MakerBot Print is our newest print-prepration software, which supports native CAD files and STL assemblies,
allows you to interact with all your printers via the Cloud, and many other exciting new features.

Download Now

Hey! This thing is still a Work in Progress. Files, instructions, and other stuff might change!

LM8UU and LM8LUU Hybrid PTFE bearing

by Coder-Tronics Aug 27, 2016
Download All Files

Thing Apps Enabled

Please Login to Comment

I would actually expect a bearing with only 3 tubes to be the best. Since with 4 tubes you need to get all the variables so tightly correct.

this is genius! thanks

Hi Antscran

I just recently stumbled on this "thing" you made and I think it's a great idea. I already purchased a bunch of igus plain bearings for my 3d printer build, but I want to give your ptfe bearing idea a try instead. My main concern is longevity. Can you give any feedback on that ? How are they holding up in your test rig ?

I'm planning on using 8 4mm ptfe tubes per bearing on a 1/2 inch smooth aluminium shaft (made for linear bearings). I actually think the ptfe tubes might hold up better than solid rods because the tubes can compress and expand a little bit. This might also help with slight miss-alignment of bearings.

One of the reasons I want to use plain plastic bearings is because they should require less maintenance (in theory) so if they wear out fast, that would be a no-go for me.


Hi xyze,

I have been testing a variant of these which incorporates is kind of a hybrid incorporating the bearing housing all in one, it's slightly longer than a standard LM8UU though at 30mm.

I purchased a few plastic rods virgin PTFE, Acetal and a few others. Currently only had time to test the PTFE which provides excellent glide, but is very soft. With the virgin PTFE rod I can push my finger nail into it and leave a mark, it's does spring back slowly (see the reply to darr below for some images of these).

I fitted the bearings to my Y axis and have used them for the past few months, probably 5-6 days of solid use over that time. I noticed after a month or so that the Y axis rod with the single bearing showed small white flecks, which were obviously from the PTFE. The other 2 bearings were perfectly fine but as time went on the deterioration got to a point where there was a millimeter or so of play on that single bearing. So I removed them and the images below show the result.

The first image shows the worn area in red and the other 2 rods are fine, the bearing is flipped and all the weight of the X axis is placed on the rods highlighted in red.

The second image shows a view along the internal area of the rod and the wear can also be seen here.

Basically virgin PTFE is too soft as the weight of the heatbed, glass etc caused it to wear. However, it is possible to buy PTFE rods which have reinforcing agents and this is something I will be testing. I am not sure if PTFE tube is 100% PTFE, so this is something I am also looking into. I think the virgin PTFE still has a user case though, as a Z axis guide is far less likely to push on the bearing assuming the printer is not leaning or has skewed geometry.

I will also be testing the Acetal/Delrin rods soon.


Thanks for the detailed replay Ant

I did some research and testing on PTFE over the weekend and it turns out that PTFE is creepy. Not that kind of creepy:). Natural PTFE has no "memory" so it will deform permanently (creep) under load. I printed some test bearings using 4mm ptfe tubes. The bearings initially fitted very snugly over the 1/2 inch shaft and was hard to move. I left one bearing overnight on the shaft and it loosened up during the night.

After some more forced movement on the shaft the bearings moved very smoothly with no play. Unfortunately this means that they will loosen up further with any kind of load. It's really a bummer. I do think they might work well with very light loads though, but for now I'm back to the igus bearings.

Looking forward to your findings with the Acetal/Delrin rods. That might actually work well, especially if you use a Delrin/PTFE blend such as Delrin 100AF : https://www.ensinger-inc.com/products.cfm?page=product&product=delrinandreg;+af+blend+-+teflonandreg;+fiber+filled


It's an interesting idea and I was going to try this out, but I realised the common PTFE you mentioned is not the usual 4mm ones we use for 3d printers. Any reason why not use 4mm ones? I don't think it would be so exaggerating to crush....

Hi darr, absolutely and I am currently testing a varient on my printer using 4mm O/D solid PTFE rod, just not ready to release it. If you watch the end of this video I show the bearings briefly at near the end.

I was flushed by the fact that it uses the common PTFE so I presume it refers to the 4mm ones. I just did a sizing and realised the 4mm ones exceeded the wall limit just by a bit.....lol. I also just check the mechanical specs, the wall is actually 3.5mm only. I saw yours at the end of the video and I think maybe your PTFE is 3mm? The common 4mm ones are too big.

