Y axis linear upgrade.

Another source of metric rods is VXB.COM. On their website go to Linear Motion Systems, and then Metric. They have several lengths of 8mm rods and the LM8UU bearings. Located in Southern California.

 

back from vacation, checking into that source.

 

I've used VXB before. They were good.

 

As someone who knows next to nothing on how to upgrade a 3D printer yet has been having a lot of prints ruined due to Y-Axis shake lately (lost a couple of smaller pieces and had to figure out some fixes; refer to "knows next to nothing"), how easy would installing this mod be?

 

@Electrocutioner - How much of the original Y-Axis range of motion are you trying to preserve?

 

I don't see why it wouldn't be possible to keep the same range of motion as before.

 

@Seshan,

Right now we have about 10.5" of travel. Spec is 10". Heat zone is about 9" .

Glass bed is about 15", minus 2 x 0.5" for end supports, minus about 3.54" (9 cm) for edge to edge on the linear bearings gets us back to about 10.5". So I guess it is doable.

One concern I have is that 3.54" span of bearing support means the bed gets heavily cantilevered at ymin and ymax. In some ways, this is no different than a CNC mill X axis, where we only care about the position under the mandrel. Still we don't have the same rigidity as a steel mill. I guess if it works it works.

Has anyone adapted the thingiverse project to 8mm and a glass bed? Did 8mm workout ok? Since the rails move with the bed, I am nervous about going to 10mm because we are adding even more mass. Opinions?

Prusa i3 has massive stationary Y rails. It doesn't have the cantilever issues, but it does have much longer span to cover. Mounting the bearing on the bed is an option for the Robo 3D, but the long rails would destroy the small foot print.

 

I'm looking forward to this mod. Every print i have has a slight wave to it. Can you also incorporate bed leveling?

 

Stas This mod is targeted at improving the Y Axis. Is your wave only along the Y? Also double check your hobbed bolt that the teeth are cut uniformly. I am still waiting for my eBay replacement to arrive.

Not sure the thingiverse attaches well to the glass bed. Incorporating leveling springs would be a good idea.

I'd still to hear about results of anyone who has done this upgrade with 8mm hardware.

 

I'll be doing this upgrade (or the official Robo 1.2 upgrade in the not-too-distant future) but I've got enough on my plate with the EZStruder and E3D upgrades in the short term.

I'm hoping that means that someone takes the time to do a redesign before I get a chance to.

 

Any strong preferences against 8mm rails which would mounted to the glass?

Does the non heated bed have the same dimensions and hole pattern as the heated?

 

Nickster, the wave its all the way around and uniform, repeating every 10 layers or so. I know i need the bed springs. but the Y rails are weak and slightly bowed.

 

I'm going to be doing this mod (or something very similar) soon. Problem is that my Y rails are so messed up that I probably can't print those pieces within tolerances. With that in mind, I looked into all-metal options and decided on mountable open bearings riding on a 12mm rail. My order arrived yesterday, but the seller accidentally sent me a 20mm rail (which they sell for $200, so I'm debating whether I should return it or keep it in case I ever need it in the future). After doing some more thinking, however, I realized that my method wouldn't work because I'm restricted to the height of the y-axis belt. The bearings don't even fit in the space between the birch and mounting plate and that's not even considering the height of the rail, so it looks like I'll be printing after all. I guess I may be able to get a usable part if I print at 10mm/s

 

Leon is working on something here http://forums.robo3dprinter.com/index.php?threads/leons-printer.1227/page-2#post-10039

I'm also playing around with some ideas, there are numerous obstacles that need to be overcome for the upgrade to work properly. Like you mentioned the Y axis belt, that could be solved with some spacers and longer bolts, or a newly designed bracket. Also y axis movement needs to be considered. Mounting something in the original holes will make you lose some Y movement. I thought about mounting the rails to the robo and the bearing to the bed, but then you would need the rails to stick out pretty far, it just wouldn't look that nice.

 

It would be a whole lot more stable if the rails were mounted to the unit though. That actually got me to thinking... If you were to just use a longer belt, could the RoBo's Y range be extended? There's a project I'm working on (which is awesome and will possibly make its way to Kickstarter in the future) and getting a 12" print bed would be a godsend.

But I digress I'm thinking the best all-metal upgrade would be a very wide, flat rail. The car goes on the printer and the rail goes on the bed. I noticed in one of RoBo's recent videos that the heatbed is now 15" long and they've done away with the birch, though I may be mistaken there. If it's true, I'll have to remember to order one with them.

 

Oh, and there is a workaround for putting the bearings on the printbed: using a third bearing on the center of side. That way you have enough stability for the bearing that escapes the rail to still be aligned when it comes back. Yeah, probably a terrible idea

EDIT: ...Unless you were to either taper the rail or flare the bearing...

EDIT 2: Or, you know, line the entire bed with bearings. I'm done with this train of thought now, I promise

 

This may be a good option, by the way: http://trinitylabs.com/products/linear-rail-and-car-418mm-no-mounting-hardware

Nice and wide, and it looks like it'll leave plenty of room for anti-vibration bushings.

 

I was getting ready for a Y rebuild, but ended up rebuilding the Z and X. My worst rail was 0.09mm undersize , which is only useful in a stick welder. C*R*A*P. With the X belt held tight, I had over 0.38mm of lateral movement on the opposing side of the X carriage. With new VxB rails, the slop is not perceptible. Y stepper and idler bearings and Y belt were loose. Black base plate shows evidence of cold plastic flow in many spots. Large fender washers on Y idler bearings were indented in the plastic. Why would you make a 9mm hole in the base plate for an 8mm idler bolt? Ended up machining an aluminum fender washer replacement that has a 9mm flange and 7.82mm ID to engage the hole so the idler can't move. I'll start a new thread on a precision Z coupler upgrade. Also cut a new hobbed bolt. Replacements from Amazon and eBay were non uniform or had tiny sharp teeth that clogged resulting in non-uniform erratic filament feeds. Maybe I should start a thread on Nickster's Robo.

But lastly, one piece of info is that linear bearings compress when they get highly force fitted into the X carriage causing binding on precision 7.998mm rails. This is relevant to a new Y carriage design if using rails and linear bearings. Fix will be to go for a snug fit for the linears in the plastic housing and glue the end sections of the linear bearings into place with silicone RTV. So in a carriage with two linears, only the outer parts of the bearings will be tightly bound and they can be made to self align on a rail until the glue sets.

Leon saw your Y thread. Looks like the right direction. Right now, I am going to back burner the Y upgrade; just doesn't need it. Going to be looking at dual extruder fans and ducts, and a silicone jacket around the extruder. Lets me increase airflow to reduce part warpage and upturned edges. Putting down filament on top of hot gooey plastic under layers is a recipe for unhappiness.

 

You're lucky you don't have to replace the rails, mine move up/down about 1-2mm when they slide out, makes it impossible to level the bed.

 

yeah...about that. I have only been checking surface quality, smoothness and uniformity. Was expecting jiggle on Y faces if the slides were bad. Haven't noticed a big variation, but have not measured it yet. First layer has been pretty uniform from HOME to center of bed. Do you have a favorite measurement method?