Ballscrews into New Zealand. Im buying, anyone want to get onto this boat?

Ive decided to upgrade my ballscrews and am buying some DFU4005 at 2700mm long. This is via LCL freight, out of China. And the freight is a minimum of 1m3 so I have space to spare.

If you want to hop on board and purchase at ex factory prices let me know, but do so asap, I am not putting any margins onto it you will just be sharing the freight costs which is minimal.

As for the quality of the ballscrews, its China, so is always a risk which I cant take responsibility for. Im taking the risk though, and always have. Also there is always damage risk but I have instructed my supplier to send in wooden crates.

Oh - I am also buying a couple of vices too.

Will be confirming my order in the next day or so - time is tight if you are keen.

Reply to this post if your interested. Oh - prices - I am paying about US$200 for one DFU4005 with BK/BF end machining and inclusive of the 30mm blocks. Its a one off chance - no high maintenance people with a thousand questions please ha

I randomly stumbled across this post when I looked at "today's posts".

Is this an upgrade for the router you mentioned in other posts? Are you committed to this deal?

4005 @ 2700mm long is a really bad choice IMO for a router. What motors are you planning to use to drive these ballscrews?

A lead of 40mm with gear reduction might work, but I'd have to look at the math.

40mm diameter ballscrews at 5mm lead will have ALOT of rotary inertia to overcome. Machine would be very slow with very low acceleration ability. Unless perhaps if it was paired with huge servos.

Please elaborate on what you're doing if you'd like more of an opinion. I thought about just not saying anything because you didn't ask for feedback, but I also realized that saying something here might really help you out.

Hi. Nope. Not committed at all. Definitely keen to hear what you have to say though. Sorry, the lead of these screws is not 40mm - you mean the pitch? These are diameter 40mm with a pitch of 5mm. Is lead the same as pitch?

Currently I have two SFU1605 at a span of 2m on the Y. They are being driven by a couple of Nema 24 closed loop steppers and tend to whip because of the size. Speed is not a problem because I machine alloy and want to be able to do so well. Max speed I need at most is 2000mmMin I have had this setup for quite a few years and its been really good.

The Nema 24, is slightly larger and better than the Nema 23 I have which is lifting a 12kg 4.5kw spindle on the Z. Also the Nema motors I have are P series. So my thoughts were, the Nema 24s I have will have plenty of torque to drive these 40mm 5mm pitch screws proven by the one Nema 23 lifting the massive spindle I have. The Nema 24s, if they work, I can always upgrade at a later date.

Oh - SFU 4005-5 is 5mm pitch there is a SFU 4005-4. I tend to not work out things correctly I just try them out. I did consider a 4mm pitch. What do you think?

Ok. That was a lot of burble on my behalf but there was no simple way to put it ha. Regards

Hey so Nic 77 what do you think? The 16mm ballscrews I have, have been ok up to now, I thought it would be ok to upgrade. Should I go smaller than 40? I just looked at span minimums and with the torque required, well, the Nema 24s I have are quite the little muscle machines.

Quote:

---

Originally Posted by boydage

Hi. Nope. Not committed at all. Definitely keen to hear what you have to say though. Sorry, the lead of these screws is not 40mm - you mean the pitch? These are diameter 40mm with a pitch of 5mm. Is lead the same as pitch?

---

I meant if it was something like a 3240 or 2540 you might be able to make it work with gear reduction from the motor, depending on the motor used and what you wanted to accomplish.

Pitch and lead are often used interchangeably, I have misused the terms myself sometimes. The lead is the linear distance moved per rotation of the ballscrew. The pitch x the number of starts = the lead. Most ballscrews we use only have 1 start, but multi start ballscrews do exist. Multi-start leadscrews are common.

For a single start ballscrew the lead and the pitch equate to the same thing. But the best term to use is "lead" as then there is no confusion about what you're talking about.

Quote:

---

Originally Posted by boydage

Currently I have two SFU1605 at a span of 2m on the Y. They are being driven by a couple of Nema 24 closed loop steppers and tend to whip because of the size. Speed is not a problem because I machine alloy and want to be able to do so well. Max speed I need at most is 2000mmMin I have had this setup for quite a few years and its been really good.

---

The way to fix the whip is to increase the diameter of the ballscrew or increase the lead of the ballscrew so that it doesn't have to spin as fast.

