[peteeng]

Hi Craig - I agree an 80mm lump of steel is significant. It's how to get the best value out of that lump and that is up to Lars to figure out Personally after analysing many many single column designs I'd go high rail with walls. Will be much stiffer in the same footprint... But again, its Lars machine and he makes the final decsions....Peter

[joeavaerage]

Hi petenng,
couple of issues with high walls:

1) Unless a high rail design can be disassembled it is an outsize structure that needs be fitted into a suitable mill for machining the lands. Column designs tend to disassemble into largish rectangular
parts well suited to mid sized mills.

2) Access to parts is much restricted. The width of a part in a high rail design (in the X direction) is the width between the walls LESS the X travel of the toolpath. I've had a six foot shaft in my machine,
that the shaft overhangs in the X direction is no issue.

3) Column machines tend to have much more Z axis travel. This is good not just for tall parts but if you ever mount a fourth axis you'll need that extra travel.

Downside of a column machine is the stiffness required of the column, base and headstock.

Swings and roundabouts.

Craig

[peteeng]

Hi All - Not trying to high jack this into a high rail design thing but:

1) A high rail machine can be designed to dissemble so it can be machined as you discuss
2) All my wall designs have a front shelf so you can put in objects longer than the width of the walls, I have cut 6m objects in my router for instance
3) Z axis length is a design issue not a limitation of the wall design. All my designs have a least 350mm Z travel... next one will be more to accommodate a tilting head, 450mm maybe 500mm. Just working through that now...

All of these issues are design issues and a clear understanding of what you want the machine to do is needed to create the "best" machine within that scope. Peter

The image attached is the development path from C frame to high rail...

[Fatamorgana]

Hi Lars,
the pics you posted of the linear rails/cars is that they are 30mm rails...correct? If so they a 42mm high.....correct?
The ballscrews you pictured are 'cut flange types'.....correct?

If so they could be as high as 48mm....correct?

In that case the gutter need be 6mm ONLY!!!!

More like you want some clearance and so the height of the ballscrew and ballnut mount might be 55mm say, the gutter still only need be 55-42=13mm deep. In an 80mm slab
of steel a 13mm gutter is hardly going to ruin its stiffness!!

If the gutter were 40mm deep in an 80mmm slab then I'd say yes, you probably need to re-think what you're doing but a 13mm gutter??....., do not introduce a whole bunch of complexity
with respect to construction and multiple setups in your machine just to solve a problem that is just not that important.

Craig

Hello, I have put in the "high" cars from HIWIN and yes there are 30mm rails, as you can see this is the challenge with the gutter. With this setup I will need 20mm gutter. For simplicity and machining in mind I think I prefer to do the gutter the "classic way" - up mid. I will though increase both y/z columns to 100mm steelslabs, and after gutter has been machined out I still will have 80mm left, after 3 days of simulation I cant really put to lines under an answer, I must admit this lies abit over my head, but I cant understand why two of this heavy slabs wont be rigid enough for these dimensions of my machine, I have to put some time in the right places where it will take me further along the path of evolving against something physical. Im btw planning for a z-height that allows me in much later stage in the future to install an ATC, how much space would I clear for having that possibility along the z-axis you think?

Thanks!

[peteeng]

Hi Lars - There is no end to chasing stiffness. At the end of the day its as stiff as it is. I'm sure it will be stiff enough. Peter

[joeavaerage]

Hi,

but I cant understand why to of this heavy slabs wont be rigid enough for these dimensions of my machine,

I think they are. 100mm thick with a 20mm gutter should be any amount stiff enough. I think you may be thinking that it has to be nnn N/um, but your estimate is too high.
It would be nice but the geometric size would also need to be larger. As peteeng has repeated geometry wins hands down over materials. With a solid slab of steel
with a very modest gutter the only possible way you can make it stiffer is increase its geometry, the material is a stiff as ANY practical material.

That you can get such big chunks of steel at such good prices is an advantage that I can only envy....were I to buy the same thing I would be paying 25%-50% more.
I would say go for your 100mm slabs. They are still probably less expensive than the ballscrews and linear rails/cars are they not?. Just the Y and Z axes are going to be about 250kg
each, and that will make a fine machine.

As far as the Z axis goes I would suggest that you try to retain some flexibility with regard the arrangement. For instance if you made the Z axis 1.2m long rather than 1m long, then at a later date
you could unbolt the rails/cars and ballscrew and shift them upwards by 200mm. This might allow you clearance to install a fourth/fifth axis say, or allow for an ATC. The ballscrews you have have fixed
the Z axis travel, and short of replacing them they are fixed. The only thing you can do is 'slide' the Z axis up or down to maximize the machines usability. This is a large reason that I went
with a column design with modular axis beds, it allows me to shift those axis beds to accommodate future developments.

Another area that requires very close thought is the space between the column and the Y axis....when the Y axis is at its closest approach to the column. Making that space large allows you to fit bigger parts in the machine
without fouling the column when being machined......and that is highly desirable, but that places very high demand on the stiffness of the base. There is a balance to be struck. Another consideration
is that space will have to be enough to accommodate the way covers when they are fully collapsed.

