[thegolfer]

Hi All,

I am a third year mechatronics student in Melbourne, Australia (nearly 30, however, so definitely a mature age student). After being interested in CNC for a while, and having built several CNC routers, I have decided to design (and hopefully build) a small CNC mill. Depending on approval from my university, I might also turn this design process into my final year project next year. I would like to work in the machine tool design field after university, so this kind of project might be one that gets me a foot in the door.

The basic goals of my mill design are:

• X, Y, Z travel of 400 mm x 280 mm x 280 mm (15 in x 11 in x 11 in)
  
• 2.0 - 2.5 kW BT30 spindle (2.7 - 3.4 HP)
  
• 15 - 20 m/min rapid speeds (590 - 787 in/min)

The structural elements of my machine will be made from cast iron and I will employ servos, linear rails and ball screws on each axis.

The first steps I will be taking in my design process are:

• layout out a rough design of the cast iron elements of the mill to get an idea of the mass of each major moving element;
  
• undertake a "first pass" selection of the linear rails and ball screws;
  
• calculate the power requirements to accelerate the mass;
  
• select suitable servos;
  
• calculate the rated linear force available from the selected servos (and define this as the maximum cutting force of the mill);
  
• find the deflection of each major moving element under the above calculated linear force; and
  
• update the design on the cast iron elements to minimise deflection.

Below is an image of the beginnings of my rough design. Currently it's around 300kg (660 lb).

Attachment 236354

Any comments would be greatly appreciated!

[awerby]

Are you going to cast all this iron yourself? Or make patterns and send them out to a foundry? The design you've sketched is interesting and not typical. Most cast iron mills are based on a big "C" - shaped casting, which maximizes rigidity. Is there some particular reason you've decided not to do it that way? Moving that big iron block up and down with the Z axis will tax your motors and motion parts without particularly affecting rigidity, although it will add mass. Usually the Z-axis spindle quill moves while the structure is fixed. Making the column and base separately makes more sense from a casting standpoint. But why the small square table? You generally want more length, so you'll have room for clamps and fixtures.

[Topdecking]

Thoughts on the table.
Flip your assembly over. The plate between the X and Y axis should be as short as possible, and large enough to mount the linear slide blocks upside down, and the ballscrew mount. This gives you a large table to work with and improves rigidity by reducing the lever length.

[mactec54]

thegolfer

This thread will give you some help/idea's, on building your machine, I think the first one that he built is to small, to be of any real use, but his bigger one is a good size, anything in between the (2) should be a worth while protect

http://www.cnczone.com/forums/vertic...nning-end.html

[Tkamsker]

I was about to send Same link Even if i think the whole Story is strange ...
I am About to Design a 5th axis for that size of Mills like tormach novakon klippfeld and so on ... so if you want to jo in ...
Should in the end also be an cast iron ..

Gesendet von meinem SM-N9005 mit Tapatalk

[thegolfer]

Are you going to cast all this iron yourself? Or make patterns and send them out to a foundry? The design you've sketched is interesting and not typical. Most cast iron mills are based on a big "C" - shaped casting, which maximizes rigidity. Is there some particular reason you've decided not to do it that way? Moving that big iron block up and down with the Z axis will tax your motors and motion parts without particularly affecting rigidity, although it will add mass. Usually the Z-axis spindle quill moves while the structure is fixed. Making the column and base separately makes more sense from a casting standpoint. But why the small square table? You generally want more length, so you'll have room for clamps and fixtures.

Hi Andrew,

I am not going to cast the iron myself. I think that casting iron is mainly left to foundries given the high temperatures that are required. Once the design is finished I will look into making the patterns myself, however I'll have to assess whether I can do it well enough to get a good cast result. Following the advice of a number of iron casting books I looked up in my University library, the general consensus in designing cast iron parts seems to be to design the part for function only and leave the casting details to the foundry (drafts, parting lines, etc.). Due to this, making my own patterns might limit the foundries ability to draft the parts and place parting lines correctly. All of that said, though, there seem to be a number of important elements to the design of a cast iron part that I need to consider to avoid stresses from non-uniform cooling and the like.

As far as the shape of the mill is concerned, I think it's pretty much like any other knee-style mill. I looked at a lot of images of mills before starting my design and have tried to base my design on those. Below are two reference images I have used.





Regarding the Z axis, the block shown in my first image is just a representation of the Z axis spindle head. Below is an updated image of my design showing the spindle head with more detail. I think that the whole spindle head assembly should be around 75 kg. If this is too much weight, however, I may add a counterbalance inside the column of the mill.





