Linear Rail design

[mrpi]

I'm mocking up a linear rail that is part of a "lift" system.

I would like to use the type of linear rail Pacific Bearing sells that is self lubricating and suitable for "harsh environments" as this system will be used outdoors.

The linear rail slider itself is "pulled and pushed" by a ball-screw actuator. The slider on the rail is attached to a linkage system that does the lifting.

The question I have is that the linear rail itself does not support any moment about is sliding axis, and I would not like to introduce any lateral forces to the linkage system. Hence, two parallel rails would solve this problem.

Are there accepted layouts for the rails and the number of sliders to minimize the chances of bending/binding of the sliders? I.E. two rails, three sliders? Two rails, four sliders? Two rails, two sliders?

How about mounting of the slider itself? Do I need to design in some degrees of freedom? Or will this depend on the tolerance of the rail/linear bearing?

I'm primarily concerned with robustness and ease of manufacture/setup and maintenance. I.E. the ability to field repair the unit without needing $10k worth of metrology tools.

Axial force seen by the ball screw will potentially be upwards of 10,000 lbf.

Beat to fit, paint to match.

 

[BrianPetersen]

Back when I designed stuff that used bearing rails and ball-screws, both bearing rails would be mounted to a machined-flat surface with one side of one rail designed to be butted up against a datum edge and the other side bolted in using clearance holes. For the bearing blocks, same thing, and wherever possible, the datum edge of the set of bearing blocks on the same side as the rail having the datum edge, would be similarly bolted up against a datum edge, and the other one bolted on using clearance holes. It all has to be bolted to machined surfaces, otherwise there will be binding. The ball nut was bolted to a machined surface perpendicular to the machined surface for the bearing blocks using clearance holes to allow it to find its reference position. The ball screw needs one end fixed, one end free arrangement also, to avoid binding.

But ... this was for special-purpose pick-and-place machinery and occasional special-purpose machining operations, in an indoor industrial environment. I don't know if those components would last outdoors, and they certainly aren't meant for sloppy loose-tolerance operations.

 

[whammett]

We use roughly th same system in my plant, and it works very well. As far as specific layout, we use two rails and two sliders. I was looking at how much force can be handled by the screw actuators and it doesnt say. At our plant we move roughy 2 tons of material at a time, though we use the belt driver actuators.
To answer your question, under your axial force requirements, I think you should contact a Pacific Bearing rep because it is not directly stated on their website or the phamplet that I have.

 

[tygerdawg]

Get yourself a variety of linear bearing equipment catalogs from other manufacturers and educate yourself, most can be downloaded for free. Usually they have very good Engineering sections that describe how their products should be mounted.

TygerDawg
Blue Technik LLC
Virtuoso Robotics Engineering

http://www.bluetechnik.com

 

[mrpi]

Brian, I'll be using a self-contained ball-screw actuator to pull on the carriage so I dont have to worry about supporting the screw or aligning it to the linear rails. The parallel misalignment seems to be the hard part here.

Whammett, does your system use any kinds of seals or wipers on the sliders? The catalog for the linear rails claims that operation in "harsh environments" is acceptable, but I was wondering just how harsh they had in mind.

Tygerdawg, I found the Pacific Bearing application section that has basically everything I was looking for! I'll take a look at some other mfg's to expand my horizons.

Thanks for the replies.

Beat to fit, paint to match.

 

[whammett]

Our system is not only in a "harsh enviroment", it is also exposed to brackish water, so we use seals, but replace them every 3-4 months because of wear. We have had this system for 9(ish) months and it has worked exceptionally well. We also use a silicon based treatment on the sliders ever week or so, especially after heavy rains.
The only thing we were worried about was withstanding hurricanes, but we seemed to get lucky this year with no large hurricanes making landfall in South Louisiana.

"I came, I saw, I made it better."
-Ode to Industrial Engineers
Will ChevronTexaco Corp.

 

[unclesyd]

As an alternate you might want to take a look the Serapid Rigid Chain for you application.

Look at the vertical lift applications.

http://www.serapid.com/gb/documentations/documentations_industry_gb.html

 

[kingnero]

This makes me think of CNC machinery. accuracy into thousands of a mm (0.001 mm)

one rail mounted against a datum adge, as mentioned before, the other one adjustable.
Bearings are positioned using a similar system.
The holes in the rails for the screws are plugged shut.

Wipers (scrapers) are mounted in front of, and after every sliding bearing. They withstand very harsh environment: cooling oil, which is aggressive, and metal chips and particles are all wiped away. Wipers/scrapers are to be replaced regularly.

Renewing the rails/bearings is relatively easy done, but needs a millimess for alignment. And quite some time, and an operator who knows what he's doing.

Ball-screw actuators usually aren't designed for sloppy fits, so maybe you should consider another way of driving the thing?
Bronze threaded housings for the lead screw? Like old-skool mills?

 

[williedawg]

THK is a very good brand and they have catalogs (hardcopy) with tons of pertinent information; and explain the installation procedure very well.

Would shields/covers help in this situation, or is it pointless to use them?
If so, you could have some made of SS....perhaps make a "labyrinth" arrangement to minimize the effects of the environment, and make the seal replacement less frequent.

 

[mrpi]

Unclesyd: wow, that Serapid rigid chain is interesting! That's new to me!

I would really like to stay away from linear ball bearings in this application since it will be outdoors in sand, ice, etc. I'm worried that telescopic shields will have the same problems as the slider in regards to needing sliding bearing surfaces and their own set of seals.

I wont be designing the pivoting part of the mechanism, so the whole thing needs to fit into a specific "envelope", which is making this challenging! I'll see if I can post some sketches to give a better idea of what I'm trying to do.



Beat to fit, paint to match.

 

[EspElement]

Im not sure if this will work for your application but if you check out lee bearings they have cams on the linear shaft.. they can handle some very large loads

http://www.leelinear.com/components/roller_bearing_pillow_blocks.aspx

i use these in a very abbrasive enviorments.. those frelon bearings require a pretty tight tolerance and i know junk can get cought in there and tear them up

 

[mrpi]

Esp, thanks for the link, I was looking into the trade offs of going with rolling element bearings. The static loads are quite high, and I think that's the advantage of the plane bearings.

It looks like we didnt get the contract anyway, so I've stopped all work on the mock-up.

Thanks for the all the replies.

Beat to fit, paint to match.

 

[EspElement]

I use those lee bearings alot.. definitly keep them in mind for rugged applications like that