You can see from the pics that I gave this a lot of thought. I wanted to make the way covers as simple as possible and therefore secure the most reliable and trouble free operation. To that end
the way covers are three pieces, one attached to the axis, one attached to the frame , and one sliding section. Because I chose to have but one sliding section the collapsed length is quite large.
If i had more sliding sections the collapsed length is less and therefore the column to Y axis space decreases. It is a balancing act. In my case I rather think that I have made the gap larger than ideal.
The way covers work well and there is plenty of room to fit parts on the table but the extra length of the base counts against rigidity. Were I to re-design I would reduce the gap by 25%-40%.

Craig

[Fatamorgana]

I think they are. 100mm thick with a 20mm gutter should be any amount stiff enough.
Craig

I have changed the slabs now in the design to 100mm with 20mm gutter, also boxed in the whole column with sidessupports all the way up, also added a plate on its back to stiffen even more, I have as mentioned earlier now clue on FEA, but I have changed something here, and something there, and along the way I have gotten marginal better numbers all the time, also did some more on the design on the spindle head so I know more accurate the weight of the assy. About 150kg.

As far as the Z axis goes I would suggest that you try to retain some flexibility with regard the arrangement. For instance if you made the Z axis 1.2m long rather than 1m long, then at a later date
you could unbolt the rails/cars and ballscrew and shift them upwards by 200mm.
Craig

Great Idea, I wrapped my head around this and came to the same conclusion, changed the column and increased the height with new FEA.

Another area that requires very close thought is the space between the column and the Y axis..

I have about 150mm between the end of travel and the column, and about 440 on the maximum travel, planned for 3 pieces of way cover on 150mm each.

Have gotten around to order some more parts, IKO rails on the way from our favourite shop.

The Z and Y slab are to big for me to machine on my mill Im afraid, also I want this 100%. I have to go to a shop to get it done, got a quotation for around 1600$ for machining the lands, all sides parallell, the gutter, also all of the threads that I need for the assembly on all sides, I think I will go for that.

Next thing to order is some motor bracket mounts and ball-supports, the servo motor is a 80x80 flange, and the ball-screw is 20mm dia. Can someone help me with linking some suitable and good supports? And I really cant understand this, do you need two supports for each ball-screw like in the picture attached, or will one support at the end of the ball-screw and the motor connection do the job?

Thanks!

[joeavaerage]

Hi Lars,
your model is unrealistic. In particular you have constrained the base at six places to an 'infinitely stiff universe', and that is not realistic. The infinitely stiff mounting points
means that the base reflects that stiffness, not infinite but close to, and again that is not realistic.

Go back to #103. You see I constrained the square face of the upstand on the Y axis. The base is not constrained.

In your attached pic you have shown a vertical force in the direction of the Z axis. This is equivalent to the spindle pushing downwards, and that force is opposed by the material/vice/X axis.
The load path is then through the headstock, through the column, through the base and then through the upstand (representing the material/vice/X axis). Your model has the load path
through the headstock, through the column, and the short circuited to 'an infinitely stiff base'.

You have made the column uber stiff by virtue of the sides and closing the 'tube' so now the most flexible or compliant part will be the intersection between the base and the column......
and yet you have constrained the base. BS in....BS out.

Craig

[joeavaerage]

Hi,

do you need two supports for each ball-screw like in the picture attached, or will one support at the end of the ball-screw and the motor connection do the job?

Commonly a ballscrew has two mounts, on at each end. One, the motor mount end, is called the 'fixed mount' in that it has two opposing angular contact bearings and those
bearings restrain the ballscrew fore and aft, ie axially. The other end is called the 'floating' end and is commonly just a single or maybe a pair of regular deep groove bearings. They rotate nicely
to provide radial support so the ballscrew does not whip but it does does not constrain the ballscrew axially as the bearing can float for and aft by a few tenths of a mm, thus accommodating any thermal
growth in the length of the screw.

The 'fixed end' is critical as it transmits the thrust generated by the ballscrew to the axis bed. The angular contact bearings are very high quality and can cost many hundreds of dollars a pair. Were you to buy new
good THK or NSK BK25 or FK25 (suitable for 32mm ballscrews) you may well pay $400 each!

Craig

[jackjr-123]

The angular contact bearings are very high quality and can cost many hundreds of dollars a pair. Were you to buy new
good THK or NSK BK25 or FK25 (suitable for 32mm ballscrews) you may well pay $400 each!

There is really nothing special about those bearings, they are just standard 7xxx bearings in P5 or P4 quality. The ridiculous price tag from THK/NSK doesn't mean anything.

Also these support units (BK/FK, ...) can be very inadequate for machine tools, their stiffness is comically low (in the 60-100N/µm range). In a proper ballscrew assembly they can easily be the weakest link.
The good support units (WBK) use thrust bearings with 60° contact angle, and the axial stiffness is in the 600-2000 N/µm range.