2000mm / min = 79 IPM. That's really slow for a machine of this size. You don't want to cut wood with it ever? If you're taking the time and money to upgrade the ballscrews you might as well pick ones that will give you the best upgrade performance as far as diameter and lead are concerned.

The amount of torque the motor has to put out to accelerate and decelerate the ballscrew in rotation is usually more dominant than the effect that the mass of your gantry has on the system, even with a fairly heavy gantry. It depends on many factors and the max acceleration used.

I have a document from THK that gives their rolled ballscrew moments of inertia for different sizes. The values will be almost identical for any rolled ballscrew.

A 16mm diameter ballscrew has a moment of inertia of 5.05 x 10^-4 or 0.0005 kg*cm^2 / mm length

A 40mm diameter ballscrew has a moment of inertia of 1.97 x 10^-2 or 0.0197 kg*cm^2 / mm length

0.0197 / 0.0005 = 39.4 meaning that the 40mm diameter ballscrew has 39.4 times more rotary inertia !!

Quote:

---

Originally Posted by boydage

The Nema 24, is slightly larger and better than the Nema 23 I have which is lifting a 12kg 4.5kw spindle on the Z. Also the Nema motors I have are P series. So my thoughts were, the Nema 24s I have will have plenty of torque to drive these 40mm 5mm pitch screws proven by the one Nema 23 lifting the massive spindle I have. The Nema 24s, if they work, I can always upgrade at a later date.

---

Unless you want to upgrade yourself to servos over 1300 watts (that's just a preliminary guess) it's not something easily fixed at a later date.

Quote:

---

Originally Posted by boydage

Oh - SFU 4005-5 is 5mm pitch there is a SFU 4005-4. I tend to not work out things correctly I just try them out. I did consider a 4mm pitch.

---

The 40 should be the diameter and the 05 should be the lead for a 4005 ballscrew. At first I was wondering if the -5 and -4 could be referring to the number of starts and if these were multi start screws but a quick online search makes me think that the -5 and -4 are referring to the ballnut style. For example, the -4 might not have a flanged ballnut. That's something to ask the seller.

Quote:

---

Originally Posted by boydage

What do you think?

---

Off the top of my head, something like a 2525 ballscrew with 2:1 belt driven gear reduction might be your best option.

If you don't want to hassle with the gear reduction then 2510, 2020, and 2525 are options. I don't know at this point which would be best for you.

If you were going to upgrade to 750W servos for a router of super performance then I might even say something like 3220 to 3240 with 3:1 to 5:1 belt driven gear reduction, but I would need to look at the critical speeds and do some math to give a better answer.

The point is that the 4005 is just a really really bad choice.

Quote:

---

Originally Posted by boydage

Hey so Nic 77 what do you think? The 16mm ballscrews I have, have been ok up to now, I thought it would be ok to upgrade. Should I go smaller than 40? I just looked at span minimums and with the torque required, well, the Nema 24s I have are quite the little muscle machines.

---

Do you have a torque vs speed graph for these motors?

Yes, you need to go smaller than 40.

Are you thinking of upgrading to servos? The ballscrew and the motor need to be paired for best performance. Are you willing to do any belt driven gear reduction or is that too much of a hassle?

I don't believe a 16mm diameter ballscrew is a good choice for a length of 2700mm.

I am laughing and laughing at myself! Well. On the inside. My machine is something that I built without any form of knowledge. And I get good results because the actual build of the machine is tight tight tight.

I really did not even consider the inertia of the ballscrew itself and that makes so much sense. Thankyou for writing that.

Ok so the last large number of years I have gotten away with the 16mm ballscrews - and they are actually at 2300mm full length. And they worked well. But they are def too long for size.

I was going up to 2700 to future proof if I decide to build a larger machine ( I have some 30mm linear rails somewhere) Minimum length I would like to go is 2600mm to give me a useable ballscrew of 2400mm. But I also just would like to get my machine going for now the best way I can without too much more work. I have just spent several months on the rest of it. Just want to use it now.

I have DIY CNC maintenance fatigue. Yep. Done enough for this year. Done quite a lot.

Speeds I am absolutely ok with. I can jog around on 2400mmMin. Hey - if I can improve it I will that would be nice.

I would prefer to not be gearing my motors. I can absolutely change my order for screws its only in "talks" with the supplier.

Ok. What are your thoughts of what I would get away with, no gearing of motors, maximum length above 2300mm, with the torque chart attached?

