Linear Bearing Idea

Post your build logs here

Re: Linear Bearing Idea

Postby bdring » Sat Sep 11, 2010 12:15 pm

dirktheeng: Tolerances, Deflection, Straightness.

Those are all really good points and do apply, but more so at the higher end of things. The target for this is a cheap and simple replacement for what is already used and working in the the DIY end of CNC. All of your points apply even more to my existing bearing system. I use a stock extrusion and manually drill and tap the parts and bolt them together, yet it still works fine. An integrated system only needs to be as good or slightly better.

When I set out to build a large laser cutter for under $1000, I was only able to allocate about $70-$80 for the linear guide system. Misumi also sells precision machined extrusions. They take a stock extrusion and machine it to improve the tolerances. I decided that it probably was not going to add that much to the design. My tolerance expectations are basically to exceed my current configuration.

DIY people make CNC machines out of electrical conduit, black plumbing pipe, plated all thread lead screws with plated coupling nuts as the drive. They have gantries made out of particle board, they use drawer slides as bearings. They use unsupported shafts and angle iron. The best part is that they actually get great results. I was able to machine working PCBs with 0.012" traces and spaces on a machine made from wood, all thread rod and belt drives. You built it, you tune for a while, maybe re-tune it in 6 months, but it works.
Bart
"If you didn't build it, you will never own it."
bdring
Site Admin
 
Posts: 2966
Joined: Sun Nov 22, 2009 7:33 pm
Location: Chicago, IL, USA

Re: Linear Bearing Idea

Postby dirktheeng » Sat Sep 11, 2010 1:14 pm

Bart,

I agree... after designing my cnc in solidworks, blacktoe 4.0 was released. In his new design, he addressed a lot of my concerns (not all.... but most of them) and you can get a working 4x8 CNC for about 3500. That is very impressive. I think it mills wood with fairly high precision (not sure about aluminum.... I think the system complience is going to lead to chatter, but I won't know until I try). I couldn't touch that kind of price with an aluminum frame and rails. It's also nice because I don't have to do the development work (though that can be fun).

The only thing about trying to get more accuracy than the current system is that you are going to run into issues because the linearity of the system is based only on the tolerances of the extruded part. With the current system, the linearity is really based on the bolted on rails. The extrusion can be all caty wampus, but at least your xy tolerance is based mostly on that rail as you bolt it on. The z tolerance could still be an issue, but could be overcome with shiming in the right spots... the disadvantage of the extrusion is that there is no adjustment at all... kindof just stuck with whatever you get.

I think for small projects this could work out great and will certianly have it's place. If I can help, let me know. In all honesty, if something were available for this project, I probably would go for it as my travel distances are low and needed tolerances are not real high.
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

Re: Linear Bearing Idea

Postby bdring » Sat Sep 11, 2010 7:09 pm

The V rails by themselves are pretty worthless and floppy. They are not much of a reference for anything. When it comes to using them any burr or misaligned hole will affect the straightness way more than any extrusion tolerance. I recommend over drilling the holes and using extrusion as the guide for straightness. The Misumi extrusion is the only one I have found that does a good job of this due to the really small radii on the corners.

That Blacktoe machine is state of the art DIY. But if you do all the tolerance, stiffness, straightness, dimensional stability analysis your find that the bumblebee can't fly.

Wikipedia: Filght ...
According to 20th century folklore, the laws of aerodynamics prove that the bumblebee should be incapable of flight,
Bart
"If you didn't build it, you will never own it."
bdring
Site Admin
 
Posts: 2966
Joined: Sun Nov 22, 2009 7:33 pm
Location: Chicago, IL, USA

Re: Linear Bearing Idea

Postby dirktheeng » Sat Sep 11, 2010 10:53 pm

Bart,

I suppose your right.... I was just thinking that with no stress applied to the rail, it should be strait... if the holes are oversize and clean, you should be able to get one rail attached to the extrusion, one end first, then the other then workign through by going to the middle. Tighten the screws loosly first and then go through after they are all snug and get them tight.... perhaps a bit of epoxy too so that you don't have to apply a lot of stress to get the rail fixed.

