E/G mill for steel and aluminum

  1. #1
    Registered
    Join Date
    Jul 2017
    Posts
    2
    Downloads
    0
    Uploads
    0

    Default E/G mill for steel and aluminum

    Hello Zone

    Long time lurker but first time poster. A friend and I are about to dive into our first CNC-mill build. The design is pretty much sketched out, but we would like some opinions and thoughts, before we rush out to build moulds and buy steppers :-) The machine is intended for precision machining of both aluminum and steel, so we would like to keep tolerances tight. It is not your usual C-frame-design, but it seems to have some merits and some flaws. We will probably build the spindle ourself, but this has not been drawn yet. Some of the facts and thoughts about the design:


    • Work envelope is ~400mm*250mm*270mm
    • Epoxy granite base, column and ram. The moving beam has internal steel bracing. Rebar is embedded in the base and column.
    • Steel inserts for all mating/mounting surfaces
    • Work envelope relative to total machine size is larger than common C-frame mills, but smaller than gantry-style machines.
    • Worst case deflection of structure is 3mu for a radial cutting force of 1kN, but rigidity increases closer to work surface. Most deflection is due to twisting of the column. This is ignoring compression of the ball screws.
    • The spindle is able to move completely clear of the table. This provides good clearance for planned ATC and toolsetter
    • All 3 axes are moving on 20mm rails.
    • Ballswcrews are 1604 with 0-backlash nuts. C7 precision with screw mapping for compensation
    • X- and Y-axis driven by 2Nm closed loop stepper motors(china). Should provide plenty of torque, inertial ratio of about 1.5. Z-axis possibly 3Nm since the moving beam is heavy(~70kg). It is however also supported by a gas spring.
    • A concern is the moment due to the offset spindle compared to z-axis rails, which would seek to twist the moving beam. However, most cutting forces are horizontal, and thus the relevant moment arm is only ~40mm(vertical distance between rails and spindle nose)
    • Way covers for Y-axis are problematic


    Similar Threads:
    Attached Thumbnails Attached Thumbnails E/G mill for steel and aluminum-img1-jpg   E/G mill for steel and aluminum-img2-jpg   E/G mill for steel and aluminum-img3-jpg   E/G mill for steel and aluminum-img4-jpg  

    Last edited by moldo2; 09-19-2017 at 01:59 AM.


  2. #2
    Member
    Join Date
    May 2011
    Location
    Canada
    Posts
    790
    Downloads
    0
    Uploads
    0

    Default Re: E/G mill for steel and aluminum

    Quote Originally Posted by moldo2 View Post

    We will probably build the spindle ourself, but this has not been drawn yet.
    That's ambitious! An ATC spindle of your own design? Do you have experience with that? Please post how you do it and your calculations for figuring out the motor windings, torque vs speed curve, Amp draw, ATC system, all that good stuff please. If you know what you're doing I'd like to learn from you. If not, this aspect will fail miserably. Either way, it would be an interesting read.

    Quote Originally Posted by moldo2 View Post

    Worst case deflection of structure is 3mu for a radial cutting force of 1kN, but rigidity increases closer to work surface.
    What's an mu? Keep it simple mate, inches or mm. Did you do an FEA on this?

    Did you also include dynamic deflections due to machine accelerations in your calculations?

    Quote Originally Posted by moldo2 View Post

    Ballswcrews are 1604 with 0-backlash nuts. C7 precision with screw mapping for compensation
    16mm seems small to me for a machine that is good at milling steel. What are your goals for top speed and acceleration?

    Quote Originally Posted by moldo2 View Post
    Z-axis possibly 3Nm since the moving beam is heavy(~70kg). It is however also supported by a gas spring.
    Pneumatic counterbalance using pneumatic cylinders or gas struts? I don't see it in your drawings.


    Have you considered looking for some really flat big box square aluminum tubes? I say aluminum because in my limited experience it generally comes flatter and straighter than steel ones. I have to think it might be easier to make the structure out of square tube sections, then fill the entire thing with epoxy granite, leaving the more attractive and wear resistant aluminum as a permanent skin.



  3. #3
    Registered
    Join Date
    Jul 2017
    Posts
    2
    Downloads
    0
    Uploads
    0

    Default Re: E/G mill for steel and aluminum

    Quote Originally Posted by NIC 77 View Post
    That's ambitious! An ATC spindle of your own design? Do you have experience with that? Please post how you do it and your calculations for figuring out the motor windings, torque vs speed curve, Amp draw, ATC system, all that good stuff please. If you know what you're doing I'd like to learn from you. If not, this aspect will fail miserably. Either way, it would be an interesting read.
    We are only making the spindle and drawbar. Drivetrain will be using a pre-wound motor. Possibly BLDC, but this is still undecided. The ATC itself is way down the line.

    Quote Originally Posted by NIC 77 View Post

    What's an mu? Keep it simple mate, inches or mm. Did you do an FEA on this?