Hi, The ones in the video use a new one I am experimenting with that is bearing (LM8UU) and housing (SCU8UU) all in one, it's slightly longer at 30mm.

Couple of images of this with 2 variants; M4 knurled brass nuts and M4 standard nuts:

Still testing although all see the my reply to xyze (13/05/17)

I see. Thanks. Yep your closeup pics show a much thicker wall....but your idea is great, I will have to do a design using thinner tubes or filament.

I'm planning to try the 8.2mm variant with PTFE swapped out with some 1.75mm POM (Acetal/Delrin) filament.

It has a similar coefficient of friction as PTFE, but it's much more rigid.
Additionally, a solid "rod" should be more stronger than a tube of PTFE.

If I run into the thickness not being enough (1.75 vs 2),
Should I just try resizing the 8.2mm down slightly?

Btw if this works, then others can also buy 1mm of 1.75mm Acetal/POM filament for just $2 on eBay.

I have some Acetal to try as well in 4mm rod form factor, the tolerances on Acetal rod is much better than the PTFE I have found.

Comments deleted.

I think you will have a hard time forcing thick silicone into a small tube. It's not that firm anyway. I'm thinking that it would be better to smear some silicone onto some filament and shove that into the tube.

Also wondering if you need the tubing to be that long. A cm or so at either end would work just as well I think and would be easier to fill.

You could be right something I need to try at some point, shorter lengths of PTFE tubing are an option as well.

Love this design. I designed my own bearings using your idea for a camera dolly tha slides on 1" round aluminum rails.
I have been struggling to get a desin that was light weight and sid not involve a lot of components like wheels etc for the bearings.
This has worked very well for this purpose. I will post pics of the finished project when done. Thanks for your work!

Thanks Chris and glad it proved useful, funny enough I have made a camera dolly using this principle and 10mm carbon fibre rod, can be seen here:


I think Aluminium will provide an even better glide surface than the carbon tubes I used. Would definitely be interested in seeing your final camera dolly when finished.

Camera Slider / Dolly

Man that's a really great idea and PTFE is excellent stuff for all kinds of gliders, so I can't wait to get the ordered PTFE tube and to have them in use instead of my waggly chinese standard bearings.
I love the way you work out the tollerances to give the best possible result to everybody!

Do you plan 10mm and 12mm bearings too?
That would be great.

Hi FZS1_Fazer,

Thanks and hope it will be useful.

I have various other designs in the works taking into account all the feedback, using 3mm O/D and 2mm I/D thicker tube, as well as playing around with some PTFE and other solid plastic rods I have on order. Will make sure I do a 10mm and 12mm variant.

Unfortunately my Melzi board smoked a few weeks back I have a replacement, but waiting for some extra fans from China to ensure it doesn't happen again. I want to be able to print anything first before uploading.

Did you ever find the time to do these in 10 and 12mm variants? :)

Do you still need them? I can make them for you. Just give me an OD and a length you want.

Comments deleted.

as you can read i am clueless about bushings and bearings... but... do you thing that adding some sort of hole to put some lubricant would help ?

I think you might go down a slippery slope here......

I am afraid I cannot give a qualified answer on this subject.

For me personally I would say try to avoid lubricants as I find dust collects and I get dark lumps collecting at the edges of travel for my bearings, not sure what effect this has if it also collects inside the bearings. Without grease you could probably aerodust any parts out and this design could potentially benefit from that, but the main goal is to reduce weight and inertial forces.

man i am trying to make a smartrap, i live in mexico and... finding locally a 8mm shaft is super expensive...
All i can find in at a good price is 3/8" and... on the other hand, i cant get any cheap linear bearing in 3/8 all the cheap ones a are 8mm...

i wonder if you could help me to use your design but that fits a 3/8 shaft (out diameter at the same as Lm8uu, but with a inner diameter for a 3/8 shaft)

(by the way i hate milimetrical and standard stuff, i wish we could have only one, no two sistems) :(

(could you posible make a variation of your model for me?) :D

i saw a lot of parametric linear bearings here in thingiverse... but... i wonder how good is your idea compare to simple pla linear bearings with out the teflon hose... is the teflon hose a wooow factor?

best regards from your mexican fan


I made a parameterized version of this in OpenJSCAD which will allow you to easily edit the dimensions to fit whatever size rod you like.


