Chinese ball nuts & screw on Ebay

[VooDooMan]

diycnc,

If you can some pictures would clear a lot of questions up, If you don't mind?!

Thanks!!

[wildwestpat]

Hi diycnc

Sorry made a few typos the worst was that the 5201 dimensions are 12 x 32 x 15.9 mm and as such will fit the standard 12 series mounting blocks leaving a small "clamping" gap.

The SKF documents are worth a quick read as they give a clear guide to the choice of bearings and the loading factors with some neat on line calculators to save on the maths.

Hope this helps and sorry about the typo as I think the 5201 could provide you with a quick cheap solution.

Have fun and stay safe with CNC


regards

Pat

[wildwestpat]

Hi diycnc

Since we are both on a value for money track here is some more info I uncovered whilst looking for cheap 8 mm ball screws to fit the Seig C3 mini lathe and its look alikes.

NSK are a well respected supplier of precision ball screws and here is a direct quote from their sales blurb.

"Though conventional support units use angular contact
ball bearings, support kits for the RMA Series use two
flanged bearings. They have no axial play and provide lowcost
alternatives."

Link to full paper from their Motion and Control mag is :-

http://www.tec.nsk.com/weblibrary/Li...l%20Screws.pdf

Since the RMA and RMS series of minature ball screws are C3 ground you can draw your own conclusions about the use of precision bearings. I am begining to believe that the CNC professionals are not cash strapped - may be they soon will be as the global economic situation dictates the bottom line and squeezes out all excess costs! The need for high speeds - long periods between maintenance - quick back into service time - are all requirements that push up the cost of the bearing. There are a lot of other factors but these I think are the main ones.

The first step is to determine what you are wanting from the end machine. In my case it is week end ocassional use with reasonably low by CNC standards travel speeds. Since I intend to use stepper motors the load imposed on the bearings will be higher as the torque is applied in impulses even using microstepping. This means that the bearing dynamic loading needs to be conservative if the bearings are to last well.

When mouting the bearing the housing compresses the outer ring and the shaft expands the inner ring and these two effects combine to reduce the running clearances. Bearing manufacturers make several types of bearing with both reduced and increased internal clearance for different applications. For low cost we need the normal internal clearance grade and to stay within the manufacturers limits for the bearing. The link to the SKF site gives the necessary calculations and you can explore the more exotic bearings from there.

If we had the cash then the choice of bearing becomes wider but the law of diminishing returns rapidly sets in so unless you need high speeds and long time between replacements think vwery hard about the very steep cash to value slope to perfection!

As a side thought are you sure your bearings are shot. Some makes of angular contact bearing have a lot of slop in the reverse axial direction even under finger pressure. Unless you have run the bearings mounted the wrong way round you may be lucky and find that correcting the mounting so that they point in opposite directions is all that is required. This is well worth a try BUT and it is a very big but any bearing is easily damaged and force can only be safely be applied to the part of the race that is being extracted. i.e. the outer shell to extract it from the bearing mounting block or the inner to shift it off the shaft. (This comes back to the need to consider the effects of force fitting on the bearing inner clearances!!!!!)

I paid £8.25 for the 5201 2RS for an existing 16 mm ballscrew and found the NSK 8 mm ballscrews on ebay for less than £50. So there is value out there!

Good luck in your quest for value

Pat

[15mgtar]

Hi guys, I'm about to purchase from ebay 1 anti backlash ballscrew RM2505-1600mm-C7+1set BK/BF15 <<< is the nut really have anti backlash system??? I'm just double checking with you guys the one that had bought similar items from linearmotionbearing guy named Chai.

[VooDooMan]

15mgtar, they state its zerobacklash, But unless its over sized balls I would say forget it.
I would just buy 2 ballnuts per axis and make your own preloaded nuts.

That is what I plan to do.

[SCzEngrgGroup]

15mgtar, they state its zerobacklash, But unless its over sized balls I would say forget it.
I would just buy 2 ballnuts per axis and make your own preloaded nuts.

That is what I plan to do.

You seem to be convinced that it's not possible to have zero backlash without double nuts. That is simply not true. If you look at the best ballnuts, you will *never* see double nuts, for the simple reason that spring-loaded double nuts are less stiff in one direction, due to the springs. You will *never* see double-nuts on a modern production CNC machine. The best nuts look exactly like single nuts, but are manufactured with a slight irregularity in the lead that ensures zero backlash. They have the added advantage that they take up a great deal less space than a double nut assembly.

Regards,
Ray L.

