Still no attatchment here, do you have a web link?
Revolutionary Linear Drive System
Nexen has a new revolutionary linear drive system that is similar to rack and pinion but eliminates it's short comings. This makes it a better choice for many applications that typically use traditional rack and pinion, ball screws, and linear motors. See attachment for details.
Similar Threads:
Last edited by allanconway2; 11-16-2004 at 05:02 PM. Reason: Add Attachment
Still no attatchment here, do you have a web link?
(Note: The opinions expressed in this post are my own and are not necessarily those of CNCzone and its management)
There is also the 'Rolling-ring' drive that is interesting and has been around for a while, I think it is used mainly in lower load applications, I have come across them in CNC back gauges for shears & brakes etc , Amacoil is one manuf. http://www.amacoil.com/ due to the possibillity of slip under stall or high load, a linear scale is usually used to detect actual position.
Al
CNC, Mechatronics Integration and Custom Machine Design
“Logic will get you from A to B. Imagination will take you everywhere.”
Albert E.
I picked "other" in the poll. I would have thought timing belts should have got a mention.
Thanks for posting the PDF file of the nexen drive system.
How comparable in price is it to other methods?
It look like It is made out of plastic. Is this correct?
Being outside the square !!!
Belt drives are fast and inexpensive, but not generally considered precise when it comes to linear positioning.
The roller pinion body and rack is made from high quality steel with excellent heat treat qualities. The pins are bearing grade steel. The black coating is Raydent. It is a high performance blackening process with excellent wear and corrosion properties.
Plastic racks may be considered in the future for low cost and force applications that need nonmagnetic, or high corrosive resistance.
Cost of the roller pinion system is comparable or less than ball screws and more expensive that traditional precision rack and pinion except for split and dual pinion systems that can be very expensive.
Last edited by allanconway2; 11-17-2004 at 12:06 PM.
Excuse my basic observation, but the literature claims the accurate positioning eliminates the need for "costly and complex split and dual pinion systems". I still imagine you would opt for rack and pinion on both sides of your gantry? I don't know if this is called dual pinion.
Dual pinion does not mean separate rack and pinion systems driving opposite sides of an application. Dual pinion systems have two pinions operating on a single rack and are preloaded against each other to eliminate backlash. This is done with either a single motor with two coupled gearboxes or two separate servo drives where the controller maintains the preload.
The attachment says it "eliminates the cumulative error problems experienced with ball screw systems". Could you please explain this further? It seems to me the cumulative error will consist of the accuracy with which the screw or rack is ground plus the thermal expansion.Originally Posted by allanconway2
In a ground and preloaded ballscrew with preloaded angular bearing block the precision is pretty much built into the unit at delivery. It seems that with your system some of the accuracy must be provided by the machine builder in that any guide system error perpendicular to the rack will affect the positioning error in the direction controlled by the rack/pinion. The effect of such errors are not described in the data sheet?
What is "Backlash Compliance". Ground and preloaded ballscrews have no backlash, but does have some compliance, usually expressed as a distance/force figure. What are the 3.2um? Are they backlash? Or are they compliance, if so at what force?
What about protection from the environment in which it is working. We're mostly concerned with processes generating swarf, dust, etc. here. What are your methods of keeping this debris from affecting the drive? When using ball screws there are usually 2 levels: An outer shield, often provided by the machine construction itself and the placement of the screw, a coiled spring or other shields. Then a secondary in the ends of the nut where many systems are used, some very effective, others not so.
Ball screw cumulative error is a widely recognized issue. There are numerous sources on the web that discuss this issue in depth. If you look at a ball screw accuracy graph you will see that the lines are not horizontal and angle away from the zero axis indicating a certain amount of lead error per unit of travel. Over short distances it is typically inconsequential, but over longer distances becomes a problem forcing the use of linear encoders to get accurate positional feedback. At lengths less than 9 inches ball screws are generally as accurate as the RPS system, but beyond this the RPS is much more accurate than a ball screw. As you have seen, the cumulative error graph for the RPS is horizontal meaning there is none regardless of the distance.
As for accuracy being built in to ball screws and guiding system design effecting the RPS systems accuracy. Precision and accuracy are not the same thing and you can have one without the other. Ball screws can be precisely manufactured but are not necessarily accurate. More on this can also be found by a web search. The guiding system requirements for the RPS are not significantly different than ball screws. Typically profile guide rails are used to guide the system and dictates the precision of component installation. Ball screws cannot tolerate side loading so the parallelism requirements of the system are high in either case. If the guiding system allows the RPS to loose preload due to poor machine design or installation then accuracy would be compromised.
