Linear guide ball counts?

ManualEd · Sep 2, 2015

I F*cked up.

I pulled the X-axis covers on my SQT15MS to clean behind it.
Then I pulled the steel scraper from the rail, cause that was filthy too.
Then I discovered the screws holding the scraper in place also hold the end cap that keeps the balls in it...

Thankfully its only one corner. Whats my next step?

Pop the entire truck off from underneath, and see what it looks like when I load up the balls?

Put a dab of light grease on the balls and start pushing them back in?(Or drop them in from the top)

If you're missing 5/100 balls is it going to jam anything up?

Anyways, some advice would be greatly appreciated. I'm not too worried about wearing out the rails more quickly with a few less balls as they're on their way out already.

Thanks,

Ed

DavidScott · Sep 2, 2015

I would call an engineer at the rail manufacturer to ask these questions.

There are plugs you insert into the block (truck) as you pull them off of the rail to keep the balls in place. Like ball screws there may be spacer balls so confirm this before trying to put things back together.

If you are a few balls short I doubt it would cause any harm, but to be honest I don't know.

I like the idea of trying to put some grease on the balls and poke them back in. I doubt you could cause much more trouble by removing the return cap, just don't move the block on the rail until the return is back in place.

Oh, and good luck!

ManualEd · Sep 2, 2015

DavidScott said:
I would call an engineer at the rail manufacturer to ask these questions.

There are plugs you insert into the block (truck) as you pull them off of the rail to keep the balls in place. Like ball screws there may be spacer balls so confirm this before trying to put things back together.

If you are a few balls short I doubt it would cause any harm, but to be honest I don't know.

I like the idea of trying to put some grease on the balls and poke them back in. I doubt you could cause much more trouble by removing the return cap, just don't move the block on the rail until the return is back in place.

Oh, and good luck!

Thanks,I'll need it!
Just got in contact with NSK. The engineers are all in Michigan and are gone, so I'll give them a shot tomorrow.

I measure .0002 or so variation between some of the balls, so I would alternate them as people in other threads have pointed out.

atomarc · Sep 2, 2015

Ed,

I just replaced all the balls in four new Hiwin blocks that needed a very low effort to move, the factory preload was too much for this manual application. The retrofit involved 500 new balls from 'Bal-Tek' and the swap worked perfectly. The engineer/owner of Bal-Tek indicated that in his world, .0001 was a huge change and recommended I start with balls no more than .0003 smaller than were delivered from the factory. He also indicated that a block will run fine with less than a full factory complement of balls.

I'm not a engineer but I think a .0002 variation in your balls is way out of wack and that from a mechanical standpoint, the larger diameter balls would carry the load and the smaller balls would do absolutely nothing but take up space.

The Hiwin bearing blocks came with .125000 balls and I replaced them, per Bal-Tec recommendations with .124697 balls and all was good!

Stuart

ManualEd · Sep 2, 2015

atomarc said:
Ed,

I just replaced all the balls in four new Hiwin blocks that needed a very low effort to move, the factory preload was too much for this manual application. The retrofit involved 500 new balls from 'Bal-Tek' and the swap worked perfectly. The engineer/owner of Bal-Tek indicated that in his world, .0001 was a huge change and recommended I start with balls no more than .0003 smaller than were delivered from the factory. He also indicated that a block will run fine with less than a full factory complement of balls.

I'm not a engineer but I think a .0002 variation in your balls is way out of wack and that from a mechanical standpoint, the larger diameter balls would carry the load and the smaller balls would do absolutely nothing but take up space.

The Hiwin bearing blocks came with .125000 balls and I replaced them, per Bal-Tec recommendations with .124697 balls and all was good!

Stuart

Thats exactly what I wanted to hear.

With regards to the .0002 variation, I read in a few other threads that they often alternate larger-smaller-larger balls so that, like you said, the larger balls will take the load, while the small ball will act as a bearing between them.

Garwood · Sep 2, 2015

I re-balled all the trucks on one of my VMC's and they used spacer balls. Everyone I spoke with about it, including bal-tek said the same thing, that linear rail trucks use spacer balls. I used spacer balls .0005" smaller than the load balls. About 150 total balls per truck.

The real Leigh · Sep 3, 2015

atomarc said:
...the smaller balls would do absolutely nothing but take up space.

Stuart,

The smaller balls are quite important for proper operation of the assembly.

