Which metal would flex more?

[hanermo2]

The better A. is that it depends.

By mass the alu is 3x less rigid - and importantly it likes to ring and vibrate.
But when you get the same mass in alu, it is thicker, and thickness scales pwr 3 or pwr 4 by section size.

Steel is == 3x better than alu, in rigidity for the same mass and == 3x cheaper by mass.
Cast iron is cheaper than steel and nearly as rigid.
But cast iron is 3x better than steel in vibration dampening, 10x better than alu.

There are effectively zero commercial machines using alu in any modern machine tool bodies, if they cut something via contact (vs laser/plasma etc.).

All modern machine tool bodies are thin-skinned stressed-skin structures. All.
This means the frames are as big as possible, and very thin, and used ribs/preload/mass/anchors.
Typical thickness is around 10 mm on cast iron machine bodies, in the centres, give or take.

An extreme example are modern optical tables.
They are very very light, for what they do.
They are very very rigid, for their work volume or free span length.

Typically, an optical table might be 40 cm thick for a 1 m table, and == 20x more rigid than the best japanese mazak/integrex mill-turn of similar size, or 200-400k$ cost for the mill turn.
My numbers from tlar.

The optical table is two thin skins, and tubes inside.
When I researched them they seemed to be mostly vertical-only tubes, sometimes with various goops/ballast/members/unicorn snot/magic.
The guaranteed rigidity of optical tables is totally unbelievable.

My point.
No-one uses alu.

Here is a data point - I use Thorlabs stuff for some things.
They are a top-shelf supplier of excellent rep.

https://www.thorlabs.com/newgrouppag...tgroup_id=9222

NOTE: !!!
1.7 um / 150 kg load deflection (constraints).
= 1500 N.
882 N / um rigidity.

The best machine tools might do 30-100 N/um, or 8-24x worse.
Iirc the industrial criteria for VMCs == 30N/um... on textbooks.

[Zeeflyboy]

The better A. is that it depends.

A better answer indeed, but an answer to a different question.

[Desertboy]

I was going bed but now I have to work out how to do simulation tests in fusion grrrr.

;)

Couldn't we apply antivibration (Like headphones) to our routers (This is something I've been thinking about for a while) seems like it would be simple enough to do.

I was thinking either hack a pair of headphones (Might not be able to get the frequencies you need) or a raspberry pi.

[PaisleyPCdoctor]

I hadn't even considered vibration as a factor. Thanks.

For what I'm doing, I'll go with the alu, but when I get to the stage of building my own, I'll seriously consider a steel construction.

Thanks again!

[routercnc]

If you want to work it out by hand it is not too tricky for a simple shape like that.

1. The geometry of the part (how the material is arranged, into what shape). This is the cross section property and is called the second moment of area:
Ixx = (bd^3)/12
where:
b is the width
d is the depth
For aluminium part this is (100x25^3)/12 = 130208 mm^4
For the steel part this is (100x10^10)/12 = 8333 mm^4

2. The end condition of the beam (how it is restrained at the ends)
Ec = 192 for fully welded
Ec = 48 for simply supported (e.g. pin joint)

3. The material properties for the material being analysed (Young' modulus)
For aluminium Ym = 69000 N/mm2
For steel Ym = 200000 N/mm2

4. The force being applied in the middle of the beam
For the example given this is 200 N

5. The calculation:

Deflection = (force (N) x length^3) / (Ec x Ym x Ixx)

For welded supported ends this is:
Deflection (Aluminium) = (200 x 500^3) / (192 x 69000 x 130208) = 0.01449 mm
Deflection (Steel) = (200 x 500 ^3) / (192 x 200000 x 8333) = 0.07813 mm

These are pretty close to the Fusion FEA results from Zeeflyboy (for fully supported end conditions):
25mm Alu deflection = 0.01471 mm
10mm Steel deflection = 0.06861 mm

If the ends are simply supported pin joints it makes a big difference:
Deflection (Aluminium) = (200 x 500^3) / (48 x 69000 x 130208) =0.0579 mm
Deflection (Steel) = (200 x 500 ^3) / (48 x 200000 x 8333) = 0.312 mm

So there you have it !

[Desertboy]

If you want to work it out by hand it is not too tricky for a simple shape like that.

1. The geometry of the part (how the material is arranged, into what shape). This is the cross section property and is called the second moment of area:
Ixx = (bd^3)/12
where:
b is the width
d is the depth
For aluminium part this is (100x25^3)/12 = 130208 mm^4
For the steel part this is (100x10^10)/12 = 8333 mm^4

2. The end condition of the beam (how it is restrained at the ends)
Ec = 192 for fully welded
Ec = 48 for simply supported (e.g. pin joint)

3. The material properties for the material being analysed (Young' modulus)
For aluminium Ym = 69000 N/mm2
For steel Ym = 200000 N/mm2

4. The force being applied in the middle of the beam
For the example given this is 200 N

5. The calculation:

Deflection = (force (N) x length^3) / (Ec x Ym x Ixx)

For welded supported ends this is:
Deflection (Aluminium) = (200 x 500^3) / (192 x 69000 x 130208) = 0.01449 mm
Deflection (Steel) = (200 x 500 ^3) / (192 x 200000 x 8333) = 0.07813 mm

These are pretty close to the Fusion FEA results from Zeeflyboy (for fully supported end conditions):
25mm Alu deflection = 0.01471 mm
10mm Steel deflection = 0.06861 mm

If the ends are simply supported pin joints it makes a big difference:
Deflection (Aluminium) = (200 x 500^3) / (48 x 69000 x 130208) =0.0579 mm
Deflection (Steel) = (200 x 500 ^3) / (48 x 200000 x 8333) = 0.312 mm

So there you have it !

Nice one! I always want work it out by hand if I can! I normally do my trig on paper then check it in cad so I already know what numbers I should be getting back.

[Zeeflyboy]

Nice one! I always want work it out by hand if I can! I normally do my trig on paper then check it in cad so I already know what numbers I should be getting back.

By hand gets super complicated super fast when considering more complex shapes and even more so when you are talking complex shapes with complex assemblies... That said always good to have a rough idea as a gross error check for computer simulations (even if nothing more than "hmm... that doesn't sound right"), after all garbage in = garbage out.

[Desertboy]

By hand gets super complicated super fast when considering more complex shapes and even more so when you are talking complex shapes with complex assemblies... That said always good to have a rough idea as a gross error check for computer simulations (even if nothing more than "hmm... that doesn't sound right"), after all garbage in = garbage out.

garbage in garbage out is why I haven't watched TV in 12 years lol ;) the only thing I've watched this year is Rick and Morty and South Park and I feel better for it ;) That said I have never learnt as much since I was a child as I learnt this year lol so so much reading and the brain is better for it.

I've found the router changes you someone along the way nothing scares you any more and you can solve any problem. Drooling over yours and routercnc's work and thinking you know what I can do that is also a healthy thing.

It's good to aspire to better things, although I do wish you'd stop upping the ante every 5 seconds I can't keep up.

Can't you do a bit of crap workmanship for once so us mere mortals can feel good about ourselves for 10 seconds lol.