Anyhow, I was parusing Misumi's site and came accross a couple things:

1) Simplified slide rails - don't know what tolerances and slop these have, but could work... really cheap ($25 for 560mm, KSRT45-560) which includes rail and car. I think if they ar used in pairs so that there is tension or compression to take out lateral slop, they could be ok. I think this is a cheaper and easier solution than the v-groove setup now. I couldn't find really cheap 5/8" wide vgroove rails. The prices you list in your BOM are closer to the 7/16 unhardend carbon steel prices at McMaster. I am considering using them, but I don't want them to rust up. Misumi sells nickel coated steel rails that fit the bill for this project for $26, and $37.6 per piece. With the 30% first time buy discount, that is only a few dolars more than the mcmaster price.

2) roller slide rails - RSRR25-560, the rail is about $28, and a care is about $12. This is like the v-groove rail system but in reverse. The bearings push out agianst the inside of the rail as apposed to in against the outside. I would be willing to guess that this would be at least as accurate as a midgrade v-groove system. They make an aluminum and stainless steel version. When you consider the cost of the extrusion and the v-groov bearings, this may be at least cost competetive to what you may be able to get with the extrusion you want to make and for applications with no tool pressure and small spans, this really may work well. Granted, it has to be bolted to something, but the misumi extrusions are cheap. In addition you can get these in lengths up to 1800mm, and get connectors to make it longer. I don't know what the load rating is.

Also, I was wondering what you thought about just epoxying the rail to the extrusion rather than bolting them. The loads are very light I don't think there would be a problem with the epoxy letting go.
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

Re: Linear Bearing Idea

Postby dirktheeng » Sat Sep 11, 2010 10:55 pm

[quote="bdring"]The V rails by themselves are pretty worthless and floppy. They are not much of a reference for anything. When it comes to using them any burr or misaligned hole will affect the straightness way more than any extrusion tolerance. I recommend over drilling the holes and using extrusion as the guide for straightness. The Misumi extrusion is the only one I have found that does a good job of this due to the really small radii on the corners.

That Blacktoe machine is state of the art DIY. But if you do all the tolerance, stiffness, straightness, dimensional stability analysis your find that the bumblebee can't fly.

[url=http://en.wikipedia.org/wiki/Bumblebee]Wikipedia: Filght[/url] ... [quote]According to 20th century folklore, the laws of aerodynamics prove that the bumblebee should be incapable of flight, [/quote][/quote]

yea, I know, but for woodworkign projects 1/32"-1/64" is plenty good enough. I would think that it would be possible (with careful alignment and jurry rigging) to get that kind of tolerance from the backtoe. This wouldn't make a very good metal working machine and I don't know if I would want to try aluminum with it.
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

Re: Linear Bearing Idea

Postby bdring » Sat Sep 11, 2010 11:07 pm

Misumi Rails. Interesting. I never saw that before. I think the $12 is for plastic wheels. The bearings are single row ball bearings which have some slop on twist.

Exopy...sure, I have heard some people use double face tape. They can corrode due to gases, so it might be nice to have them replaceable.
Bart
"If you didn't build it, you will never own it."
bdring
Site Admin
 
Posts: 2966
Joined: Sun Nov 22, 2009 7:33 pm
Location: Chicago, IL, USA

Re: Linear Bearing Idea

Postby dirktheeng » Tue Sep 14, 2010 12:49 am

I think the RSRR25 could be a really great alternative product. They also make a 40mm version which is all steel. However, the 25mm version is extruded aluminum. The cart does have plastic wheels, but for really light duty applications like lasers, I don't think that is going to matter one bit. I probably wouldn't use it for a cnc milling machine with significant tool pressure, but for this kind of stuff, it may be what you need. When I get my system set up, I'm going to mess with these things and see if they are any good. If they are, I will modify my cutter to use them and compare results.
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

Re: Linear Bearing Idea

Postby bdring » Tue Sep 14, 2010 2:56 am

I worry about the single row ball bearings. Take any bearing, even ABEC 7. Hold the inner race and rock the outer race. That is a big amount of slop.
Bart
"If you didn't build it, you will never own it."
bdring
Site Admin
 
Posts: 2966
Joined: Sun Nov 22, 2009 7:33 pm
Location: Chicago, IL, USA

Re: Linear Bearing Idea

Postby dirktheeng » Tue Sep 14, 2010 11:19 pm

I think you are confused about which part I mean. I was going to attach an image of the inventor files I am drawing up, but I can't for some reason... the options are not there.

