426 (edited by pirvan 2014-12-05 00:34:07)

Re: Odd circle test results

jagowilson wrote:

And the quality of the rod is no question (I mean, it's MC), but LM8UUs are subject to exteme variation. I had to throw away 4 of the 12 I got because they would just snag.

Just because you bought something from McMaster Carr, it doesn't necessarily make it the perfect part.  Let me ask you guys a question:

Did you buy the right thing??

What is it that you bought from McMaster?   Did you buy a Rod or did you buy a SHAFT?  There is a world of difference between those two. 

The rods are intended to be used as steel stock to make tools out of them.  They are meant to be cut, shaped, sharpened, and hardened.  They are NOT meant to be used straight up as shafts for bearings.

The shafts are rods as well, but their tolerances are much higher, the surface is polished and hardened, and they have a tighter straightness tolerances.

So is this what you bought?  It should be smooth to the touch as if it was chromed.


Edit:
I do agree with the LM8UU assessment.  Most of the ones available for $2 from eBay or some of the distributors in china are crap.  I too bought a bunch of them and ended up throwing away half of them.  I even had a couple of them that got distorted and loose with long term use. 

I ended up buying my bearings from Misumi USA.  Their prices are are excellent, and so is the product quality.  Their bearings equivalent to the LM8UU are about $6.00, which while it nearly double or triple the price of the cheap Chinese knockoffs, is still less than half of the McMaster Carr cost of about $16-18/bearing.

To print or, 3D print, that is the question...
SD3 printer w/too many mods,  Printrbot Simple Maker Ed.,  FormLabs Form 1+
AnyCubic Photon, Shining 3D EinScan-S & Atlas 3D scanners...
...and too much time on my hands.

427 (edited by jagowilson 2014-12-05 00:32:55)

Re: Odd circle test results

I bought hardened steel shafts. I didn't do any of this without the guidance of other forum members. The rods are 8mm all the way. No real detectable deviation. mcMaster has high quality everything. Nothing on mcmaster is bad. If it is they will fix it because it shouldn't be. It's expensive for a good reason.

If anything is bad on my end, it's the bearings. If you didn't know LM8UUs have a high failure rate (seriously like 25%) you are a fortunate person! The thread for the lawsy carriage says buy 12 when you need 8 for a good reason. Some just suck. I didn't get my bearings from MC though because $18 each is ridiculous. I'd rather buy cheap crap and throw away half than spend that much. I paid a buck each for mine from RRD.

428

Re: Odd circle test results

Okay we had a hard time keeping up with each other's edits but yeah long story short the bearings we have aren't ideal smile can you share a link to those $6 equivalent LM8UUs?

429

Re: Odd circle test results

I think my bearings suck.  Damn chinese knockoffs from ebay.  12 for $10.  I've not replaced the original SD rods so I'm still on stock rods but I've measured them up and down.  The rods which I've been demonstrating are actually 7.92 to 7.98mm in diameter (stock x-axis rods).  These bearings do not slide on their own.  As soon as I take the force of my finger off them, they stop; even when vertical with nothing on them but their own weight.

If these bearings don't work out, I'm going to have to invest in some higher quality bearings.  I'll be looking at Misumi USA as well.  I think Privan is talking about part number LMU8.

430 (edited by Claghorn 2014-12-05 01:51:30)

Re: Odd circle test results

pirvan wrote:

I ended up buying my bearings from Misumi USA.  Their prices are are excellent, and so is the product quality.  Their bearings equivalent to the LM8UU are about $6.00, which while it nearly double or triple the price of the cheap Chinese knockoffs, is still less than half of the McMaster Carr cost of about $16-18/bearing.

Just out of curiosity, I visited their web site. I would never have imagined there were so many different ways to configure linear bearings :-). You don't happen to remember the specific Misumi part number do you? I did go through the configurator and come up with one example price for part LMU8 that was $5.85 which is pretty close to the $6 you mention, so maybe that's the one.

431 (edited by n2ri 2014-12-05 02:40:07)

Re: Odd circle test results

has anyone noticed these bearings also have seals on ends (least mine do) to wipe grease/dirt off rods? these are snug to rods and will cause slight drag on dry rods enough to maybe not let them glide by gravity from 20% tilt.

but also should not keep from moving with slight touch or flick. grabbing and pulling is too snug for stepper motors, belts, pulleys etc.

also dont force them on ends of rods. smooth ends with emery cloth to remove burrs etc so bearings dont get damaged.