PTFE Tube Based LM8LUU Parameterized Remix
by ryanarr

Hi Ferslash,

I don't have time right now to knock-up a design im afraid, but this is a possible option. 3/8" is 9.5250mm according to Google, so one route would be to scale up the design to fit the imperial linear shaft. However, consideration needs to be given to the other constraints of the bearing and the fitting that the bearing is seated in?

Most slicers offer a scale up facility Cura definitely doe's and I read in another post it also allows you to chop the design, so you can print smaller test samples.

I make the difference between the metric and imperial 17.4%, so if you can find out by what factor your slicer scales, then try the appropriate number. If anyone else reads this and has a better solution please feel free to correct me, as never scaled stuff and actually printed.

As for whether this design is good compared to others, well not for me to say and the comments here are testimony to that. I can see failings and good points, all will ultimately lead to a better design :-)

Hi MKSA, thanks for your comment.

I happen to think it's a good idea, so far in my tests it has performed very well. This is of course an iterative process so the design can be improved upon, the Igus website is a useful resource for problems they have overcome. I need to do some tests under load next as even Igus bearings have maximum load recommendations.

I agree entirely on the form factor, in fact my comment reply to O_Lampe 5 days ago stresses this: The main limitation here is the LM8 bearing form factor, but future version can be incorporated into a SC8UU housing or even moving away from these form factors, which is one of the empowering aspects to 3D printing.

Why nobody in the industry uses PTFE that way ? Do you think it has not been tried ? Why Igus developed their special blend of plastic ?
Besides, to use a tube is even worse as it will crush.
I use PTFE sheets for some cases but Igus plastic for bushings, linear or ball bearings in some cases. Better than the crappy chinese 8X22X7 that people still try to use everywhere on these hobby 3D printers.
Similarly, I don't care for the LM form factor. I integrate the bushings directly in the parts.

I suggest you do some reading about the use of PTFE, its qualities and limitations.

PTFE #3 of top 5 plastic bearings on metal (and it´s only #3 because of it´s costs) http://info.craftechind.com/blog/bid/259158/Top-5-Materials-for-Plastic-Bearings-Used-on-Metal-Surfaces

Hi rj3005,

Thanks for the link, I am testing 4mm Acetal and PTFE rods at the minute.

Read and UNDERSTAND the article, THEN you will understand I am right regarding these POORLY DESIGNED bearings proposed here !
So who made them and use them ??? Did you ?
I have been using PTFE for decades, I have sheets, blocks, bars in various sizes.
BTW, I see you have a Tarentula, which is a piece of junk (the Blwk Widow is a valid design), so go ahead, use these bearings ! An other thing, your "design" is yet an other stupid auto bed level sensor. A clear indication of someone who knows shit about mechanic. So do something smart, then come back.

You don't have to be so aggressive in your comments to provide feedback.

Why nobody in the industry uses PTFE that way ?

The industry does use PTFE that way. Lookup teflon bearings. The reason teflon bearings are not more popular is because it's relatively expensive. But if you don't think teflon bearings are any good, you should realize that they use teflon bearings in buildings to help steel beams slide against each other as the building twists. If teflon is good enough for building loads, I think it's good enough for 3D printers.

Why Igus developed their special blend of plastic ?

Just because Igus has developed their own plastic doesn't mean that there aren't other good alternatives.

Besides, to use a tube is even worse as it will crush.

This is a valid criticism. You could try to address this by making the tolerances around the tubes very tight, which would then leave no space for the tubes to deform and it would then take tremendous loads to crush them. The downside is that it would be difficult to get the tubes in.

Similarly, I don't care for the LM form factor.

The form factor is useful to a lot of people. I'm sorry that it's not useful to you, but there's no need to berate someone providing their work for free. Feel free to make your own designs around his freely provided work.

Nice to hear it wasn't just me thinking this reply was a little on the negative side, any how some positives have come out of it so all is good.

Cheers Centenary.

