Carbon Fiber Bearing Rods

Has anybody tried this?

I've found this store that has ~8mm tubing.

it's lighter, which might mean we can use smaller motors or move the head faster with the existing ones.
it might be cheaper for similar tolerances.

There are two questions:

1) Is it rigid enough?
2) How long will it last?

1) This question can be easily answered. Someone with a machine needs to order some and replace their steel rods with CF rods and print a sample to see if it affects quality. I think rigidity would be fine for the Z axis, but would have little benefit except making the whole mendel lighter.
edit : answer is yes. CF tube with 8mm diameter is rigid enough for thermoplastic and paste extrusion.
Weight savings is significant: ~150 g less per bearing rod.

2) Steel will defiantly last longer. The question is will CF last for the life of the printer, or if not that long, is the shortened life of the bearing rods made up for in weight savings?

Edited 2 time(s). Last edit at 10/05/2010 06:40PM by Buback.

Hi Buback,

I have just CNC'd a Mendel Isaac, and I have carbon fibre rods for the x and y axis.
They are actually a little thinner (7.85mm) than the silver steel ones I was originally going for.. not sure if that is going to be a problem at the moment.

At the length that they are cut, they feel adequately rigid, but I wouldnt want to make them much longer without going to a larger diam.

I have not assembled it yet, but I will put some pics somewhere when I do...

Lucastar

Great news! You can be our guinea pig, unless someone else beat you to the punch and is using them in a working machine.

Are they solid rod or tubes?

Mine are solid rods.. I guess that tubes would be more rigid...

Not for the same diameter they won't. Remember that a rod is just a tube with walls all the way in. Tubes are better weight for weight.

I'm going to try extruded CF tubes, instead of rods. they have a 2 mm wall thickness, so i hope they'll be strong enough.

I'll post an update once my mendel is up and working.

I'm still waiting on some parts, but I got the CF tubes in yesterday.

wow they are light!
(I got 2 48 " lengths of tubing, so i had to cut the axies a couple mm short each. weight is for length listed)
x axis - 493 mm - 16 g
y axis - 404 mm - 13 g
z axis - 328 mm - 10 g

I have a length of chromed steel i got out of a printer that is 435 mm (so between x and y) and it is 171 g!

The main concern i have right now is the bar clamps cracking the tubing when it is compressed. when tightening the y axis in place i heard some creaking, so i stopped for the night. I think i'll go get some dowel and glue a 2 cm piece into the ends of the tubes, so that the wood helps resist the compression.

I'll post pictures later, but for now i'm off the the NYC maker faire!

So I have the tubes all installed. I did put some dowel in the ends of the tubes and so i can really crank down on the clamps if i have to. They are light enough, though, that they don't need a lot of pressure to hold them tight.

I had to put a 4-5 cm piece in the x axis ends, because the captive nut assembly clamps further down the x axis rods.

I had some concern about using wood, as it might swell with high humidity and crack the tube from the inside. I'm not too concerned, because the clamps will prevent that much swelling, but if it happens I'll post about it. Another option is to fill the end with epoxy, or cut up a metal shaft instead of wood (5mm fit perfectly).

Since I don't have steel bearing rods, I don't have a control, but when I get my electronics in and get everything calibrated I'll post about my print quality.

---
I'm thinking about the friction with these tubes. The one steel rod I do have (from an old printer) rolls much smoother. I would think there is less surface contact with the steel because it is much harder, and the surface is smoother. the steel is solid and not tubular; tubes will compress a tiny bit and increase surface area as well.

Overall, steel will have less friction. What i don't know yet is at what point would friction cause major problems. more friction will require more torque from the motors, but i don't think NEMA 17s will have any problems as long as ball bearings are used. maybe some axes could even get by with hardwood bearing rods, or plastic, since they are cheaper and much easier to cut (i'm thinking z axis, as it moves slowly anyway and it isn't load bearing)

Lost my camera, so still no pictures :-(

Does friction between the rods and the bearings matter? Isn't it the friction in the bearings themselves that is relevant.

---

[www.hydraraptor.blogspot.com]

yeah I think your right, although it does seem to grip a bit more. The CF tubing seems a bit tight between the ball bearings so maybe it just needs some adjusting for smoother movement.

I observed the difference by trying both in the x carriage. When i held the carriage level, the steel bar moved easily on it's own when i tilted it slightly off-level. the CF didn't move on it's own when tilted the same amount. but i'm now thinking that the mass of the steel rod has a lot to do with this observation. The heavier steel can easily break the static friction inside the bearing, whereas the CF would have to be at a much steeper angle. Also the adjustment thing I mentioned above.

I've taken some measurements and pictures.

I've worked out the angle at which the two types (CF and steel) start to move. Steel starts rolling at ~8 degrees off of level, CF starts at ~14.5 degrees.

My numbers aren't very scientific, mind you. the CF is tubular and cut to size, the steel is solid and 42 mm longer. However, I'm pretty sure that it's mainly the difference in mass between the two types that is causing the different angles.

Here a picture of my mendel with the tubes.

From reprap Mendel

This is a shot of the z axis end with the dowel visible.

From reprap Mendel

Not so sure about me being the guinea pig! - you are well ahead of me - and it looks great!

I have nearly all all my kit now (I decided not to start building until I had got the complete kit together...)

Just need my nozzle/hot end assembly now...



The mass of the rod will have a lot to do with the angle you have to tilt it, to get it to roll throught the bearings - try letting a 'carriage' of predetemined mass roll down the tubes instead...?


Lucastar

oh yeah, duh! I should have realized that. thanks for pointing that out.

You're Welcome

Actually, for true weight saving, I'd like to see a (folding?) Carbon Fibre Photonlabs style molding fitted with your CF tubes - linked to an Epad (running Android).
It would be a Lightweight Portable Repstrap - really special visually, and priced as regular Mendel.

I was originally thinking of a CNC birch ply frame with CF rods, but I like this much more

Next project lined up then....... race lightened reprap (....?)


Lucastar

I finally got the y carriage and bed installed, and I've notices that the rods bend a lot. Actually, it seems to only bend about 1 mm, if that, but there seems to be a lot of backlash somewhere in the system. I probably need to use more study materials for the bed and squashed frog, but my dad had the scrap wood at home.

1 mm is a lot of movement, though, so i'm not sure tubular material is the best for the y axis. I'm still not printing yet, so we'll see how it goes.