are you happy with the quality of your bearings? i'm just worried about buying from the cheapest ebay store i can find in terms of quality. Have you heard of any horror stories about cheap ebay bearings?
What kind of steel did you get and what was the cost?
My design uses a flying gantry of smooth rods/bearings supported inside a mdf box. The z axis will be be supported with threaded rods and smooth shaft guides. I plan to make most of the supports for motors and such from mdf while using a printed extruder and a kit hot end, which if you have any advice on i'd love to know.
i plan on using it for cnc routing too as it has a pretty large possible build area (450mm by 450) which would multiple toolheads.
from what i can tell the cost of axes should scale well and i am skimping on a lot compared to other designs.
If you are using a gantry you need to use larger rods, say 10mm or 12mm as the weight of the garnty will be a factor
Also if you are going to use lead screws on your cnc thing you need to acount for that they will be slow for 3d printing and belts will be to fast for cnc.
I have brought heaps of cheap bearing form ebay and only ever had one bad one.
As for rodding you are best to use 304 stainless stell.
You should also use 16mm mdf but Alum would be better.
Hi Russ, It sounds like your design is fairly similar to the one that Nophead is developing. Have a look at his blog.
I'd second Auzze's comment wrt rod size. With a build platform that size, you should really look at 12mm rods. The deflection in the centre is going to be quite large with 8mm rods, even with quite a light extruder.
For your printer, it's preferable to use stainless steel (any grade is fine). The mild steel ones will corrode, and flex too much. It's also much harder to get smooth rods which are completely straight.
Edit: on second thought, it sounds like your design 'flying gantry inside mdf box' sounds like some of the Darwin-derived printers. I think that Jos at mauk.cc is making something similar.
Edited 1 time(s). Last edit at 01/10/2012 04:22PM by Shibboleth. ___________________________________________________________________________ systematictechnology.net
> For your printer, it's preferable to use stainless
> steel (any grade is fine). The mild steel ones
> will corrode, and flex too much. It's also much
> harder to get smooth rods which are completely
I'm always intrigued by people's recommendations to use "stainless steel - any grade" in preference to mild steel rods. Is this based on experience and research or just presumption?
For the record, pretty well all grades of "steel" (mild, stainless, hardened tool steel, etc) have pretty the same stiffness / deflection characteristics as each other, because the Elastic Modulus / Young's Modulus of all grades of "steel" are pretty much the same. (Assuming we are talking about "small" bending deflections well below the yield strength of the material - and I think we can agree we don't want to see "large" bending deflections on any of our axes!)
If it's wear that you are concerned about, then hardness is what matters. Hardness depends upon both the alloy you choose AND its heat treatment / cold working, etc. Many common stainless steel grades are actually quite a bit softer than mild steel - if you really want to minimise wear rates, you are probably looking for a hardened shafting steel - which is not the same thing as common grades of "stainless steel".
For corrosion, stainless steel and bright shafting steel will outperform mild steel - but in a typical indoor environment, a light coating of sewing machine oil in conjunction with the constant motion of the bushes / linear bearings will keep even mild steel polished and "rust free".
For the record, I built my Prusa machine using "unidentified" 8 mm steel rod (definitely not stainless or shafting steel). I first fitted it with standard PLA bushes [julianh72.blogspot.com] - they were a bit stiff at first, but freed up nicely with a bit of machine oil and manually running them up and down the rods 20 or 30 times. I have since upgraded to LM8UU linear bearings on the X and Y axes [julianh72.blogspot.com] on the same 8 mm rods, and they are running REALLY smoothly! OK, I have only been running this way for a few months, but I honestly don't see any real NEED for stainless steel or shafting steel rods. I can always upgrade later if I see any problems develop.
Edited 1 time(s). Last edit at 01/10/2012 03:10PM by julianh72. Follow my Mendel Prusa build here: [julianh72.blogspot.com]
Julian, you're right in that mild steel has the same (or less) deflection for a given load than stainless. I must have had my brain switched off in the post above.
My recommendation for stainless comes primarily from the fact that it seems that the manufacturing tolerances on stainless are higher - it's much easier to find (in my experience) perfectly straight stainless rods than mild steel.
I don't think any of the other concerns from mild vs stainless - corrosion, wear rates, etc. have a significant bearing for 3D printers.
Auzze: Thanks for your advice on the bearings. I guess if Chinese bearing weren't good enough quality they wouldn't be used in anything else.
Thanks everyone for your inputs on the steel. I was looking at the rail from here [stores.ebay.com.au]
but I will try some local steel suppliers instead.
> on second thought, it sounds like your design 'flying gantry inside mdf box' sounds like some of the Darwin-derived printers. I think that Jos at
> mauk.cc is making something similar.
The design uses the same axes layout as the darwin, but instead of using four threaded rods at each corner and trying to connect them using a belt and associated hardware, i'm thinking 4 rods at each corner and then 2 threaded rods each powered with a stepped motor to lift the build platform at each end.
With the bed being lifted in the middle, it seems like there is opportunity for the bed to tilt either way unless the bushings on the smooth rods are extremely tight, allowing no slop. With the micron-tolerances that we are attempting to achieve with these printers, a tilting bed will rapidly throw a print out.
I've thought that the 'box with gantry' is in many ways an ideal design shape for a printer. My thoughts were that it might be better to use four motors, one on each corner. Pros: accurate, no belts. You could also set up auto-bed levelling. Cons: Expensive for all those motors.
Now that you mention it that is a very good point.
After looking at the makerbot designs I thought that one could enlarge the design by supporting a platform at both ends.
I'm now thinking i would need to place extra bearings axially to the 4 currently at the corners in the platform.
Looking at the just released Makerbot replicator, it looks like they have changed to a gantry mechanism for supporting their extruder. Which if it can support dual extruders must be able to take a bit of weight.
It's interesting, looking at the MB Replicator, that they've chosen to use a cantilevered print-bed. The cantilever material would need to be super-stiff to resist sagging, particularly under the weight of a heavy print.
The 6mm linear bearings I bought off a Chinese EBay seller for my mini-Prusa were terrible. The first one I tried, all the balls fell out as soon as I put it on the rod. The rest seem to be holding together but they run REALLY rough. I've tried oiling them, I've tried silicon grease, I've tried smoothing the rod, I've tried 5 minutes of running them along the rod in the hope that they settle down...nothing has worked. It actually feels like the balls aren't quite spherical or something. They're probably usable if I keep my belt tension really tight and deal with backlash issues later, but I think I'm going to write these ones off and get some better quality ones from a local supplier.