I was really amazed that you replied in such a level-headed manner. Thumbs up mate

Always best to keep your cool in any sort of discussion or you loose sight of the debate, I also like to remember this quote "Never argue with an idiot. They will only bring you down to their level and beat you with experience"

Obviously you failed to understand what I wrote. Anyway, here is an example of someone who is not an idiot, for he understands pretty much what has to be done, yet should investigate further regarding the surface contact, deformation, creep, use of pure PTFE and its limitation for 3D printers bearings ..... https://pinshape.com/items/9235-3d-printed-cheap-linear-bearing-with-ptfe-semi-printed-lm6uu
At least, his solution is usable.
As for your fantastic idea, go ahead, install them, use them and come back ! In fact, no one should be allowed to present something unless fully reviewed, validated, tested .... The "likes", "this is great" etc... have no value in engineering.
So much free crap that the really good things are lost..
BTW: I did not "argue", I stated facts. Oh, I have used PTFE before you were even born.

Here be a troll. Each reply gives him an organism.

Wow, after two years a moron pops up out of his hole to make a personal attack he think funny, when in fact it is because he can't argument !
These wonder bearings are still nowhere to be seen, being quickly replaced by regular or Igus one after a few tries., Yet because they go the way of the Dodo bird, they gets reinvented now and then with the same lame results.
Mine, made out of Igus filament have been copied over and over and used successfully.
Now let's see what this new genius will "design". A spool holder, a filament guide, an other lame bearing made with soft drink straws may be ?

You sir, are a great example of how not to communicate. Thanks for that, you have reminded me to give new things like this concept a chance and try them out, if they don't work quite right then offer constructive positive encouraging criticism. If I don't I might sound a little like you, which should be avoided at all costs!

Thanks again!

Go ahead, make them and USE them. replace all the LM of your 3D printer or whatever with this "thing" and come back after extensive use.
No "constructive positive criticism" ( laughable) required for bad ideas. In my time, you were just kicked off. Not for a bad idea but for presenting it without proper analysis, testing We were aiming for RESULTS, FACTS, NO "wow brillant" " wow nice" "great" !

You project your emotional fantasies constructed to boost your ego as if they are truths, not very logical. Does not make me believe you are capable of recognizing results or facts over learned knowledge that may or may not be correct but you are cost loss invested in to utterly irrational extents.

Free energy, that sounds great! Got a link?

also, i checked your profile you have zero contributions and are obviously a troll, either emotionally disturbed teenager pretending to be grandpa or maybe paid by some other site to just make things lame as you can, whatever, I'm trolling you back and having fun doing it but done now, bye have a great weekend :)

On the reprap forum, I posted how I improved my Geetech Alum Prusa I3. You can still see some of it although I am not registered anymore. Same handle. What I encountered was a very few knowledgeable people (DC42, digital-dentist ...) and a majority of jerks reinventing the square wheel, congratulating each others for their underachievements and unable to accept they are wrong. People claiming to be "designer" who can't measure a torque, compute a beam deflection, yet pretend to achieve "micron" accuracy (they don't even know what accuracy is).
I don't mind people making mistakes. I mind when they spread them leading people to waste their time and money trying these stupid ideas.
BTW,my Prusa prints pretty well at 100mm/s. Only the frame (stiffened) is still from the original kit. The last big improvements was to replace all the bearings by my own made from Igus filament. Now time for something different.
No contribution here, indeed I registered recently. Is it worthwhile though ? I wonder.

You can still see some of it although I am not registered anymore.

Sounds like another way of saying banned. Judging from some of the toxic comments you left there, I wouldn't be surprised.

DC42, digital-dentist

You should strive to be more like them. They know how to leave constructive feedback without being toxic.

I don't mind people making mistakes. I mind when they spread them leading people to waste their time and money trying these stupid ideas.

You're free to leave feedback, but why be toxic about it? And it's only stupid in your opinion, I don't see why this would be stupid with solid rods.

BTW,my Prusa prints pretty well at 100mm/s. Only the frame (stiffened) is still from the original kit.

Doesn't seem unexpected once you've replaced everything.

All my remarks were about poor design/implementation and became "personal" when people started it ! I refrain from calling someone an idiot unless confirmed by his stubbornness and in case of personal attack.
All what I said proved to be true, either their "design" failed miserably, either the "designer" refrain from building it. Computer animation is just that, an animation !

You find "obvious" that by replacing parts, we gain performance ? OK, go ahead ! BTW I replaced about everything by MY OWN DESIGNs that I made and tested BEFORE presenting them. AN plenty more since. I am quite pleased by the Igus filament and the freedom it gives in term of integration.