Draw a pair of balls pushed tightly together and rotating in the same direction.

The two sides touching each other are rotating in opposite directions.
This produces considerable wear on their surface due to friction.

If you put a slightly smaller ball between them, it rotates in the opposite direction.
This eliminates the frictional wear completely without affecting operation at all.

- Leigh

machtool · Sep 3, 2015

The real Leigh said:
The smaller balls are quite important for proper operation of the assembly.

Draw a pair of balls pushed tightly together and rotating in the same direction.

Its always the electrical guys that push this myth. They think there like electrons pushing each other.

First and foremost there is no tight in the ball train. There's always a considerable gap in the circuit, normally one or two balls are left out of the circuit. You wouldn't be able to get the end caps or return tubes back on otherwise.

Secondly, it doesn't apply in pre-loaded assemblies, with in Machine Tools as is always the case. No one likes a spindle or slide system to have clearance in it.

While we are drawing pictures, draw one ball, it is under half the preload pressure where it touches the race in the rail block, and the other half of the preload where it touches the race in the rail. Move either relatively to the other and the ball is positively driven to rotate. It has to, it has positive preload at each contact point.

Add the other balls in, each and every one is driven by its contact pressure via the races.

This nose to tail crash of the balls is ball shit.

Has any one ever found the master ball? The Hadron Collider ball. He's the one with the whip at the end of the ball train, going mush, mush. giddy up little ballys, and makes them crash into each other. Under pressure if you buy into this counter rotation wear myth.

That would be to believe that some balls travel faster down the same length of rail than any other does. Or "push" the other balls. It's frog shit, just being parroted on what they have heard here.

Phil.

DavidScott · Sep 3, 2015

I talked with quite a few people when reballing some screws in my VMC, including a very informed man at Bal-Tek. By far the most accurate advice I got was from the engineer at NSK, who went well out of his way to help me. Kuddos to machtool for all of his help when I needed it on that project!

The real Leigh · Sep 3, 2015

machtool said:
First and foremost there is no tight in the ball train.
There's always a considerable gap in the circuit, normally one or two balls are left out of the circuit.

Then how do you define preload?

And kindly explain this statement from post #4 above.

atomarc said:
I just replaced all the balls in four new Hiwin blocks that needed a very low effort to move, the factory preload was too much for this manual application.
...
The Hiwin bearing blocks came with .125000 balls and I replaced them, per Bal-Tec recommendations with .124697 balls and all was good!

- Leigh

CarbideBob · Sep 3, 2015

machtool said:
I......
While we are drawing pictures, draw one ball, it is under half the preload pressure where it touches the race in the rail block, and the other half of the preload where it touches the race in the rail. Move either relatively to the other and the ball is positively driven to rotate. It has to, it has positive preload at each contact point.
.......

It would be nice if bearings behaved this way.
If you design machines with these type bearings you know better. Precise needle roller guide rail and block assembly, custom ball leading screws supplier from China Cle auto Source Industry (just a nice picture I found, my source has pages to go through) ....
Even under many thousand pounds of preload you will find the cage will walk out the ends without provisions for bearing creep.
No matter the preload, lube or cleanness the bearings don't always follow the motion and this style is way stiffer and higher in accuracy than any rail/truck system
It's the whole stopping, reversing and starting thing. Not like a conventional bearing that just keeps moving.
This problem is why you see "caged ball" rail/truck systems and pay more for them.
Bob

machtool · Sep 5, 2015

CarbideBob said:
. Precise needle roller guide rail and block assembly, custom ball leading screws supplier from China Cle auto Source Industry (just a nice picture I found, my source has pages to go through) ...

I'm at at a total loss. Bob wants to link in needle rollers. Leigh hasn't heard of preloding

Ox · Sep 5, 2015

ManualEd said:
I F*cked up.

I pulled the X-axis covers on my SQT15MS to clean behind it.
Then I pulled the steel scraper from the rail, cause that was filthy too.
Then I discovered the screws holding the scraper in place also hold the end cap that keeps the balls in it...

Thankfully its only one corner. Whats my next step?

Pop the entire truck off from underneath, and see what it looks like when I load up the balls?

Put a dab of light grease on the balls and start pushing them back in?(Or drop them in from the top)

If you're missing 5/100 balls is it going to jam anything up?

