[bmolinar]

hello all, thanks for having me in the forum..
I am totally new to this hobby and I want to build a router for me and my sons so I can teach them my trade, and we can have something to do together. I am totally in the dark when it comes to what size, pitch or type of ballscrews to buy so Im hoping someone here will offer some guidance and help me get things sorted out. I want to build a 4' x 4' router. I know that I want to be able to machine at upwards of 450ipm. I was thinking either 20mm or 25mm ballscrews but I dont know what pitch I would need or how to calculate things to make sure Im on the right track or not with my selection. Any help would be appreciated
Thx
Ben

[awerby]

I'd go with the 20 mm screws. The length isn't enough to get much "whipping" going, which is the only reason to use fatter ones. The extra mass of the wider screws can work against you when you're trying to change directions, since they get a lot of momentum going. Figuring out the pitch is dependent on the motors you're using. Servos go pretty fast without losing torque, so you could use relatively fine-pitch screws and still hit your target speed with them. But steppers lose torque as they go faster, so if that's the type of motors you're using, coarser-pitched screws would be advisable. The other thing to consider with ball screws is precision. All ball screws are not created equal, and some are just good for opening garage doors, etc. Look at the specifications, and just choose the ones that can hit the tolerances you're looking for, over the length of your typical parts. A ball screw's "thread drunkenness" is specified as +/- x units per unit of length, so be sure to do the math.

[peteeng]

Hello Ben - 20mm or 25mm diameter ballscrews are too big for a typical 4ft machine that uses nema23 motors. 16mm is fine. The rotational inertia of large ballscrews is considerable and can take up must of the torque of the motor if your not careful just to turn them, especially at speeds your talking. 450ipm is 11,250mm/min. This is incredibly fast for a hobby machine, even a small commercial machine. To get to this sort of speed in 4ft is doable but will require large motors to be able to achieve the accelerations required. Why do you need 450ipm plus feed speed? Now if your using steppers they peak out torque at say 1000rpm so 11250/1000= 11.25mm so you need a screw pitch of 10mm or bigger say 16mm. But at 1000rpm steppers have little torque for cutting. So if your serious about 450ipm you will need AC servos I expect as they have a flat torque output from 0-3000rpm. More info on what you want to machine and do would be good to take this further. Peter

[pippin88]

https://www.engineersedge.com/calcul...crew_15635.htm

16mm screws will whip at ~1600rpm
Ok for steppers where you might be around 1000rpm. Not ok for higher RPM servos.

Do not get 5mm pitch screws if you are using steppers.

I have 1616 (16mm diameter, 16mm pitch) screw for a 1300mm travel axis. Works fine for stepper use.

[joeavaerage]

Hi,
be very careful about choosing the size of the machine. I note that you have nominated 4' x 4'....and that is perfect doable, I would have thought 20mm diameter screws appropriate
for that size machine, especially if you want to get to your target speeds. The size choice has extremely important consequences on the cost and rigidity of the overall machine.
It will determine amongst other things the diameter of the screws, and that in turn will determine the size of the steppers/servos. It will determine (most likely) the style of construction,
say welded steel verses aluminum extrusions etc. All these determine the price.

As a rule of thumb I would say that to double the size of a machine is to commit to five, maybe ten times the cost. The cost escalation with size is extreme....do not underestimate
that fact or you'll end up in the situation of a half built project having 'sold off all your children to slavery' only to realise you are going to need a lot more children to raise the funds to complete the project.
How do I know????

I would suggest you consider again what it is you are hoping to achieve. Being able to work and share with your sons is a very laudable purpose. The question then becomes what is the smallest machine
that would satisfy that requirement and be useful? I have no doubt peteeng will chime in that it makes sense to build small and cheap to start with, learning as you go. Then attempt a larger and more ambitious
project. Many of the lessons that CNC has are often hard earned and come as a result of failure to meet ones goals...my own CNC experience has plenty of disappointments and outright failures
to lend authenticity to my convictions!

May I suggest that you consider a smaller machine. Lets say the machine has 600mm x 600mm travels. Additionally imagine that the X axis is unobstructed so that you could place a half sheet
(600mm x 2400mm) in the machine. Admittedly you'd only be able to work on a 600mm x 600mm section before sliding it along in the X axis and dealing to the next section and so on.
I understand that you might find it inconvenient to operate in such a manner but according to my rule of thumb above such a machine will be something like 1/5th the price of a 4' x 4' machine.