On the issue of backlash you say that ball screws have none. This is not entirely true of ball screws or any mechanical system. Some clearance is always needed for parts to slide or rotate, although it can be very small. Even with bearings clearance is required for ball or needle lubrication or it would be metal on metal. Yes, mechanical nut preloading takes most of the play out of a ball screw but there is still clearance for lubrication that can be squeezed out when sufficiently loaded. If you look at so called zero backlash ball screw specs in a catalog you will typically still see a backlash value. The RPS uses a preload to achieve a similar effect. Equally important is system compliance or elasticity, ball screws are much more elastic due to deformation of the ball nut and the unsupported length of the screw. The 3.2um linear backlash stated for the RPS takes into account backlash and compliance. It was measured with high precision rotary and linear encoders while applying the maximum linear force while the pinion shaft was locked. In addition to this ball screws are much more susceptible to thermal expansion due to environmental variation and heat build up in the nut due to high speeds. The RPS rack is rigidly mounted and with it's 99% efficiency very little thermal expansion is generated regardless of how hard you run it
On the issue of environmental contaminates. The RPS system is susceptible to environmental contaminates as are ball screws and traditional rack and pinion. If the rack is not oriented with teeth up and ideally downward contamination is minimized. The lubrication requirements for the RPS are much lower than ball screws. The rack gets a light greasing at installation and then typically twice a year thereafter. Harsh environments may require more frequent servicing. There are many lubrication options from food grade to dry lube that can be chosen based on the application. Lubed for life ball screws aren't and are based on clean room conditions and light loads. The thing to keep in mind is that the needle bearings that support the pinion roller pins are permanently lubricated and sealed. As the rollers engage the rack it is mostly rolling friction. Each time a pinion roller engages and disengages a rack tooth it only turns a small fraction of a revolution. This gives the RPS its 99% efficiency rating compared to a ball screws 80%. In other words very little friction or wear takes place giving the RPS much longer life in most applications. We have had customers use the RPS for life issues alone and they are very satisfied with it. If contaminates get into a ball nut they quickly eat it up due to the high amount of work the balls do in the nut and opportunities for friction. As with any drive system if shielding can be installed to protect it from contaminates it is a wise thing to do and should extend the life of the system.
@Allanconway2: I appreciate your reply. Although I do not agree fully, your comments are good additional info to the data sheet.
Do you have any mounting suggestions with sketches/drawings showing preferred mounting and contaminate shielding?
Also it would be nice with some prices. I assume most in these forums are interested in one off prices for the hobby and small volume custom machine builder.
I would be interested in knowing what you disagree with. If you can show me that any of what I have said is in error I would appreciate it. I am not big on hyping things and only deal in facts as I know them.
For more information on installation I sugest you download the user/installation manual for the RPS from Nexen's website. Go to www.nexengroup.com. The RPS is presently featured on our home page and links are provided to various RPS information. I am not giving out price lists at this time but can give quotes on specific applications. Let me know wheather you prefer direct or through distribution and the quantities involved. Keep in mind that when comparing prices of the RPS to ballscrews that many aspects of the machine design may change and it is the final package cost that matters. From our experience the RPS is price competitive with precision ground ball screws and offers more value. If you are interested in seeing the RPS Nexen has a representative Belgium who may be able to meet with you.
The big linear guide manufacturers (INA etc) have guides with built in decoders, wich gives an absolutely no-backlash positioning. The only minor thing is - the price...
But at the last tech fair I visited there was a Chinese guide manufacturer that had the same solution but to a price that was reasonable. Still not cheap, but very interesting if precision is an issue.
Regards,
Sven
svenakela,
What is the web site for the INA manufacturer? Also, what is the name and web site for the chineese manufacturer the makes cheaper but similar devices that you mentioned?
Roller chain/sprocket inside out
It's said there's nothing new under the sun...
I Look at this and I see a roller chain drive. The only difference is that the "sprocket" has been turned in to a "rack"; and the roller chain itself becomes the pinion.
Since I've had good luck in the past with roller chain drive; I'll probably just keep using that inexpensive system; knowing that this "update" confirms what I've always felt about its suitability for axes driving...
Ballendo
P.S. Allan, Are you an employeee of Nexen?