If I need to stay with 2300mm I will, although if I can go longer I will also. Hey - I was just about to buy some massive 2700mm 4005 screws that would not have worked. Oh - correct about the part number BTW it was the size of the ballnut flange.

Regards

For the 2700 mm ballscrew, what is the bearing to bearing distance between free and fixed ends?
For the 2300mm ballscrew, same question?

Do you have a link to a datasheet for these motors? I need to know the rotor inertia of the motor, it should be on the data sheet. If you can't find one, I can guess, it's not a huge deal. You can probably find it on eBay.

What is the approximate weight of the gantry including Z axis and all parts that move with the gantry. Doesn't have to be exact.

This gantry is driven by two motors and ballscrews, one on each side, is this correct?

What acceleration value do you have inputted in your control software right now? Are you happy with this value? What control software are you using, just curious, is it Mach 3?

I'll throw those in my spreadsheets and give you a better opinion when I have some time.

Wow I really appreciate this. I know it takes time out of days.

Ok.

2700mm bearing to bearing is 2552mm

2300mm bearing to bearing is 2200mm

Datasheet attached,

Gantry was 44kg before the upgrade now is 58kg

Yes, two motors one each side.

Oh, I would like to get some more speed out of the machine. I am happy with its repeatable accuracy though very much. Its spot on and very accurate. I have attached a screenshot of the motor tuning settings. Yes, I would like to make it go a bit faster on the Y. Previously I had Nema 23s that would miss steps. Now though with the closed loop Nema 24s I can probably push them a whole lot more.

Really looking forward to your reply. Like I said, only went to 2700 for the future if I build a larger machine.

Have you thought about changing to rotating nut setup on your existing screws?

Quote:

---

Originally Posted by pippin88

Have you thought about changing to rotating nut setup on your existing screws?

---

I have, I dont want too though I do prefer to have minimal mechanics involved.

I actually wound my machine up today to 3750mm Min. It was fine, except for the whip on the Y. I think there is a small bend in it. Was funny watching the "whip" get whipped into a more resonant higher frequency (appeared) as the length of the screw shortened. How do I explain that - ummm - plastic ruler wobble it with 3/4 of it over the edge of a table. Then whist its wobbling, pull it into the table. Whats the engineering term? Dunno.

I want to keep with the plan. Upgrade the ballscrews to a larger size. Just it appears 40mm is to big.

Question - If I went for a 10mm pitch, how does that go for resolution and accuracy? I like the 5mm because it is finer. Or am I completely out of wack and the Nema motors are sweet with a 10mm pitch?

Thoughts?

40mm dia ballscrews are huge.
Good if you have an industrial vmc.
I think about 32mm is as big as you want to go.

Quote:

---

Originally Posted by v2tiller

40mm dia ballscrews are huge.
Good if you have an industrial vmc.
I think about 32mm is as big as you want to go.

---

Well I WAS thinking.... that if I used 40mm screws, that I could shim under the blocks with say some 0.5mm shim stock. And that would take some of the weight of my gantry off the linear rails. win win. hahaha.

Ok well I am pleased I posted here even if it was off topic. I just thought, I have really powerful Nema 24s, no problem. Imagine it, in my head it was a great design. NIC 77 has made me really think, and I will wait for his response. Yes, I agree. I was looking at a piece of 20mm screw I have here and it is quite a lot more than the 16. Gee, 16mm hs been working for years, I am beginning to think 25mm will be fine.

I will struggle with one thing though and thats moving away from a 5mm pitch if that is what NIC advises. I am machining parts out of alloy, not too small and some of them are 46mm thick/deep, that have a sliding fit. Ie my sliding fit is 0.005mm. If I went too pitch 10 thats half the resolution. I am currently running ny nemas at 400 steps per mm. Have done since day one. 10mm pitch, I would then be changing the driver to something like 4000 steps rev and...... oh, stays the same 400 steps mm.

What about backlash on a Chinese ballscrew. Is backlash increased by a larger pitch? Currently the screws are C7 and barely any backlash at all.