Anyhow, look up the part RSRR25-480 and is $27. The car that goes with it is RSRS25, and is $12. This is a 480mm track with 18.25" of travel. It has center line holes to bolt to anything you want (I'm going to use t-slot extrusions). The track is aluminum alloy, the car body is 304 stainless, the wheels are a very very hard plastic (polyacetal) (the same stuff they make really hard scate board wheels from). There really isn't much room for any slop. This is not a roller slide like a drawr guide. Given that the loads are so light, a hard plastic does not concern me, and frankly it will wear better against an aluminum surface than hard steel. I can do the calculations for deflection, but from experince I can tell you that the expected loads are so low, that plastic will not deform more than 0.001" I would only expect a few ounses of live load for a laser cutter application.

This length would be used for the Y axis, so I need 2 of them. The total cost for the guide, car, and bearings (in this case included in the car), is $78. That doesn't include the first time buyers 30% discount (which brings it down to $54.60)

Now lets compare to the current design. To get simmilar travel, I have to order the 2ft v-groove track from mcmaster. If I get the cheapest 5/8" rail (which is what is specified in the drawings you posted), I can get them for $12.78 a piece. Mind you, these are unhardened steel without any chrome coatings or pre-drilled holes. I have yet to find a cheaper source for these rails and I don't particularly want to use an uncoated steel piece given that I know it will corrode. VXB sells the v-groove bearings for $9.95 each. I am assuming you got some sort of volume discount on the bearings to be able to offer them at the price you do in your bearing kit. Lets assume you got 30% off and can pass those savings on to us. That means the bearings are $6.97 each. The design calls for 3 each for each car, so the minimum price for the car is now $20.89. That doesn't include any hardware or materials to make the car (which is included in your hardware kit and the xy kit). Lets assume that there is somewhere around $5 of cost for the car plate and hardware for each car. That means that each car is about $26. I need 2 cars and 4 v-groove rails. That totals $103 and I still have to drill the v-groove rails and drill and tap the extrusion, and I still have to worry about protecting the rails from corrosion (which could be a problem if you ever cut any plastic with chlorine in it as it will make hydrochloric acid vapor).

Furthermore, I am still going to be heavily influenced in straitness tolerances to an extruded part even if I use the steel rails. I am really beginning to think that this missumi extrusion and car may be the most cost effective way to build a linear rail for laser cutters. They also offer steel rails and cars with steel wheels in a larger size rail system, but it is more expensive and (in my opinion) a large overkill for this kind of system. People get great results from cheap chinese lasers that have nothing more than square tubing for rails and plastic bushings for wheel/bearings. I want more than that, but I think this is a reasonable way to go that is going to be about half the cost of using v-groove rails with about 90% less headache and assembly complexity.
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

Re: Linear Bearing Idea

Postby bdring » Tue Sep 14, 2010 11:36 pm

When I bought my v rail I don't think the smaller one was an option. I used the smaller $3/foot on my z axis upgrade. There is no reason to use the bigger version other than the design needs to be rescaled. Regardless, this thread is not about endorsing the existing design it is about making it cheaper and easier.



Anyway, this is the cross section of a true angular contact guide bearing. There is virtually no play in any direction. It is almost a hybrid of a standard ball bearing and the thrust bearing. Note the center line indicating the line of contact. I am sure VXB gets them cheaper than I do, they just sell them at a profit ;). It this what Misumi has?
.
kiit8351-1.jpg
Bart
"If you didn't build it, you will never own it."
bdring
Site Admin
 
Posts: 2966
Joined: Sun Nov 22, 2009 7:33 pm
Location: Chicago, IL, USA

PreviousNext

Return to Build Logs

Who is online

Users browsing this forum: No registered users and 32 guests

cron