The comments farther down suggest that the pitch error of ballscrews and nuts needs to be corrected by encoders. I disagree; and feel that there is some advertorial going on here... It--lead error-- is pretty simply accounted for by adjusting the steps per unit to accomodate this increasing or decreasing error. The rolled thread techniques used for precision rolled ballscrews will favor a lead error over a cyclical error, since lead error is EASILY reduced/removed; whilst cyclical error is the bane of every linear motion drive system.
From what I can see, the nexen drive will have the same issue od "lead" error. It will be due to the pinion PD not being perfect. And can be accomodated the same way.
The problem with chain and this nexen drive is that it has an uneven speed of motion. The pinion is essentially a polygon instead of a circle; and since the effective radius changes for any polygon inversely to the number of sides, there WILL be uneven motion speed. The "Rack" tooth profile will attempt to minimise this, but it will still be there.
The epicycloidal flank shape of gears is designed to eliminate this by providing a constant diameter circle at the PD where two gears intermesh. This flank shape is also designed to minimise relative sliding motion between the gear flanks to minimise wear. The nexen drive will suffer from the same wear effects as roller chain, albeit perhaps to a lesser degree since the rollers of the pinion ostensibly have better bearings than the ones used in small roller chain. But over time the rollers will wear and their diameter will no longer fit the rack tooth profile exactly. And the rack profile will similarly wear and no longer match the roller diameter. Without a means of compensation backlash will result. Since there is not a "wedging effect" as found in "normal" rack and pinion--the root of the nexen tooth form appears to match the roller diameter exactly--there appears to be no ability to "deal with" this wear outside of replacement.
I think the single most determining factor in a " home built CNC " or a hobby use machine is availability. We like ballscrews because they are cheap and plentifull. I dont see a very large number of people on this forum buying things new, unless they are surplus items. Im sure if you dumped a ton of "Revolutionary Linear Drive Systems" on the surplus ebay market, they would become very popular overnight.
Halfnutz
Unless, of course, they are cheap as **** new. I know, I know...wishful thinking.
(Note: The opinions expressed in this post are my own and are not necessarily those of CNCzone and its management)
Check Out My Build-Log: http://www.cnczone.com/forums/showthread.php?t=6452
I work for INA USA. I don't have anything to sell, just my 2 cents. Actually I work out on the shop floor. As far as price goes, you will never beat China! Coolman here is a link ----> http://www.ina.de/inaupdate/homepage4_gb.asp?flash=1 Do a search for Linear at the bottom of the page. Im not really sure what svenakela is refering toINA has so many products.The big linear guide manufacturers (INA etc) have guides with built in decoders, wich gives an absolutely no-backlash positioning.
I looked at these Nexen rack/pinion. I opted not to use them simply because they couldn't take the power of the 640 oz steppers I planned to use. If you jump up to the next size pinion, then you loose resolution, its a dog chasing its tail type thing.. to get the resolution you'd need to up the reduction, which up's to torque which means you need a bigger pinion.. etc etc..
I settled on steel rack/pinion... besides.. its cheap and you can purchase it locally..
Jerry
JerryFlyGuy
The more I know... the more I realize I don't
(Note: The opinions expressed in this post are my own and are not necessarily those of CNCzone and its management)
thats a nice system and anyone with a cnc mill can easily build that system ,ive seen it a while back and its been stuck in my head for a while now, the rack can easily be made out of 6061 layed on its side ,contour the profile ,then hard anodized , same goes for the pinion housing with dowel pins pressed in , considering a 6061 bike sprocket can take the abuse , it would take a long enough time to wear the rack out , and its light weight easy to machine
OH CRAP now im rethinking my current router design im working on
I run 0.8" PD X and Y rack gears on a NEMA 34 5' x 10' CNC router. I'm not that enthused about the microstepping needed to smooth out movements. The motors barely turn, about 40 rotations for the 100 inches of travel it has. Yes, it has hit 1200 inches per minute, but most of my work is in the 100 inches per minute range. It would only exaggerate the issue with a larger pitch diameter. Ballscrews are like a free gearbox being added, multiplying torque, increasing resolution (reducing microstepping), and driving an axis, what a deal! The Nexen system looks better and less noisy than a rack system, but with the increased PD, router users will run out of stepper pulses and mill users will run out of torque. I'm a free torque multiplying, increased resolution, drive my axis guy.
Jim
Posting Permissions