A 2'x 2' machine would be a slam dunk for 16mm screws, either very good 23/24 size steppers of 23/24 size servos if speed demands, and viable with aluminum extrusions. I'm not a fan of aluminum
extrusions, they are a stiff as a cooked noodle.....but in shorter lengths, say 2' x 2' become very attractive. By reducing the size of the machine you have come right into the 'sweet zone' with regard
to availability and cost of materials, parts and components and techniques.

Craig

[ger21]

If you are using steppers, you want between 10mm and 20mm pitch. With servos, I'd go 20-25mm pitch, and use a belt reduction from servo to ballscrew.

[joeavaerage]

Hi,

I'd go 20-25mm pitch, and use a belt reduction from servo to ballscrew.

Why? Use course pitch ballscrews only to then use a belt reduction on servos....complication for what gain?. If you used 5mm or 10mm pitch screws
you could direct connect them to servos. The high speed of the servos will ensure that you can still get decent rapids while the (somewhat) finer pitch allows
for high thrust and acceleration without the need for reduction.

My machine has 32mm diameter screws of 5mm pitch. At maximum tune I can get 25m/min rapids and 0.25g acceleration when direct coupled to 750W servos.
The combination means I have a stall thrust (without servo overload) of 3kN and at overload 9kN. Speed, thrust and acceleration....what's not to like?

Craig

[phomann]

Hi,
Just for reference, my YM84 pick and place machine has 750w ac servos on the X and Y axes. The ballscrews have a 20mm pitch and have around 550mm travel. I’d have to check but I think they are 16mm diameter, but may be 20mm.
The servos max out at 3000rpm and have a 1:1 belt between the motor and ballscrew. This gives a max rapid of 60m/min.
The original 750w DC servos maxed out at 3600rpm (72m/min)

Cheers
Peter


Sent from my iPhone using Tapatalk Pro

[ger21]

Why? Use course pitch ballscrews only to then use a belt reduction on servos....complication for what gain?.

So you don't spin a 60" long 16mm screw at 3000 rpm.

[ardenum2]

So you don't spin a 60" long 16mm screw at 3000 rpm.

1000 RPM + pitch 20 (3:1 belt reduction at best) vs 3000 RPM + pitch 5 (direct drive).

I might be mistaken but...isn't direct drive max m/min in this case lower?

[davida1234]

1000 RPM + pitch 20 (3:1 belt reduction at best) vs 3000 RPM + pitch 5 (direct drive).

I might be mistaken but...isn't direct drive max m/min in this case lower?

What Ger21 is saying is that the direct drive needs to spin the screw at 3000 rpm vs just 1000 rpm for reduction.

How is the direct drive screw rpm lower than the reduction rpm?

[joeavaerage]

Hi,
my suggestion was that OP consider a 2' x 2' machine for which 16mm diameter screws would be a good fit. At that length they could be spun
at high speeds if so desired. The point being that by making a smaller machine not only does the cost go down bigtime, but the choice of components
like ballscrews go up. At 2' x 2' any reasonable pitch could be made to work whereas if the machine is 4' x 4' then 16mm diameter screws would need to be 20mm
pitch to avoid likely whipping. Such screws are available but a much reduced choice.

My experience is that every time you think 'if I just had that extra 100mm of travel.....'. When you analyse that thought then you realise that 'little bit extra' costs
a fortune. You have to match the size of the machine to your budget. The thought process is to decide 'between the size I want verses the size I need (and can afford)'.

Craig

[phomann]

What Ger21 is saying is that the direct drive needs to spin the screw at 3000 rpm vs just 1000 rpm for reduction.

How is the direct drive screw rpm lower than the reduction rpm?

They are referring the speed of the axis, not the ballscrew.

It’s because of the ballscrew pitch. Turning a 20mm pitch ballscrew at 1000rpm will move the axis faster than turning a ballscrew at 3000rpm

The maths;
For a 20mm pitch screw, at 1000rpm, it will move the axis at 20m/min.

For a 5mm pitch screw, at 3000rpm, it will move the axis at 15m/min.

