[NP-rosbud]

Hey, i know the information if in here scattered amoungst different posts but im still going to do it again... i am sorry though.

Im working on selecting a ball screw for a conversion im working on.

Id like to i guess understand the cause and effects of the different veriables involved.

Diameter, accuracy class, pitch, others i dont know about, with formulas.

I have estimates on dimensions
and mass on the computer at home i can add later.

The orginal mill is a Optimum BF16V, very small, it is heavily modified though, most notably a replacement saddle that is about 3 times as long, and much heavier. The original lead screws were D=14 mm. Given it is so small i may be very restricted with diameter, mainly due to the nut. My current workng estimate is a D=12mm to make it fit.

Some of the kore specific questions i suppose i have are what pitch would work best, in my case the priority is precision and repeatability.

What methods have people used to estimate weight allowance for materials, workholding, ect.
How well did it work, and more importantly when did it not work.

Similarly methods for estimating max speed.

How much benifiet is there in trying to find a D=14mm (non standard) ball screw or even 16mm, or would D=12mm be possible?

Does anyone know of good products or brands for low profile or extra small ball screw nuts to help fitting them in?

And similarly how are people finding low backlash single nuts as apposed to double, and can you recommend a product. Links incouraged.

There's alot there. If you made it this far thank you!

[joeavaerage]

Hi,
the problem with ballscrews is that the nut tends to have a much larger outer diameter than a plain ACME nut and you may struggle to fit them in.

If the original screws are 14mm in diameter what makes you think 12mm would be adequate? I would say 16mm or even better 20mm.
Max speed is about how fast you can spin the screws before they start to whip, so its all about diameter and length. In the first instance decide on the biggest
and stiffest screws you can fit in there, THEN determine the rotational speed to attain some reasonable speed, and check the calculations that its not going to cause the
screw to whip. This is usually not the case, but if you go for long, skinny screws you could run into trouble.

Next thing to decide is the grade of screw. Most cheap Chinese CNCs have rolled C7 screws which are good to 50um per 300mm. Downside is the cyclic variation, uncontrolled in C7
grades and could be 35um or more in any one revolution of the screw.

The next step up is ground ballscrews C5 or even better C3's Beware there are plenty of Chinese manufacturers claiming C5 and C3 which would scarcely pass C7...don't be fooled.
C5 is good to 18um per 300mm but the cyclic is much better at 8um per rev. C3's are better again.

If a C7 screw costs $100, a genuine C5 will cost nearly $1000 and a C3 more again. So it makes a huge difference to your budget.

For my new (two years old) mill I bought near new second hand 32mm diameter, 5mm pitch, double nut BNFN, C5 ground screws by THK, a world class manufacturer and paid a small fraction
of new price. I could not have afforded to build my machine otherwise. If you are careful and patient you too might find a good deal on some genuine high quality screws second hand or new old stock.
You have to be realistic, your design has to be a bit flexible to use the parts you can actually find verses those wish are a perfect fit, and you have to be patient. Try:

https://www.ebay.com/str/industrialp....m47492.l74602

Craig

[NP-rosbud]

So. As a quick update

Mass of the table assembly (X-axis) as according to my CAD model
1.9 kg, not including ball screw assembly mass

Mass of table and saddle (Y-axis)
8.6 kg + the mass of the x axis ball screw assembly

And I dont have an accurate reference for the machining head (z-axis) ar this time. But for a point of refference i did find an reference from an equivalent model come in at 29.0kg but i believe the BF-16V may come in a bit lighter, also, i uavent thought it all the way through but i anticipate i will end up adding kind of counter wieght on pullies or a gass strut.
I have a lot more room on the Z-axis and could much more easily use a bigger diameter ball screw, but i do worry about the inertia of the larger diameter.

These are all likley to be under estimates however and x and y will obviously have to deal with vices material tool holding, fixturing ect. Again any recommendations for estimating would be awesome.

I have an inital calculation for the length of the travel of the x axis, current the length of the entire ballscrew is 440mm and further dependent on things like bearing blocks.
I dont have a number for Y and Z respectively but is might guess ~~250 and 500 respectively, hopfulky less.

This is a very small mill so honestly dont mind if my rapids aren't amazing.
As for feed rates was talking to a friend and they suggested 1000mm/min would be ample. Im flexible if its under. It's not a super heavy duty mill anyway, and when dealing with pitch I'd prefer fidality and accuracy over speed.

And some other points of reference about the mill.
https://youtu.be/AU_nLOP7YrA?si=86Er3To4aOlZkoun

https://www.machineryhouse.com.au/m649

https://www.reddit.com/r/CNC/comment...version_album/

And a couple of ps. My background is predominantly mechanical, taining in machine repair and reconditioning. electricals are still a bit foreign.

Ill be mounting mild steel supports as shown and attatching to a mild steel sencodary base. Very seriously concidering epoxy granite filling the base and column which wont effect the weights of moving components.

I have also scraped in the ways gibs, reference surfaces, and alot else. Bronze gibs, turcite on the short componets ways. One shot lube.

Havent dissmissed useing linear scales for feed back, currently runs a dro, might use the scales from that if it is possible(?)

I dont know that this effects much but i also intend on using three indervidual MPG handwheels to simulate manual use, as a time same, and idiot proofs me against having it set to the wrong axis and crashing. Might have someone smarter then me write script to identify manually imputed "continuos" motion with a high sample rate and have it output a true continuous speed with assigned rapids and feeds.

And your standard e stops, limit switches and sim.

Again, Thanks all!

[jaguar36]

I know we all like to do lots of analysis to figure out the optimal parts and design, but in reality unless your budget is such that you can just go to Rexroth or whoever and spec something out, you're going to be severely limited by what is available. You can go the used route and find something that is kinda close and make it work, or you can go the new Chinese route, in which case you're going to be limited to which of the few sizes/pitches is readily available. Rather than try and do a whole bunch of analysis to figure out what is optimal, I'd instead start by looking at whats in your budget. Anything beyond the few 'standard' sizes is an order of magnitude more expensive. For instance more accuracy is always better, but to jump up in accuracy class you're looking at a massive price increase that likely is out of your budget, or if it is in your budget, the money would be better spent elsewhere. It doesn't make sense to put a $1000 ball screw in a mill that wasn't much more than that to start with.

So rather than try and figure out what the optimal size/accuracy/pitch, see what is available that will fit. I think you will find that you have very few choices

[joeavaerage]

Hi,

So rather than try and figure out what the optimal size/accuracy/pitch, see what is available that will fit. I think you will find that you have very few choices

Excellent advice. When I was designing my machine, I knew roughly what I wanted, but I could not finalize anything until I had settled on motion parts as I knew that
I would have to go second hand or new old stock to have any chance of getting the quality of parts that I wanted at a price I could afford.
Eventually I found matching parts of the quality and size (approximately) that I wanted, and got them. Then and only then could I actually design the rest of the machine.

Ground C5 and/or C3 ballscrews are often the most expensive parts of any build, maybe only exceeded by the spindle or perhaps the servos.

Look around EBay and see what you can find that might work for you, and then do a few calculations to ensure that the physics of them matches your needs.

Craig