New Machine Build Rigidity about cnc machine - Page 2

1. ## Re: Rigidity about cnc machine

Yeah we fully agree upon the ideal situation, which is why i got something that's 380 volt and a bunch of kilowatts.

Yet he could buy a kress spindle which has very little runout and weighs 1.4 - 1.5 kilo and 1050 watt. Costs over here (Netherlands) a 150-160 euro. Of course milling some mild steel with that takes forever and lots of cooleant, yet milling with very small diameter carbide tools is also possible. Of course that thing is bloody noisy and milling at such high RPM requires some good calculations on how to mill

The ebay stuff in general has a very bad runout - that's asking for problems on tool wear isn't it?

No one said If you want to mill cheapskate that you do not need to be a very good machinist to do it

Furthermore if he finds even a couple of hundreds of dollar too much for shipment of something great it's obvious that has to rethink to pay for 500+ kilo of steel and a lot more in concrete which he should to build such gantry

2. Could you recommend some good text books that focus on rigidity of milling machines? Any help would be greatly appreciated. I am about to start my senior year of ME and besides trying to tie together what I’ve learned from statics and Kinematics, I don’t have any examples of problems that could help me learn more about the general design requirements for milling machines.
Originally Posted by diepchess
Aloha!

Old thread - month ago you posted question - yet let me drop a few coins.

It's possible to buid a reasonable gantry that works on steel. The whole thing is about how much time you want it to cut such sheet.
The trick is to use a very lightweight spindel motor.

For example my spindel motor is very heavy. About 9 kilo. That means if i put it in a steel box, that box is soon 40 kilo if not more. If that box then hangs onto a long tube on a gantry, that means i need some very strong stiff tube.
Now next is to limit the Y-travel.

Deflection of your gantry bridge that moves over the machine is Newtons force times length of the Y axis to the power 3.

For example simplistic point deflection:

Say we have my 9 kilo spindel that in total has a weight of 60 kilo with motor and box and so on and then we add 50 kilo of Z-axis force it is using to hammer onto the workpiece.
Far over 1000 newton. Let's use 2000 newton. Now let's use a cheap tube as our moving gantries bridge that rolls over X axis. Let's use a tube of 100 x 100 millimeter thickness 5 millimeter.

First we calculate how strong that tube is: (100 ^ 4 -90 ^ 4) / 12 = 2,865,833
We then take length of our spanning of the gantry bridge say 3.5 feet => lets take 1000 mm.

I would then want to calculate with deflection (millimeters) = (2000 newton * 1000 ^ 3) / (48 * 210000 (steel) * 2865833) = 0.069 mm deflection
And we didn't even add the weight of the tube itself yet, though we did calculate as if all force is concentrated in the middle of the tube whereas it probably
is not entirely in the middle located. So in reality probably we measure 0.05 mm deflection.

That is a big problem in short if we get nearby 0.05mm as human eye will see that easily.

Now we do the same calculation for a cheapskate spindel of 1.5 kilo that we hang at a light structure of 4 kilo and because we mill with a smaller mill we also generate less force onto workpiece. Say maximum 10 kilo.
That's a weight of something over 15 kilo. Let's take double that in force. 300 newton.

So now if we do exactly same calculation with 300 newton, we suddenly are having a deflection that is nearly 7x less, or in short less than 0.01 mm.

Easy as a pie.

Yet of course milling time suddenly is very long as you can take away very little steel from the steel plate each second.

3. ## Re: Rigidity about cnc machine

Hi Yodda - follow the Maximus thread it details the design of a router but could be a mill, same process. In it5 there are references to info about machine building and calculations. Cheers Peter

4. ## Re: Rigidity about cnc machine

Yoddha - the huge difference is whether you want to build 1 machine for yourself - or whether you want to build a series of machines.
We can talk endlessly about the advantages of cheaply pouring pot metal - yet how useful is it if you do not have a professional factory that's doing a series production?
It needs casting, pouring the metal - then sit and waiting 3+ months until the pot metals internal strengths have settled - then have some giant machine to make it very flat for the linear rails.

If you just want to build 1 machine - go the epoxy concrete method. That's easy to do on your own. Yet pretty expensive for a heavy 10k+ kilo machine.
Doesn't matter what sort of machine you build - if you just build 1 - then epoxy concrete is going to win it in rigidity of everything and it's easy to pour couple of hundreds of kilo's.
Add a few percent of epoxy (not 10% - like 5% is enough already). Yet that epoxy is expensive.