[joeavaerage]

Hi,

The ridiculous price tag from THK/NSK doesn't mean anything.

BS. A pair of matched P4 or P5 bearings can easily be hundreds of dollars. The bearings with a broader contact angle are more expensive again.
Either way they are critical to a machine. Many Chinese support bearings just use deep groove bearings and they sell them as 'fixed' bearings and are plain rubbish.
Even many of the cheaper fixed mounts that use angular contact bearings, the bearings are not matched, so are $100-$150 whereas any that used matched
bearings are $300-$500.

Craig

[Fatamorgana]

Any thoughts on this supports from damencnc? Seems like a serious shop from netherlands.

https://www.damencnc.com/en/fka20-fi...-quality/a1435

[joeavaerage]

Hi,
do yourself a favor and ring up your favorite bearing supplier and price a matched pair of 7005 P4 bearings.

I needed a set for my FK25 fixed mount. I got lucky and my local NSK stockist had one matched P4 set in the country, and I got them for only $350NZD ($210USD).
Now, can you tell me how damennCNC can sell the entire mount for a fraction of what a matched pair of bearings cost???

They may be in the Netherlands, about 18500kms away, but I can smell the BS from here!

If you were to get those mounts and then replace the bearings with genuine matched bearing pairs they'd be fine, but the non-matched bearings are substandard for the machine you are planning.

Craig

[joeavaerage]

Hi,
this listing looks to be better, but I'm still dubious that they can sell them at $250USD with matched bearings:

https://us.misumi-ec.com/vona2/detai...HissuCode=BK25

Craig

[Fatamorgana]

Hi,
do yourself a favor and ring up your favorite bearing supplier and price a matched pair of 7005 P4 bearings.

I needed a set for my FK25 fixed mount. I got lucky and my local NSK stockist had one matched P4 set in the country, and I got them for only $350NZD ($210USD).
Now, can you tell me how damennCNC can sell the entire mount for a fraction of what a matched pair of bearings cost???

They may be in the Netherlands, about 18500kms away, but I can smell the BS from here!

If you were to get those mounts and then replace the bearings with genuine matched bearing pairs they'd be fine, but the non-matched bearings are substandard for the machine you are planning.

Craig

So buy a regular cheap mount and change the bearings is the way to go?

[joeavaerage]

Hi,

This looks to be the genuine article:

https://www.uberbearings.com/BK25-TH...2&sku=BK25-THK

Note particularly that the bearings are listed as '7205 in a DF' set. That is critical, it means they are the genuine thing....but see the price 490GBP!!!.

So buy a regular cheap mount and change the bearings is the way to go?

That is possible. My ballscrews (3) came with two genuine THK FK25 fixed mounts per ballscrew. I had to replace one after I crashed my machine and damaged one. When I removed
them I saw the manufacturer marks that indicate that are a DF matched pair....and that costs a fortune.

The fixed mount transfers thrust from the screw to the machine. Any flexure or lash will be reflected in every part you make. A good quality fixed mount is mandatory. It is entirely probable that a cheap mount
is sufficiently well made that if it were upgraded to DF matched bearings they would be fine. Either way you need good quality mounts or all your effort to build a rigid and accurate machine is wasted.

The broad angle contact bearings that Jack talks about are typically used in large machines and are $1000 each and more. Look at an SKF catalog.....and weep.

Craig

[joeavaerage]

Hi,
these are a matched pair but universal ground, good but not great:

https://www.globalindustrial.com/p/s...m-bore-52mm-od

These look OK, and the price is good, probably universal ground but still useful:

https://www.ebay.com/itm/172696856114

As you can see as soon as you start looking at matched bearing pairs the price goes through the roof.

Craig

[jackjr-123]

Any thoughts on this supports from damencnc? Seems like a serious shop from netherlands.

https://www.damencnc.com/en/fka20-fi...-quality/a1435

These are support units from SYK (Taiwan) and they are top notch. Exact same performance as THK/NSK ones for a fraction of the price.


Craig,
Again, there is nothing special about matched bearings. You can get a genuine pair of 7005P4 NSK/SKF or whatever for about $80 straight from Aliexpress.
Today bearings are manufactured so accurately that matched sets are less and less common. Instead you buy universal bearings and pair them however you want (DB, DF, DT).

[joeavaerage]

Hi,

Again, there is nothing special about matched bearings.

BS. Why do you suppose all the major machine manufacturers pay big dollars for top quality fixed mounts....because they like paying big dollars when cheap would do?
Its your money, and its your choice and ultimately your machine will reflect your choice. My choices are different than yours.

Craig

[Fatamorgana]

Hello

Got supports, rails, cars etc. just before the weekend.

Did some measurements, my gutter needs to be 37 something mm with my sourced parts, thats to much.

Wondering if a solution could be to Get two parallells machined and bolt them to the base, like on attached drawing.

What do you think?



Sent from my iPhone using Tapatalk