[VooDooMan]

You seem to be convinced that it's not possible to have zero backlash without double nuts. That is simply not true. If you look at the best ballnuts, you will *never* see double nuts, for the simple reason that spring-loaded double nuts are less stiff in one direction, due to the springs. You will *never* see double-nuts on a modern production CNC machine. The best nuts look exactly like single nuts, but are manufactured with a slight irregularity in the lead that ensures zero backlash. They have the added advantage that they take up a great deal less space than a double nut assembly.

Regards,
Ray L.

Hey Ray,
Well I am quite convinced.
A. These are not the best ball nuts to start with.
B. I am no magician and thus I can't tell from the sellers half Chinglish in terms of having the special lead in angle that like you suggested will take care of the irregularities.

I don't expect miracles from a single ball nut. To many questions and not enough answers to not justify buying a pair of ball nuts per axis, then smack thy forehead and buy more later.

[veteq]

Hello guys,

been reading and searching about ballscrews for a long time, the Chinese screws on ebay are c7 types, this means that the lead deveation is max. 0,052 mm per 300 mm.
If you ask Hiwin or Thk if they cann preload a c7 screw they say it is not a good thing to do by using oversized balls or a offset in the nut, if its done this way the balls will suffice from wear very soon.

This is logic, because if the nut is at one point at the screw the preload cann be 0 and at another point it can be as much as 0,052 mm. The deveation is to big. So the Chinese dont claim to be 0 backlasch or even preloaded, on ebay it says 0,015mm i think.

Would we really need 0 backlash or a higher preload, think if you really need a good preloaded screw you would pay at least 700 dollar a set.

Give Thk or Hiwin a call and ask, they cann explain it faster and better then me, since my English isn`t that great.

Regards,

Roy B

[veteq]

Bye the way,
In my opinion, preloading by oversized balls is a very good method, the balls are always under pressure and are in the territory off elastic deformation.
This last thing isn`t bad, since they dont damage from it, if there is a extra force on it by accelerate or milling and it stays in the zone off elastic deformation (Law off Hooke) the ballscrew stays at same quality and dont damage. For pitch offset this is the same.

The material off the balls are a lot stiffer then the spring used in a double nut, so the stifness is at least in 1 direction greater by the first 2 configuration than by double nut.

Dont forget that a preloaded c3 screw needs a very precise base and linear motion system to operate at high quality for a longer time. If the linear motion system isn`t that precise (a dovetail Chinese mill) the expensive ultra precise screw will wear out very soon and becomes a c7 or worse.

For the Dovetail mills the Chinese ballscrews are a good match in my opinion, for the money it doesn`t get better!!!!!!!!!!!

[veteq]

The double nut configuration is possible, al great manufacturers sell them,
dont forget that there are a lot off calculation done to get it right. you cann simply put some spring washers between them and preload them that way, but it isn`t by the book. When i asked Hiwin about a double nut they would need to know a lot data off forces and cycles the ballscrew would be going throw, then they make the calculations and preload the double nut using high precision measuring equipment to get the right preload.
Don`t get me wrong, but doing it the way by just hoping it will be good is the same as trying to build yo`re own space shuttle.

[SCzEngrgGroup]

Bye the way,
In my opinion, preloading by oversized balls is a very good method, the balls are always under pressure and are in the territory off elastic deformation.
This last thing isn`t bad, since they dont damage from it, if there is a extra force on it by accelerate or milling and it stays in the zone off elastic deformation (Law off Hooke) the ballscrew stays at same quality and dont damage. For pitch offset this is the same.

The material off the balls are a lot stiffer then the spring used in a double nut, so the stifness is at least in 1 direction greater by the first 2 configuration than by double nut.

Dont forget that a preloaded c3 screw needs a very precise base and linear motion system to operate at high quality for a longer time. If the linear motion system isn`t that precise (a dovetail Chinese mill) the expensive ultra precise screw will wear out very soon and becomes a c7 or worse.

For the Dovetail mills the Chinese ballscrews are a good match in my opinion, for the money it doesn`t get better!!!!!!!!!!!

"the balls are always under pressure and are in the territory off elastic deformation" - That is not necessarily true - It depends on the accuracy of the screw. With a typical rolled screw, the lead error can FAR exceed the elastic limits of the balls. Were this not the case, there would be no point in making spring-loaded double nuts.

For a hobby mill, a carefully optimized double-nut assembly is not required, as the service life (total lifetime number of cycles) is a tiny fraction of what it is for a production machine. The goal of carefully optimizing pre-load, support, etc. is to get maximum service life. Achieving near-zero backlash with more than acceptable life for a hobby machine is quite easy, and does not require a PhD - simply set the preload to 10-20% of the static load rating of the screw, which is precisely what off-the-shelf double-nuts do.