You want to overdo rigidity. Obviously with any form of CONCRETE every wall needs to be very thick. Not 1 centimeter like steel - you want 10 centimeter or more at least.

For a gantry with say 600mm of X range and say 400mm of Y range it's easy. Just pour a block of 160mm thick and 1000x600mm or so. Then 4 rubber/alu legs at around 22% of the weight at both ends.
The moving gantry on top of it there is more choice.
Easist thought is an U shape structure. So 160mm in the middle and like 260mm height left and right. I give rough guesses here. Have an example worked out in CAD here if you're interested.
It's all about : what can you produce yourself, because someone else isn't gonna do it for you.
If you got 40k euro on other hand - you could buy one ready to go from epoxy concrete. In Zhermany some company sells a line there.
It is a lot of work though to build something yourself.

It's not only about rigidity it's also about how much sound do you want it to make?
You end up at water cooled milling motors then - but those are heavier.

Just do not touch aluminium extrusions - that's having a different expansion coefficient than the steel linears for example.
No math will work for you if you have different sorts of metal combined with each other.

Epoxy concrete has the same linear expansion coefficient like steel (it is very similar).

Notmal concrete is not possible by the way - it keeps working for years to come.

It's not only about rigidity and calculating that - that's the easy part - some sort of crabble notion there is enough with respect to milling.

You get onto a different league though if you want to design a machining center. That's a different league from building one (or a couple) at home in your own workshop.
Those machining centers in itself do not need to be 15000 kilo from a rigidity viewpoint seen - yet they are - and with good reasons.

5. Originally Posted by peteeng
Hi Yodda - follow the Maximus thread it details the design of a router but could be a mill, same process. In it5 there are references to info about machine building and calculations. Cheers Peter
Thanks Peter. I’ll look for the thread

6. Originally Posted by diepchess
Yoddha - the huge difference is whether you want to build 1 machine for yourself - or whether you want to build a series of machines.
We can talk endlessly about the advantages of cheaply pouring pot metal - yet how useful is it if you do not have a professional factory that's doing a series production?
It needs casting, pouring the metal - then sit and waiting 3+ months until the pot metals internal strengths have settled - then have some giant machine to make it very flat for the linear rails.

If you just want to build 1 machine - go the epoxy concrete method. That's easy to do on your own. Yet pretty expensive for a heavy 10k+ kilo machine.
Doesn't matter what sort of machine you build - if you just build 1 - then epoxy concrete is going to win it in rigidity of everything and it's easy to pour couple of hundreds of kilo's.
Add a few percent of epoxy (not 10% - like 5% is enough already). Yet that epoxy is expensive.

You want to overdo rigidity. Obviously with any form of CONCRETE every wall needs to be very thick. Not 1 centimeter like steel - you want 10 centimeter or more at least.

For a gantry with say 600mm of X range and say 400mm of Y range it's easy. Just pour a block of 160mm thick and 1000x600mm or so. Then 4 rubber/alu legs at around 22% of the weight at both ends.
The moving gantry on top of it there is more choice.
Easist thought is an U shape structure. So 160mm in the middle and like 260mm height left and right. I give rough guesses here. Have an example worked out in CAD here if you're interested.
It's all about : what can you produce yourself, because someone else isn't gonna do it for you.
If you got 40k euro on other hand - you could buy one ready to go from epoxy concrete. In Zhermany some company sells a line there.
It is a lot of work though to build something yourself.

It's not only about rigidity it's also about how much sound do you want it to make?
You end up at water cooled milling motors then - but those are heavier.

Just do not touch aluminium extrusions - that's having a different expansion coefficient than the steel linears for example.
No math will work for you if you have different sorts of metal combined with each other.

Epoxy concrete has the same linear expansion coefficient like steel (it is very similar).

Notmal concrete is not possible by the way - it keeps working for years to come.

It's not only about rigidity and calculating that - that's the easy part - some sort of crabble notion there is enough with respect to milling.

You get onto a different league though if you want to design a machining center. That's a different league from building one (or a couple) at home in your own workshop.
Those machining centers in itself do not need to be 15000 kilo from a rigidity viewpoint seen - yet they are - and with good reasons.
Thanks. There’s a lot of info in here.

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts
•