As for the table, it will be made longer to accommodate fixtures. I still have to think about that!

[Tkamsker]

Hi i refurbish right now an bridgeport 412 where the Motor had 75 kg - which Travel distances do you try o accomplish ?


Gesendet von meinem iPad mit Tapatalk

[thegolfer]

Thoughts on the table.
Flip your assembly over. The plate between the X and Y axis should be as short as possible, and large enough to mount the linear slide blocks upside down, and the ballscrew mount. This gives you a large table to work with and improves rigidity by reducing the lever length.

Hi Topdecking,

I have not gone down that path because of the linear guide/ball screw covering I want to design. I would like to use either steel or "rubberised" fabric covers and create a machining environment much the same as a large-scale VMC, where all mechanics are completely sealed from swarf and coolant. I'll keep working on my design and this will become more apparent.

Also, I'm not sure that having a long sub-table such as mine increases lever length as the forces from the spindle are always going to be acting on a Y-Z plane through the centre of the table and sub-table. Yes there will be a lever on the end of the sub-table if the table and work piece are at the very end of the X axis travel (that decreases as the table and work piece gets closer to the centre of the X axis travel) , but wouldn't this be the same for a long table, small sub-table design?

[thegolfer]

thegolfer

This thread will give you some help/idea's, on building your machine, I think the first one that he built is to small, to be of any real use, but his bigger one is a good size, anything in between the (2) should be a worth while protect

http://www.cnczone.com/forums/vertic...nning-end.html

Hi Mactec54,

I have had a look through that thread and it has a lot of interesting information. That guy clearly knows how to build a machine and how to get things done! I also like the look of his BT30 spindles.

I was about to send Same link Even if i think the whole Story is strange ...
I am About to Design a 5th axis for that size of Mills like tormach novakon klippfeld and so on ... so if you want to jo in ...
Should in the end also be an cast iron ..

Gesendet von meinem SM-N9005 mit Tapatalk

Hi Tkamsker,

A 5 axis machine is the pinnacle! I'm looking forward to seeing your design.

Hi i refurbish right now an bridgeport 412 where the Motor had 75 kg - which Travel distances do you try o accomplish ?

Gesendet von meinem iPad mit Tapatalk

My travels are going to be 400 x 280 x 280 mm (X, Y and Z axes).

[Tkamsker]

Hi,
My design will be straight forward
a trunnion with 1 Harmonic drive gear (or 2 ) depending on the sourcing which i can get.
The idea is to use 2 Servos in one Granitedevice dual configuration.

Fortunatley i already have a VMC Mill which fits your travels - look http://www.tkamsker.at:8080/web/pub/...ctory-retrofit (enjoy the pictures)

And the guys from that factory are verry helpful we right now design am full enclosure similar to the bridgeport.

My prototype will be in normal steel but in mass production it should be cast iron so i watch your thread about the learnings -
thx
thomas

[Winterwatchers]

Have you given consideration to using steel tubes filled with epoxy granite? Cast iron is a very involved process and is not cost effective at all if you are only doing it once. It takes a long time to season the casting and machining adds more to the cost. It took rapidfire years to make his from beginning to end and he is doing it as a business.

You don't have to weld steel tubes, a carefully designed bolt + epoxy setup is just as strong as weldment and doesn't have residual stress problem to worry about. Just fill it up with non-shrinking cement if EG is too expensive. Read up on rapid machine design from MIT, I am using that as a basis to design my own vmc.

[handlewanker]

Hi, love the column.....hate the small square table.........a casting will be impractical from the cost point of view.......take note of David DeCaussin's UMC-10 CNC build on UTUBE.

As the man said, castings need to be aged......castings are only really practical if you want to have multiple parts the same without having to fabricate from scratch, but you need to really, as in extremely really, have a design that is completely R&D'd and finalised and not need modifications due to shortsighted thinking......a casting that ends up having to be reworked extensively is a waste of money.

When you go to a casting it's an object that is the final solution and only needs machining where it needs it.....you are going to reinvent the wheel without knowing what road it will ride on?

My opinion is that if you do a fabrication, based upon current CNC technology, you can adapt and "re-invent" it as the need arises without having to outlay for castings which as one offs will not be cost effective for the foundry where the total cost is bourne purely on one item and has to be passed onto the customer.

A fabrication can be re-worked constantly while you re-think the build process as you go along, and it doesn't have to look like a pile of bolted sections like a Meccano set.

By my estimate, you will not get any change out of an "investment" of $50,000 once you add up all the outside manufacturing costs, which will be approx. $50 to $80 an hour.
Ian.