Solidoodle 2 with Deluxe kit cover & glass bed with heater. and 2nd board SD2 used not 3rd and alum platform not installed yet still wood. also need cooling fan installed to board. use Repetier Host couple vers. Slic3r also have all free ware STL programs

432 (edited by pirvan 2014-12-05 03:44:37)

Re: Odd circle test results

jagowilson wrote:

Okay we had a hard time keeping up with each other's edits but yeah long story short the bearings we have aren't ideal smile can you share a link to those $6 equivalent LM8UUs?

Just got home.

Anyway, I looked at my purchase history, and I bought the LMU8.  So I used the configurator to find them:

http://us.misumi-ec.com/vona2/detail/11 … 71%3A%3A24

To print or, 3D print, that is the question...
SD3 printer w/too many mods,  Printrbot Simple Maker Ed.,  FormLabs Form 1+
AnyCubic Photon, Shining 3D EinScan-S & Atlas 3D scanners...
...and too much time on my hands.

433

Re: Odd circle test results

Excellent! I will likely pick some up because I am obsessed with doing anything within my power to improve my printer's drive system. Thanks pirvan.

434

Re: Odd circle test results

http://youtu.be/e6A959u0SCo

I made a video of the bearing movement on my stock SD4 shafts.  These are chinese bearings.

435 (edited by jagowilson 2014-12-05 04:31:29)

Re: Odd circle test results

Goodness those are some bad bearings sad where did you get them? Your stock bearings don't slide that poorly do they?

436 (edited by pirvan 2014-12-05 15:02:00)

Re: Odd circle test results

Wow, that's truly bad.  I just did a quick video of how a proper linear bearing moves on a hard chromed rod:

To print or, 3D print, that is the question...
SD3 printer w/too many mods,  Printrbot Simple Maker Ed.,  FormLabs Form 1+
AnyCubic Photon, Shining 3D EinScan-S & Atlas 3D scanners...
...and too much time on my hands.

437 (edited by nbourg8 2014-12-05 19:14:28)

Re: Odd circle test results

In an effort to save on shipping costs, I am looking at the Misumi 8mm Shafts as well. Apparently their shaft pricings are the same price as McMaster Carr's.  I'm looking at shaft model PSFJ8 @ 358mm and 334mm for the y/x axis shafts.

Ordering 10 bearings and the 4 shafts comes to $103.  Damn....

Should I focus only on the bearings?

438 (edited by jagowilson 2014-12-05 19:34:09)

Re: Odd circle test results

If your rods are not damaged in any way, are straight and are of reasonable consistency in diameter you will not need to get new rods. Consider that more of an upgrade for later.

439 (edited by redbarret 2014-12-09 21:51:46)

Re: Odd circle test results

So I got my flanged bearings today meant to replace the bushings on the sides and the pillow block (Solidoodle 4).
However, I can't get the bearings to pop off with all of my force. For those who have done this mod: how did you get the bushings out? Maybe there's a drop of glue used to keep them in place, I don't know...

Solidoodle 4

440

Re: Odd circle test results

redbarret wrote:

So I got my flanged bearings today meant to replace the bushings on the sides and the pillow block (Solidoodle 4).
However, I can't get the bearings to pop off with all of my force. For those who have done this mod: how did you get the bushings out? Maybe there's a drop of glue used to keep them in place, I don't know...

They're pressed in, so don't be surprised they're tough - my method was to get a piece of wood with a c. 20mm hole bored in it, sit the whole printer on the bench on it's side with the bushing located in that hole, and then tap it gently with a hammer from the inside until it pops out.

SD3. Mk2b + glass, heated enclosure, GT2 belts, direct drive y shaft, linear bearings, bowden-feed E3D v5 w/ 0.9° stepper
Smoothieboard via Octoprint on RPi

441

Re: Odd circle test results

Maybe not a linear bearing quality problem,
Check tolerance of shaft and bearing from misusi,
Maybe your shaft diameter too big.
Use the ball bearings or flange bearings instead of
SD stock bushings, doesn't make a sense, because
The SD frame is made by 1 mm thk. Sheet metal,
It is not rigid as a solid metal bars, and all holes
are not aligned up well, so use GT2 belt and pulleys,
It is a better way to reduce the backlash on Y drive.
Of cause if you Like to make a big mod, to move
Y motor to outside of the frame, and drives 6mm
Rear rod (shaft) directly, it is a best way to reduce
The backlash on y direction.