    Did you also include dynamic deflections due to machine accelerations in your calculations?
    Greek letter mu. Means micrometer. 1 mu = 0.001 mm

    This is based on static FEA. Thus ball screw deflection and dynamics are ignored. Would love to hear some thoughts on those effects.

    Quote Originally Posted by NIC 77 View Post
    16mm seems small to me for a machine that is good at milling steel. What are your goals for top speed and acceleration?
    We'll have to do more calculations with regard to the ball screw, but with the current parameters, it is well below buckling strength. Rapids of 4000mm/min and 0.2g should be attainable, we'll post relevant calculations later.


    Quote Originally Posted by NIC 77 View Post
    Pneumatic counterbalance using pneumatic cylinders or gas struts? I don't see it in your drawings.
    Just a simple gas strut. It is hidden behind the Z-axis cover next to the ball screw - ie. between the rails.

    Quote Originally Posted by NIC 77 View Post
    Have you considered looking for some really flat big box square aluminum tubes? I say aluminum because in my limited experience it generally comes flatter and straighter than steel ones. I have to think it might be easier to make the structure out of square tube sections, then fill the entire thing with epoxy granite, leaving the more attractive and wear resistant aluminum as a permanent skin
    All mating surfaces will either be ground, scraped or indicated using Moglice or similar allignment epoxy.



  4. #4
    Member
    Join Date
    Jul 2014
    Posts
    215
    Downloads
    0
    Uploads
    0

    Default Re: E/G mill for steel and aluminum

    Overall, I like your design. It seems like a good design for space savings and also reduced EG. Things to consider are method of securing Z axis column to the base to get it aligned to a high tolerance. That will not be too difficult to address. The one weakness I see is that when the Y axis is machining hard material way out near it's left hand limits the spindle is quite a ways away from the Z axis column and so vibrations are going to be higher on that side of the mill than on the other. This is going to become a bigger problem with the harder material that you cut. It might not be a big problem and I don't know if it will effect the machines ability to cut as much as surface finish problems will be higher on one side than the other. If you are planning on doing high tolerance work with good surface finish you should consider going with a double column gantry or a more traditional monolithic vertical column of larger size. It doesn't make sense to skim on the amount of EG you have to buy as it is a small part of the overall cost. But, then again I don't know your space requirements for the machine either or how much weight you are able to deal with moving around. I think it would work pretty well for aluminum but any weaknesses in the machine are going to show themselves when you start trying to machine stainless steel or other hard to machine alloys.



  5. #5
    Member
    Join Date
    May 2011
    Location
    Canada
    Posts
    790
    Downloads
    0
    Uploads
    0

    Default Re: E/G mill for steel and aluminum

    Quote Originally Posted by moldo2 View Post
    We are only making the spindle and drawbar. Drivetrain will be using a pre-wound motor. Possibly BLDC, but this is still undecided. The ATC itself is way down the line.
    That would still be interesting to see. A word of caution though, it's way better to design it all out with all of the parts included than to start building before your heavy spindle assembly design is finalized. What you are talking about can take up alot of space. I assume that this is a preliminary sketch. There are also units you can buy that already have this capability, perhaps for cheaper than you can make them, but also, perhaps you will make them better, so I'm not saying to give up on that aspect, only to have it all designed before construction, which may be what you are planning.

    Currently drawn it looks like a regular mount for a cylindrical spindle. Not a separate (belt driven?) motor attached to a spindle assembly.

    Quote Originally Posted by moldo2 View Post
    We'll have to do more calculations with regard to the ball screw, but with the current parameters, it is well below buckling strength. Rapids of 4000mm/min and 0.2g should be attainable, we'll post relevant calculations later.
    Don't forget to include the size of the bearings you can use with a smaller screw in your considerations. You want to mill steel. Not saying they won't work, but it's not what I would use. 20mm with 15mm end bearings perhaps. I don't know. I could be wrong about this.

    The speeds and acceleration are quite modest IMO. The work area is small so perhaps that's all you need.

    Quote Originally Posted by moldo2 View Post
    Just a simple gas strut. It is hidden behind the Z-axis cover next to the ball screw - ie. between the rails.
    My original plans for my build were to use two gas struts. I have since reconsidered. Now I am going to use two pneumatic cylinders, hopefully on a closed system that requires very few top ups. I have heard that the gas struts have too much dampening to be very good on a CNC. No first hand experience, that's just what I've heard (a few times).

    I would try to have the start of the cutting bit as close to the bottom of the (gantry?) not sure what you call that axis to reduce the cantilever moment. I'm not sure how you will accomplish this with the separate motor and spindle as you have planned. I think you will need to redesign this a few times to find the correct balance in design parameters with the dimensions of your actual spindle assembly included.

