virtual leadscrews

[GEOFF]

I have just purchased an old scrap wessel pcb cnc router.The leadscrews in X and Y are plain 25mm round bar and the nuts seem to be some sort of preloaded spiral bearing.The system does work as I already have such a machine up and running.What I would like to know is does any slip occur? The machine I am running uses the original servos and heidenhain scales. As I want to increase the X axis travel on the "new" machine (it is a twin spindle job and I intend to change to a single spindle with greater travel) the optical scale is too short and I was thinking of changing to a servo with encoder, but do not know if any slip is likely to occur. I do know that conventional ballscrews would resolve the problem, but would be an expensive solution and funds are tight at present.If slip is unlikely to be a problem steppers would also be another solution.If anyone has any experience with this kind of leadscrew I would be very interested in any comments.Thanks,G

[rokag3]

hi,
maybe you have a roller screw
http://www.cnczone.com/forums/showthread.php?t=13593
will be nice to put some picture
just to see what we are talking about

[GEOFF]

Hi rokag3,thanks for your reply,I dont know what a roller screw is, and I dont know how to post pictures!!!the leadscrew is a 25mm dia plain bar and the "nut" is a completely enclosed bearing block.I have looked on the web and the system is probably similar to the rohlix system.I have come to the conclusion that I must stick to using a servo and linear reference scale in case some slip were to occur.With the current system the carriage could wait for a cup of tea and still not lose position as until the carriage arrives no further direction change can occur,wheras using a stepper or encoder the system could poosibly slip, but continue.thanks for your interest.Regards,Geoff.

[unterhaus]

it's not a roller screw like the one Geoff posted. I have one of the ones with the plain shaft, they have been around at least since the '80s. I could never get mine to slip, but there has to be a point at which it will slip. Just don't know what that point is. I'm sure they have ratings if you can find any literature.

[rokag3]

hello,
you have here very good literature for roller screw (very big file)

http://skf.manager.nu/publication_fi...7551246328.pdf
if you have got something like that keep it because you are very lucky
and it last

[NEATman]

Geoff-
Perhaps it is one like this:
http://www.zero-max.com/products/rohlix/rohlixmain.asp

Info on the inventor:
http://www.motionsystemdesign.com/Is..._inventor.aspx

Internals:
http://amacoil.com/html/page3a.htm

Variable pitch models:
http://amacoil.com/html/page3d.htm

I hope this helps.

NEATman

[Mike Nash]

I've seen this type of "screw" on a core cutter that feeds long cardboard cores (for rolls of tape) onto a mandrel, stopping and cutting at the selected roll widths. It was run by a servo also. I don't know whether it ever slipped, but no one ever complained about the core widths being off, just all the other stuff that continued to break.

[rokag3]

hey Mr. Neatman,
I know this system the angle of the bearing with the shaft determine the speed and the thrust i have eared that some big machine use this system i may have some information with a guy who was controlling it with a stepper motor,
but with a good film linear encoder this solution could get revival in hobbyist because ball screw our purse (mean it's expensive what else !!!)

[NEATman]

rokag3-
This type of unit (http://www.zero-max.com/products/rohlix/rohlixmain.asp) shouldn't be too difficult to reproduce - especially if a dedicated fixture is made to hold the parts at the proper angle. Here's an unusual idea - MAKE THESE FUNCTION AS THE LINEAR BEARING BLOCKS ALSO. Picture one of these on each side of a gantry, perhaps 25mm OD shafts that are belt driven together. Then make these blocks perhaps 8" (200mm) long, and they may work as the carriage blocks also. The top and bottom clam shell blocks could be identical, having two bearings on one end and one on the other. You could also have one counterbore and one tapped hole in each.

Just think about this one too - to "zero" or home the machine, just run it into the hard stop - as it won't hurt anything, reset your counter to zero, and away you go. For a gantry with one on each side, you could have a screw adjust hard stop on each side. once you figure out the squareness, you could do the same thing - run it into hardstop and it would resquare without damaging anything.

NEATman

[NEATman]

There are a ton of other inventions such as this out there, dating back to the 60's. I did a patent search, and here is the result for this product (I believe the patent just ran out in August):
http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=1&u=%2Fnetahtml%2FPTO%2Fsearch-bool.html&r=19&f=G&l=50&co1=AND&d=PTXT&s1=%22Rohlix%22&OS="Rohlix"&RS="Rohlix"
You'll need to get a viewer from thier website. Check out the other referenced patents on the above page - they made one that worked with an acme screw, and this one:
http://patimg1.uspto.gov/.piw?Docid=...iew+first+page
absoutely brilliant!

NEATman

[RICHARD ZASTROW]

FYI, NB linear system calls theirs "slide screws".