Anyway, the Prusa is a dead end. The two Z lead screws and motors is an issue easy to fix (although most fail to recognize it or use improper solution), the moving bed prevents higher performance. Plus it is open. It is a good starter kit though for FFM.
Anyway, looking for something different now.
An other remark, most in the current reprap community (the "founders" were smart, real engineers, read the old posts) are just "artists" without a clue about what it takes to design a REAL machine. Fine if they have fun. There is also a few crook selling junk kits ! They didn' like it when I told them.
Good luck.

Obviously you failed to understand what I wrote.

Actually, you're the one who failed to understand what Antscran wrote. He already stated that he's thought about the deformation issue and he's thought about trying solid extruded rods. He even linked to the solid extruded rods. That would be a better solution than what you linked to due to reduced surface contact. If you failed to read that, that's your own fault.


I think it's funny you think creep is a problem at 3D printer loads. A look at any of the published creep curves would show that creep is pretty much irrelevant for 3D printer loads, even for pure unfilled PTFE.

Hi MKSA, thank you again for your comment.

I know it's been tried in the industry as liners and coatings for bushings etc, Frelon is something that works well here. These are however industrial applications with forces generally far beyond your standard hobby printer.

Crushing I have already thought about this and need to do some load tests, but a couple of options; Increase the diameter of the PTFE tube and PTFE extruded rods.

http://www.theplasticshop.co.uk/ptfe-rod-4mm-5mm-6mm-8mm-10mm-12mm-16mm-20mm-25mm-diameter.html There are plenty of other sources.

The LM form factor is restrictive for this design but it's something people have parts for generally, pending on the forces and weight of their end application crushing may not be an issue with 2mm tube. You can design around and manufacture to your requirements, even with a hobby 3D printer.

Crushing should be not a problem is you just stick a piece of filament into the PTFE tube.

The filament idea got me thinking, maybe you could simply inject household silicone into the PTFE tube or a caulking compound. Only really thinking of using this for the smaller diameter tube, as the smallest diameter PTFE rod I can find relaiable is 4mm.

Not a very good idea. PTFE and of course in tube form is not suitable. People who thinks it is a good idea are welcome to say why and try :)
Igus sells a filament that works pretty well. Of course, no way to get a proper fit and surface quality by just printing. You have to make the parts oversized and adjust them. Files and reamers are cheap. Works fine, hold well. Friction coef is greater than PTFE but iglidur does'nt creep ! And simple SS rods supplied with most kits which are totally unsuitable for LM bearings are fine with these bushings.
Oh and no need to focus on standard LM form factors, just design the parts directly to accommodate the smaller bearing. For ex. for 8mm rods, proper sizes are OD 12mm L 12mm. I saved 80gr on my Prusa X carriage and 160g on my Y by just repalcing the LM!

Hmmm, cloth irons are coated with PTFE for better gliding, thats why it is a good idea IMHO

When I have made printed plastic bushings for drone projects ... I found I could control the internal clearance by intentionally designing the bushing too tight, then carefully heating the assembly (oven / heat gun), allowing some limited stress relaxation of the plastic .. if done right, a perfect fit can be achieved (until the bushing wears from use, of course).

How about adding a "screw thingie" on the end that wedges the assembly, to loosen or tighten the bushing fit on the shaft ... then one could adjust for wear, which is the biggest detractor with using bushings over linear ball bearings. Bushing would have have to be 2 or 3 printed parts, where they wedge against each other by rotating one piece on a screw thread.

Hi BigCooter,

Some interesting ideas there, will have a think along with some of the other comment ideas and see what I can incorporate, if not me I am sure someone will remix this.

Good idea, but I'd make one version that is adjustable instead of so many versions. Just add an axial slot and two adjuster screws to clamp the housing.

Cheers, yep agree an adjustable version would be better, this is just the first iteration and should allow someone to find the model that fits and just print it.

The main limitation here is the LM8 bearing form factor, but future version can be incorporated into a SC8UU housing or even moving away from these form factors, which is one of the empowering aspects to 3D printing.

These look pretty good! I wonder if some sort of twist-on retaining caps could be added to completely prevent the PTFE tubes from sliding out of the bearing housing.

Hi thanks,

I don't think they will slide out if you get the right balance when printing. The one I show in the video has a small amount of resistance when placed on to the rod, the PTFE tube is therefore under a small amount of compression and is also snugly fitted into the 2.2mm O/D tube guide.

Time will tell ultimately and I can tweak the design pending on feedback!