    As far as the design concept is concerned, I like it! Some definite merit in this design. Even with one axis in a cantilever position, certainly there are some advantages to this design. Of course, I'm no expert on epoxy granite structures, and as mentioned before, I wonder if an appropriate outer skin of aluminum or steel would be better.

    Good luck



  6. #6
    Member
    Join Date
    Sep 2008
    Location
    USA
    Posts
    86
    Downloads
    0
    Uploads
    0

    Default Re: E/G mill for steel and aluminum

    Quote Originally Posted by moldo2 View Post
    Thus ball screw deflection and dynamics are ignored. Would love to hear some thoughts on those effects.
    Don't ignore those effects. If you are going to the effort of designing, FEAing and building a machine, don't guess here. Some food for thought:

    A 16 mm dia steel rod of 0.4 m length will compress 29 um under 1 kN of axial load. This dwarfs your 3 um goal.
    Here is a decent link about ball screw stiffness (How to calculate ball screw rigidity)
    Here is a better one, (http://www.flli-frigerio.it/wp-conte.../08/ee_A15.pdf)

    You are going to use a 4 mm pitch ball screw and steppers. Let's say you are using 1.8 degree steppers and you will load them such that they will deflect 20% of a step under max load. That's 4 um.

    Regarding the nut, I pulled a random THK nut (EBA 1605-4) for a sanity check. It has a stiffness of 210 N / um or ~ 5 um at 1 kN.

    And there is the screw bearing...

    I'm not saying, "you're doing it wrong", but it sounds like you have some stiffness goal. It would be a shame if you went to all this trouble only to find you missed the majority of the stiffness budget.

    Linear bearings have stiffnesses, too.



  7. #7
    Member
    Join Date
    May 2013
    Location
    washington
    Posts
    480
    Downloads
    0
    Uploads
    0

    Default Re: E/G mill for steel and aluminum

    the distance the bearing blocks are apart, and the distance between the rails of the z axis should be wider and longer than the Y axis length imo.

    secondly, you can put a hollow section through the z axis column to reduce the weight, and the base can have hollow sections as well. the base should be thicker perhaps, and hollow.


    the triangular gusset in the middle of the y axis is going to take all the stress, the epoxy granite outer blob that surrounds it will probably only be dead weight. if that's your intention, then you've got a lot of dead weight that is dampening vibrations by mass alone. tar mixed with sand might be better! what you have basically made with this machine is similar to a planer from 100 years ago. you can hollow out the epoxy granite cantelivered y axis as well to reduce the inertia and mass of that arm that sticks out, as you say the primary weakness is the torsional twisting of the column, but what is its fundamental frequency compared to other modes in different axis of the rest of the machine?

    the metric you should look at is this:

    stress times dampening coefficient times modulous of elasticity. if all the stress is in the steel sub structure of the machine then the epoxy granite is just dead weght.

    adding dead weight doesn't do anything but reduce the fundamental frequency of the structure. as such it "sounds" like it is dampening the vibration, but its just lowering the frequency, which to the human ear given its non linear responses sounds like its doing something when it really isn't.

    some time ago i was milling some aluminum on a bridgeport with a 2 flute endmill, i was conventional milling in the y axis, towards me. at around 900 rpm, i hit the fundamental frequency of the machine. it still made a good cut due to the light cut, but i could hear the machine resonating like a wine glass at 30hz



  8. #8
    Member
    Join Date
    Sep 2008
    Location
    USA
    Posts
    86
    Downloads
    0
    Uploads
    0

    Default Re: E/G mill for steel and aluminum

    Quote Originally Posted by Eldon_Joh View Post
    adding dead weight doesn't do anything but reduce the fundamental frequency of the structure. as such it "sounds" like it is dampening the vibration, but its just lowering the frequency, which to the human ear given its non linear responses sounds like its doing something when it really isn't.
    This! So many people think mass is damping, This is clearest way I have seen anyone explain what is going on when you add mass.

    Last edited by ebrewste; 09-22-2017 at 04:10 PM.


  9. #9
    Registered Zach_G's Avatar
    Join Date
    Jun 2005
    Location
    USA
    Posts
    98
    Downloads
    0
    Uploads
    0

    Default Re: E/G mill for steel and aluminum

    I played around with this design once, it seems like a good space saving layout and novel because no one does it, but there's reasons for that. As previous responses have pointed out, most of your deflection is actually from the linear motion bearings, ballscrews, and angular contact bearings rather than the structure. I did static FEA on my structure as well without taking account the springiness in those motion elements and the as-built machine had deflections 5 times what the FEA showed. This particular layout places the forces and moments further away from the linear bearings at the travel extremes than traditional Bridgeport types. Not that it can't work well, just have to oversize the bearing components. Ultimately I went with a traditional Bridgeport type layout because it was easier to fabricate square. It's still a fun layout ya got there, would be cool to see someone pull it off well!



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

E/G mill for steel and aluminum

E/G mill for steel and aluminum