2552 mm bearing to bearing with an end fixity of "C" and a root diameter of 21mm (25mm diameter ballscrew) gives a critical speed of 573 RPM according to the Nook calculator

https://www.nookindustries.com/resou...ed-calculator/

for a 25mm lead that would give a top speed of (573 x 25) / 25.4 = 564 IPM
for a 10mm lead that would give a top speed of (573 x 10) / 25.4 = 226 IPM
for a 5mm lead that would give a top speed of (573 x 5) / 25.4 = 113 IPM

2552 mm bearing to bearing with an end fixity of "C" and a root diameter of 27mm (32mm diameter ballacrew) gives a critical speed of 737 RPM, which is only 164RPM more than the 25mm diameter.

2552mm bearing to bearing with an end fixity of "C" and a root diameter of 13.5mm (16mm diameter ballscrew) gives a critical speed of 368 RPM.

The all up gantry weight sounds light to me. I just went with 100 kg or 220 lbs. It doesn't make much difference.

Your acceleration is currently set to 400 mm/s^2

400 mm/s^2 / 9810 mm/s^2 = 0.04G which is what it says in the Mach 3 screenshot.


From the graph you provided I used these values

RPM Torque (Nm)
1 2.1
75 1.75
150 2.1
225 2.05
300 2
375 1.9
450 1.8
525 1.7
600 1.55
675 1.7
750 1.62
825 1.585
900 1.55
925 1.5


Rotor inertia = 690 grams*cm^2 = 0.000069 Kg*m^2
Ballscrew inertia = 0.00301 kg*cm^2 /mm = 0.000000301 kg*m^2 / mm
Bascrew Length = 2700mm, note added a bit extra to this as a wag for any couplers
Gantry Weight = 100kg / 220lbs
Assumed a ballscrew efficiency of 0.9

Did not include preload torque, bearing friction, or reduce the available torque from the motor graph provided. In real life, your results will be less than what the graphs show, the are simply my opinion and offer a point of comparison.

I cut off the following graphs for 25mm ballscrews at 573 RPM due to critical speed.

25mm diameter ballscrew at 0.07G / 0.15G / 0.25G

Attachment 448826

Attachment 448828

Attachment 448830

Increased the gantry weight to 200kg at 0.15G in the graph below to show the difference that this has

Attachment 448832

Graph below is 40mm diameter at 0.04G, which is very low acceleration. The 4005 could not have possibly worked out.

Attachment 448834

At 0.1G and 40mm diameter below

Attachment 448836

Below I went with a 3:1 planetary gear reducer that has an interia as seen by the motor of 0.000025 kg*m^2 which I believe is not atypical. Belt driven gear reduction would give a bit better performance. Just went up to 1050 RPM on the motor for this graph. Back to 25mm diameter ballscrews.

Attachment 448838

And below I went with some JMC 400W servos and 3:1 gear reduction on 25mm lead 25mm diameter ballscrews, for comparison

Attachment 448840

If you ask my opinion, I'd go with some 2525 ballscrews in your position. You could use 2510 but would be limited by whip and still have marginal top speed ability, even though it would be better that what you are currently using.

If you went with the 2525 you are leaving yourself the option of upgrading with gear reduction, either low backlash planetary or belt driven at some future time.

Using 5mm lead on a machine of this size is a mistake IMO. Also 5mm lead are better suited to servos than steppers in general on smaller machines.

Like I said, the actual results you will get will be a bit worse than the graphs, but can not be better than the graphs unless the motor can put out more torque than the motor torque graph predicts which I think is unlikely, but the opposite may be true.

Quote:

---

Originally Posted by boydage

What about backlash on a Chinese ballscrew. Is backlash increased by a larger pitch? Currently the screws are C7 and barely any backlash at all.

---

I have no idea. China is a big place and there are so many different manufacturers there.

The ballscrews from China can sometimes arrive bent also. I ordered some from BST automation on aliexpress and the long screw I ordered was bent, but they replaced it for me free of charge, so that was great of them, some suppliers won't do that.

Also, TBI ballscrews might be an option but they cost more.

You could ask for quotes from both places and see where you end up.

Thankyou. Coming from a different industry I know you spent some valuable time on this. And I have read several other posts you have had some quality and obviously trained input into as well. Infact I ended up with quite a number of browser tabs open. I am the person who trained in building and maintaining the various pieces of machinery in this world not having much time in the design industry its always good to see the other side.

I am going to take the advise and will post my upgrade build in a few months once I manage to get everything in my workshop.

Far out, about 10 years ago I was offshore in Asia somewhere. I needed to brush up on my electronics and automation so was looking for a project. I was going to automate my house and only stumbled on CNC by chance.

Anyways, once again, thanks. Regards from NZ