Voron 3D printer uses just one bearing per XRod... is that a smart choice?

Hi guys, just noticed on reddit about the release of a corexy 3d printer called voron and while looking at it it hit me that the XCarriage only uses one LM<#>UU (albeit something Mitsumi, not chinese) per XRod. Even Mitsumi or whatever brand LM<#>UUs have a certain amount of play in the bearings plastic tracks, which would translate to side to side rotation and possibly affect the print especially if the hotend is not straight in the center of the bearing (ex: dual hotend).

Have you seen the voron yet? What do you think (especially about my observed issue above)?



Edited 2 time(s). Last edit at 06/11/2016 07:21PM by realthor.

---

RepRap Lander concept on Concept Forge
RepRap Lander concept on RepRap Forums
My Things, mostly experimental stuff

* The distance from the center of rotation to the nozzle, as you your self made such a nice sketch about recently.

* Tooth side of belt running on smooth idlers

* 4 point bed leveling on a alu slab

* Bed with 2x Z-screw and those screws seems supported in a way much like the Prusa, which I don't agree with. In the current design with so many rods it is probably OK.

* Bed rods held by printed parts that extends some beyond the mounting points

* Belt tightening seems a little tricky

BUT funny enough, I think it is OK with just one bearing on the X axis

The printer looks nice though. A nice finish and many beautiful details in the CAD design of the parts. Not many CoreXY printers have the Z probe these days and I think it is nice they he had the time and energy to get it integrated into the design. Also the PSU is really nicely integrated.

If I was making a new printer today (and if I should make it after a design) it would probably still be Philipe's G&C, but this one would not be far down the list.

I have to say all this talk about play in bearings and nozzle centre of rotation is very nice in theory but my corexy has vertically stacked x rods and the nozzle is wobbly if you stick your finger on it and wobble it. It does not translate into poor print quality even at 175mm/s average speed it doesn't wobble when printing. So whilst a rock solid nozzle is desirable it is measurably not necessary.

I'd prefer one long lm8uu bearing to one short one, indeed that's what I have.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Quote
DjDemonD
I have to say all this talk about play in bearings and nozzle centre of rotation is very nice in theory but my corexy has vertically stacked x rods and the nozzle is wobbly if you stick your finger on it and wobble it. It does not translate into poor print quality even at 175mm/s average speed it doesn't wobble when printing. So whilst a rock solid nozzle is desirable it is measurably not necessary.

I have an incredible respect for the Dunning-Kruger syndrome. I paste the curve here. In simple terms, spend a little time with something and you feel like a master expert. Spend a lot more and you realize you know nothing (...John Snow...). Hah, I even recently set fire to one of my printers , I am definitely on the down ramp!




Totally ignoring that; This picture is of a part I made with a hotend that could really wobble, it is my believe that the deep lines you see of a single layer offset is a consequence of this. As long as the nozzle makes the same movement all the time it is displaced the same. For this reason part of the print looks OK. Then when it moves from one contour to another it "resettles" and gets these offsets.

PICTURE OF PRINT

That is at least my hypothesis and I haven't seen these types of lines on my newer printers which have more firm hot-end supports. I should say that on the printer that made the above it was pretty grave. Both the mounting point of the belts was much above the center of rotation and the hotend much below.


While writing this I came to remember another guy who posted about his printer in here some time ago .

He did post this link which I think shows very explicit what it is we are talking about, when talking about the hotend turning about center of rotation (it is a video but well worth the watch!) - [youtu.be]

Edited 1 time(s). Last edit at 07/17/2016 01:47PM by LarsK.

I've come across the Dunning-Kruger curve before. And I accept that a rigid hot end is very much desirable, my delta has a much more rigid hot end than my corexy and does produce sharper prints. But I think there are 3 levels of acceptability, 1 - doesn't work, 2-works okay, 3-works brilliantly. The issue with 3d printers is that many only need to get to 2, getting to 3 costs 10 times more time and money.

I read all about your printer fire, and have sympathy it can happen to anyone but I hope some extra safety routines can be built into reprap firmware I'm making a foray into duet wifi soon myself.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions

Quote
LarsK
Totally ignoring that; This picture is of a part I made with a hotend that could really wobble, it is my believe that the deep lines you see of a single layer offset is a consequence of this. As long as the nozzle makes the same movement all the time it is displaced the same. For this reason part of the print looks OK. Then when it moves from one contour to another it "resettles" and gets these offsets.

PICTURE OF PRINT

That is at least my hypothesis and I haven't seen these types of lines on my newer printers which have more firm hot-end supports. I should say that on the printer that made the above it was pretty grave. Both the mounting point of the belts was much above the center of rotation and the hotend much below.

That looks a lot like a sudden temperature change, for example from a shorter time to print a single layer.

The quality of the printed parts, and the fact that there are so many printed parts on it in the first place doesn't really inspire confidence in this machine.

For me any wobble in the hot end is an issue.

Quote
Mikk36

Quote
LarsK
Totally ignoring that; This picture is of a part I made with a hotend that could really wobble, it is my believe that the deep lines you see of a single layer offset is a consequence of this. As long as the nozzle makes the same movement all the time it is displaced the same. For this reason part of the print looks OK. Then when it moves from one contour to another it "resettles" and gets these offsets.

PICTURE OF PRINT

That is at least my hypothesis and I haven't seen these types of lines on my newer printers which have more firm hot-end supports. I should say that on the printer that made the above it was pretty grave. Both the mounting point of the belts was much above the center of rotation and the hotend much below.

That looks a lot like a sudden temperature change, for example from a shorter time to print a single layer.

Hey, that part is in ABS, not PLA. Also, and more important, the offset is consistent on both sides, the layer is plain offset with the line you see on one side and it extruding to the other.

In other news;

What do you guys think about this concept he uses here:

Might heat the centre of the bed quite nicely.

---

Simon Khoury

Co-founder of [www.precisionpiezo.co.uk] Accurate, repeatable, versatile Z-Probes
Published:Inventions