So, turning the 20mm pitch screw at 1000rpm moves the axis faster.

The advantage of the coarser ballscrew is that it rotates slower for the same feed rate, compared to a fine pitch ballscrew. This reduces the amount of whipping the screw will experience.

The down side is that the resolution of the axis will be less on a coarse screw, compared to a finer pitch screw.
That said, with the 3:1 reduction on a 20 pitch screw, the resolution will be slightly worse than the direct drive 5mm pitch screw as the 20mm screw is effectively a 1:4 gain over the 5mm pitch screw.
These days with the high count encoders,its not an issue.

I’m not a that is clear, but it is what it is.

Cheers
Peter


Sent from my iPhone using Tapatalk Pro

[pippin88]

A reduction drive (belt or gear) also changes the torque

A 3:1 Motor:Screw drive will result in 1/3 the RPM (eg motor RPM 3000, screw RPM 1000) but will triple the torque (e.g. motor torque 4nM, torque at screw 12nM.
So often a smaller / less powerful motor can be used.

There are also changes to inertia, the physics of which I do not understand.
A reduction drive substantially reduces the inertia that the motor 'sees'.
Rotational inertia of large ball screws is an issue - requires a lot of torque to get the screw moving.

[peteeng]

Hi Pippin et al - There are two types of inertia that are relevant in machine design. 1) translational and 2) rotational. Inertial forces are forces that oppose accelerating masses (inertial forces has something to do with mass properties like gravity is related to mass. Its something to do with how subatomic particles react to motion) . In the case of rotating machinery these are resisting torques. When you are in a car and go around a corner you feel the centripetal acceleration or the resulting inertial force pushing you to the outside of the curve. Inertial forces always resist change so if your linearly accelerating northward the mass inertial force is southward. If the shaft is accelerating Clockwise the inertial torque is CCW. At steady velocities there are no inertial forces present. By using a reduction drive you increase the torque available from a smaller motor that has a smaller inertia. So overall the system inertia potentially decreases. The inertia of the ballscrew does not change, unless a smaller diameter screw is used... If you used the same size motor that you would use for a direct drive and the same screw diameter the inertia of the system does not change. But you will have more torque if its a reduction drive.... Until the target velocities and accelerations are specified we can't design the drive system.

Laser cutters are very fast and use belts to achieve high accels and vels. To design these the belt is considered as a linear motion mass and the pulleys and motor are the only rotational motion... Peter

[joeavaerage]

Hi,

Laser cutters are very fast and use belts to achieve high accels and vels. T

Peter Homanns pick-and-place machine is pretty damn fast at 60m/min too. The big determinant is weight, or rather momentum. It takes a lot of energy to get an object up to
60m/min..or 1 m/s, and it takes a lot of energy to slow the damn thing down again. I imagine the pick head on Peters machine cannot be overly heavy or the acceleration/deceleration
forces would be extreme.

Craig

[pippin88]

https://www.motioncontroltips.com/ho...from-the-load/

https://www.novantaims.com/technolog...ystems-cont-2/

https://www.chiefdelphi.com/t/reflected-inertia/37414/4

Reflected inertia changes to the square of the gear ratio (I think)

Important for inertia ratio

[peteeng]

Hi - The term reflected inertia is a poor way to describe the physics. But its a common term. Similarly with "inertia ratio". Once the system is defined all the masses and accels become constants so the total eqn can be simplified into an "inertia ratio". This becomes a short hand method. I kind of accept the term reflected load vs reflected inertia. "Reflected" implies there's a forward load and a back load. There is no fwd inertia load or un-reflected load just applied loads (body loads) and the inertia load... various frictions and air resistance to name the common ones that act on machines... Peter

I'll firm my stance on the term "reflected" its a lay term attempting to simplify the inertial load interpretation. But there is no "reflected" load or inertia. This is a similar circumstance to interpreting centrifugal forces and centripetal forces. Technically there is no centrifugal force.... similiarly there is no reflected inertia or load....

[phomann]

This is a short video when I was testing the new motors. If I recall correctly, I was driving it from Mach4 and an ESS.

https://www.instagram.com/p/CIejWV-l...RlODBiNWFlZA==

Cheers
Peter


Sent from my iPhone using Tapatalk Pro