442

Re: Odd circle test results

I've replaced my bushings for the drive rod with flanged bearings.
I'll post photos and how I did it soon.

What I'll say now is I can tighten the Y belt as much as I can by pushing the Y motor down and I don't get step skipping at all.
That's what the bearing upgrade has allowed me to accomplish.
So naturally I tightened the belt as much as I could to get rid of the backlash and I see improvements.
http://i.imgur.com/ZHCnkMp.jpg

However, this is very odd but I still see backlash.
There's still room for the Y motor to go down but I'm pushing it down as much as I can.
I also tried the bearings upgrade with pirvan's tensioner mod to add more tighteness to the belts that way (and the photo above is with that mod) but it still doesn't seem to be enough.
Odd.

I'd try upgrading to a gantry which use bearings as well ("lawsy's carriage"), but I don't see a point right now as I don't reach a point again where I get step skipping.

Solidoodle 4

443 (edited by jagowilson 2014-12-10 17:27:57)

Re: Odd circle test results

Those look good. Let me know what you did to replace them because I'll be doing this soon. The lawsy carriage would remove a lot more mechanical play (particularly on the right carriage piece) and should yield further improvement. My circles look about like yours now so the carriage upgrade should get you close enough that you won't have issues with tolerance.

Make sure there's no wobble anywhere in your hotend as well.

444 (edited by redbarret 2014-12-10 17:49:50)

Re: Odd circle test results

About using a carriage with bearings, sorry, I disagree. Here's what likely happened: you switched to a carriage which uses bearings and I switched to bearings in the drive rod. For both of us bearings allowed to decrease friction so we could tighten our belts some more to reduce backlash without getting step skipping because of it.

Maybe I need my Y belt even more tight, but there's no point in switching to a carriage with linear bearings for that because I don't have an issue with friction/step skippng right now and so don't have need for smoother movement of the carriage.

I can't think of what else to try right now though.

Solidoodle 4

445 (edited by jagowilson 2014-12-10 17:55:08)

Re: Odd circle test results

Friction anywhere is a problem and if you want the perfection you seek linear bearings are a requirement. The stock carriage has plenty of friction. You've just never felt the smooth movement of linear bearings so you don't know what you're looking for.

Keep in mind there are plenty of other potential mechanical culprits, the wiring clonkig into the bed and the filament feed both put lots of stress on the drive system. You're looking for perfection so you have to address all these variables. If you don't you'll never get what you want. So it's up to you. There is no smoking gun.

446 (edited by redbarret 2014-12-10 18:04:20)

Re: Odd circle test results

My point is, if friction was my problem, I'd get step skipping at the belt tighteness I needed.

And BTW, I have some 12 LM8UU bearings and tested them on my rod, I know what you mean.

Solidoodle 4

447 (edited by jagowilson 2014-12-10 18:03:01)

Re: Odd circle test results

If you have the bearings why have you not put on the new carriages? They make a huge difference and all of the information on how to do it is here. If you are having a warping problem try that slurry stuff everybody likes. Haven't tried it myself.

448 (edited by redbarret 2014-12-10 18:11:07)

Re: Odd circle test results

Because of the wasted filament, the fact that I don't have two spare 8mm rods and because I just explained why it will be a waste of time and not fix this issue. Unless you can show something wrong in my explanation, there's no point for the reasons I mentioned.

And btw, again, I'm not "looking for perfection". just ask someone else with a different printer to print this test. And again, this is a test for a problem which shows up in real prints. Small gaps between infill and shells is one example. I had such issues and that's why I started this backlash reduction upgrades. I'm printing a circle only because it uses so little time and filament and is so easy to spot backlash on, not because I want my printer to print "perfect circles".

Solidoodle 4

449

Re: Odd circle test results

what about the carriage having rods loose in it letting it flex?

Solidoodle 2 with Deluxe kit cover & glass bed with heater. and 2nd board SD2 used not 3rd and alum platform not installed yet still wood. also need cooling fan installed to board. use Repetier Host couple vers. Slic3r also have all free ware STL programs

450 (edited by redbarret 2014-12-10 18:21:53)

Re: Odd circle test results

That's a good reason. But doesn't the play go away after you tighten the belts moving the carriage for you as well?

Solidoodle 4