Regards,
Ray L.

[veteq]

Hey Ray,
think i mixed up a few things, the preload with oversized balls on a c3 screw are always by the way off elastic deformation.
Got some c3 preloaded screws here and the pdf says:
preload torque= 0,3 to 2,5 N/cm.
Are the balls here on elastic deformation or is the minimum of 0,3 N/cm th result off grease etc.(friction)?

So if we use a c7 screw with double nut, in 1 direction you only got zero backlash till 10&#37; off the static loading is reached?

Regards,

Roy

[wildwestpat]

Hi 15mgtar

Yes these are the bits I ordered from linearmotionbearings. See reply 48 for details of what I got in the way of backlash. I have also checked the bearings I got and again I cannot measure any movement with just finger pressure used on the supplied square locking nut. The free end bearing is a good slide fit in its housing. It appears you should check the fixed bearings for orientation as it looks as if there are some BF12's in circulation intended to mount a screw in tension with the bearings in tandem.

Hope you are as satisfied with your purchase as I am with mine. Incidentally the BF12 would be dificult to dismantle to check the bearings and I suggest you measure the end float with a M12 bolt before mouning the screw. I asked for the bearings to be sent separated from the screw to avoid transit damage and was glad I did as it is then easy to check the bearings before fitting the shaft. Please note the ballscrew shaft is a tight fit in the bearings and will prove difficult to remove without risk of bearing damage. Thumps with a mallet and drift are a NOT advised!

Regards

Pat

[15mgtar]

Hi guys,

Thanks for all inputs , since that's the cheapest price I can find for ballscrews and my budget only goes that far for ballscrews I'm going to give it a try and see.

[VooDooMan]

15mg,

You decide on a pair of ballnuts per axis?

[15mgtar]

I'm going with 1 ballscrew + 1 ballnut per axis. It got 3 axis so
RM2505-1600mm-C7
RM2505-1200mm-C7
RM2505-600mm-C7

[VooDooMan]

15mgtar
What sort of machine are you converting, I was looking at your recent threads.
You have some monster steppers for the machine

RF45 or similar?
Some big screws you got ( 3/4 + in size )
I calculate 2600 lbs of force ( dynamic ) on that screw. Seems large!

Let us know!

[15mgtar]

Hi VooDooMan,

I'm converting my old benchtop mill I think is the same like RF, btw I'm totally new and blind to CNC, about that motor I didn't know how big I have to get to drive my mill tables (long time ago I bought an expensive used 120oz motor and thought I could make use of them but when a friend of mine look at them he just laughing looking at the size of the motor that's why I bought a much bigger ones) and now at the same time I'm gathering parts to build my own CNC woodworking router which uses a stupidly too big 40mm shaft slide bearing open type, and for the screws I also not sure how big of ballscrew I supposed to get because my table would be 1.5 meter x 1 meter wide so I thought 25mm would be the smallest one for that size that I can afford. btw I purchased ballscrews and 2.2kw spindle n inverter from linearmotionbearings2008. Haven't arrived yet very anxious to wait. As for the design of CNC woodworking router I haven't got a solid ones yet so if you please suggest me on the design I would be greatfull

[wildwestpat]

Hi 15mgtr

Good to see you have reached a decision. (I just wish that I could accomodate screws as big in diameter as 25mm but they are way too big for the Sieg X2 mill and a 7Xxx lathe look alikes.) You could always add another ballnut per axis BUT the calculations on the spring loading are complex and the further apart the nuts are the more the imperfections in the C7 pitch have to be accounted for. In my opinion if you are after very fine machining limits then there are other factors that have an equal impact on the tolerances achieved in addition to the play (backlash) in the ballnut. Think about the rigidity of the machine as well as backlash as each and every joint is a possible contributor as well as general flexing of machines frame. Sorry if this sounds pedantic and grim - press on and get building and see if the end results are acceptable. You can always spend more money no matter what you purchase. In my experience the law of diminishing returns sets in at the point you can just achieve the required accuracy for the work in hand. There are those of us who want to squeeze as much accuracy as we can out of the hardware at our disposal and this can becomme and obsession - I know the bug is getting a grip on me and I only want to make clock gears!!!! Horologists please don't tell us that this is not so as I need to get the lathe and mill finished.

Kind regards

Pat

[diycnc]

here is how i set mine up on my Bridgeport conversion