Anyways, some advice would be greatly appreciated. I'm not too worried about wearing out the rails more quickly with a few less balls as they're on their way out already.

Thanks,

Ed

Running it a few balls short of a full load will be fine.
I have one machine that had spit balls out of the trucks for several yrs before I had to replace - once too many balls were gone.

I blame the ball loss on too many pumps from the grease gun early on, as well as cracked or softened plastic end caps later on.

... but a cpl balls missing in one of these should not be a problem.

Now - if you lost a needle down in the crank case from your un-caged wrist pin, and then start it up, the lost needle will cause may more damage being where it shouldn't - than it will not being where it should. (Not that we were originally talking about needles/cylinders)....

------------------------

Think Snow Eh!
Ox

CarbideBob · Sep 5, 2015

machtool said:
I'm at at a total loss. Bob wants to link in needle rollers. Leigh hasn't heard of preloding

They make the same way systems with balls in place of the needles with the same creep problem.
They are only used on light stuff as they won't carry much weight.
Same thing happens in a preloaded bearing like on the ends of your ballscrew.
If you bring the screw to the same home position after many cycles the retainer will not be where it was oriented when you assembled it.
Even when heavily loaded the balls "skid" during reversal and don't perfectly follow their tracks.
No matter how much you load it is not a fixed drive system like a gear is.
Bob

The real Leigh · Sep 5, 2015

ManualEd said:
the larger balls will take the load, while the small ball will act as a bearing between them.

Correct.

It does not matter whether there is space between some balls or they're all jammed together.

The spacer balls provide a direction of rotation that matches the load balls at each interface.

- Leigh

atomarc · Sep 5, 2015

My comments were in regards to Hiwin linear guides. These are pretty nice units made in Taiwan. I went to the shop and measured the balls I swapped from the new blocks...there were about 500 balls and when each factory block was opened for the ball change out, there was room between the runs of balls, they were not packed tight. Using a Starrett electronic micrometer and a Mitutoyo standard, all the balls I checked measured .125000, there were no small balls that I could find.

I found it interesting that a metric bearing assembly would use inch balls but the Bal-Tec gentleman said this isn't out of the norm as the US makes the most accurate bearing balls.

The preload I spoke of was not in the number of balls crammed into a run but rather the fit of the balls into the block and guide rail interface.

While all the Hiwin balls appeared to measure the same, I can see the benefit of alternating small balls to achieve the dynamic Leigh points out. I didn't measure all 500 balls.

Stuart

garyhlucas · Sep 5, 2015

I built a window system with Harken ball bearing sailboat blocks running on vertical rails. These blocks ran smoothly under heavy loads when horizontal. However in the vertical orientation the balls all fall to the bottom and then motion is very much rougher. I could not source the Torlon balls in a smaller size to alternate with the larger balls so we had to live with it.

Heavey Metal · Sep 5, 2015

garyhlucas said:
I built a window system with Harken ball bearing sailboat blocks running on vertical rails. These blocks ran smoothly under heavy loads when horizontal. However in the vertical orientation the balls all fall to the bottom and then motion is very much rougher. I could not source the Torlon balls in a smaller size to alternate with the larger balls so we had to live with it.

Bullshit

Proly mounted the rail on an uneaven surface and now want to blame it on some dum shit like this

The real Leigh · Sep 6, 2015

machtool said:
Leigh hasn't heard of preloding

You're absolutely correct... I have never heard of "preloding".

I have, however, heard of "preloading", and am quite familiar with the concept.

- Leigh

garyhlucas · Sep 6, 2015

Heavey Metal said:
Bullshit

Proly mounted the rail on an uneaven surface and now want to blame it on some dum shit like this

If you look up Harken travelers for sailboats you'll see that they have large clearances so they can be BENT on a vertical or horizontal arc and still function properly! Also you can see the balls in operation so I could see the balls bunch up against one another and jam when mounted vertical. The tracks in question were specially made for me in 18 ft lengths, and I bought 100 lengths along with over 200 cars. These rails and cars have been in operation for about 15 years now. So no BS to see here, just move along.

Linear guide ball counts?

Stainless

Diamond

Stainless

Diamond

Stainless

Diamond

Diamond

Diamond

Diamond

Diamond

Diamond

Diamond

Diamond

Diamond

Diamond

Diamond

Stainless

Banned

Diamond

Stainless

Similar threads