Z axis on wrong side  

Update: Z axis on wrong side

I discussed this with Michael in a support chat, asking whether I should swap the slider to the correct side by taking both off their rails or whether I'd have to disassemble the printer to the point where the left and right side z-axis assembly could be installed on the proper side. Michael strongly encouraged me to disassemble and start the build over. Thanks for the input, but I ignored that advice.

What I did instead was to take out the green stops at the top of each z-axis rail. Then I carefully turned the 80% assembled unit on its side so that the left rail was horizontal. Then I slid the slider off the end of the rail, being careful that my cupped hand caught all the roller bearings that fell free once the slider was free of the rail. BTW, there are 15 1mm balls each side, for a total of 30 per slider (X2 sliders). Then I slipped the unit over on its opposite side and slid the right-hand slider off its rail, again being careful to catch all the balls.

When I put all the balls of the rotating slider back in place and carefully slid it onto the still horizontal rail of the z-axis rail installed on the right side. As I put the slider on, about 15 of the 30 roller balls fell down into my cupped hand. I put them in a container, rotated the unit into an upright orientation, and spent the next 40 minutes or so carefully "spoon-feeding" the balls into their proper place. This was done by picking up each ball using a slightly magnetic jeweler screwdriver. Then I would slowly, and carefully slide the slider up about 4 mm past the top of the rail. This would, in theory, give me just enough room to push the ball into the roller-bearing channel. If I'm making it sound like this was easily accomplished, it was not. Sometimes the magnetism would cause one of the balls already in the channel to come out, drawn as it was to the screwdriver tip. But after a time, I was able to get the bearing channels refilled with all 15 balls on each side of the two sliders - 60 balls in total.

As difficult as it was to fix my mistake this way, it did work out. I'm convinced that the time I spent doing this was about half the time it would have taken to disassemble the printer, swap the z-axis assembles to their proper sides, then reassemble to that point. I only felt comfortable solving the problem this way because I sat down to the task with the mindset of a surgeon. I was exceedingly careful, patient, and methodical. You’ll have to be all that as well if you try this. Believe me, I understand why I was discouraged from taking this course.

Yes, I’m an idiot for not picking up on which z-axis assembly went on which side. But again, I strongly encourage the Prusa team to use stickers or some other visible means to visually distinguish the left and right z-axis assemblies.

RE: Z axis on wrong side

Hi ,

My XL was dropping some balls too last week.

I had some bed leveling issues, and followed a mod in another tread about removing the upper Z-spindle bearing caps.

After a few weeks of printing I got bed leveling issues and started all over with the calibration proces from the printer menu.

In this proces the bed went all the way up without touching the nozzle and not stopped by de caps.

With a closer look I found ball bearings on top of the linear z rails on both sides, which came out of the bearings, luckily they were held to the top by the grease.

Disassembling the  whole machine was not an option, so I found out that it is possible to remove the linear rails by removing the 7 torx bolts.

They loosen up quite easy and the spindle gives enough room to let pass the torx screwdriver on an angle (don't force it tho).

The bolts from the rail are mounted within a strip in the extrusion with threaded holes, which needs alittle help with assembling because it is a little shorter than the extrusion.

It can be easily hold up with an awl or small screwdriver and then install one of the bolts

All the bearings where removed over a large container and I came to the conclusion the bearings have 2X33 balls each.

In this way it is would also be easy to swap the rails from left to right.

I still haven't found why the callibration test drove the bed to its end, but all is running smooth again.

Cheers, Roland

RE: Z axis on wrong side

Thank you so much for posting this! I stupidly removed both arms and lost all the bearings. 

One pro tip that we found out was using a silicone grease sort of as a glue to hold all the balls in place while loading it onto the rails. With out the grease several attempts failed and balls went everywhere.

RE: Z axis on wrong side

So - just checking: You were suggeting to remove the rail. But if you do - why would you have to remove then still the ball bearings?

(I mean - there is still the problem that the bed holder is in the spindle - so I suppose that would need to be moved to the top (to get it out of the spindle) - at which point removing the rail doesn't really help? What am I missing?

Many thanks in advance!

RE: Z axis on wrong side

Moooment. I have found this page here which shows how to remove the linear rail to swap it with a new one. 

It talks about a special tool which needs to be created if you remove the rail - BUT - it also shows how to remove the Z-Axis arms without removing the ball bearings. I wonder if even the z motor needs to be removed. I think that can be seen if the on rotating screw can be removed or not, but either way according to images it looks like that the ball bearing case is the same on the left and right side. So... remove left, remove right, attach old left right, rotate printer again and then attach the other again. That should do the trick. Will try that on Friday and update!

RE:

If your following the instructions it very clearly shows the difference between the left and right sides. 

On the 3rd picture on step 7 there is a giant red triangle with an all caps attention stating how to tell the difference between the left and right side. 

I'm sorry your made a mistake, but I don't see any fault of Prusa here.

RE: Z axis on wrong side

I am confused. Nobody was pointing at Prusa for a fault here?

I am talking here about how to correct the problem without taking the whole printer apart.

But as an improvement idea: There are apparently several people which did this wrong. So it appears to be a good idea to attach a small label to the two pieces and label the, Left / right. Seems simple and would have prevented this.

RE: Z axis on wrong side

Posted by: @unicorn3d

I am confused. Nobody was pointing at Prusa for a fault here?

I am talking here about how to correct the problem without taking the whole printer apart.

But as an improvement idea: There are apparently several people which did this wrong. So it appears to be a good idea to attach a small label to the two pieces and label the, Left / right. Seems simple and would have prevented this.

Unicorn, my comment was not aimed at your fix.  Did you look at the instruction step I mentioned in my comment? It very clearly, with large capitol letters and symbols states to make sure your assembling the correct side.  If one is to not notice that in the instructions what makes you think a sticker will work?

I just feel that when people make mistakes they are always looking for a reason as to why they made it.  This is natural of course, but all too often we look to blame others.  In this particular case the instructions are very clear, so I think the blame is not reading the instructions.  Saying Prusa should include a sticker seems an an awful lot like saying Prusa is at fault without actually saying it. 

Anyways that's my 2¢.

RE: Z axis on wrong side

I am only looking to fix the problem - I am not pointing fingers or blaming anyone (but maybe myself that I missed this thing as a newbe to 3D Printing at the time).

Also - I do not think that I was indicating anything else here? So as long as we are talking about a potential fix for the situation - I am all ears! 😀 

RE: Z axis on wrong side

Just for the case someone is running into the same problem and is interested in a fix:

Yes, it worked with a modification of the link.

1. Removing the exterior things as directed,
2. Flipping the printer, have the rotating arm on the bottom,
3. Removing the Z-Motor,
4. Removing the rotating arm,
5. Putting the motor back and fixing the downstairs
6. Removing the upper motor
7. Remove the arm
8. Screw the fixed arm in the bottom
9. Screw the rotating arm to the upper rail, slide
10. Put the upper motor back
11. Make sure all screws are as tight as before and flip the printer back.
12. Add all the exterior stuff again.
13. Perform the calibration to see that all is good and do a test print.

==> Seems to work - without disassembling the entire printer - which is clearly far more work.

It also feels safer than removing the ball bearings (but that is only me 😀 ).

I did do a test print and it could be that the print quality might have improved - but I am not sure yet. 

RE: Z axis on wrong side

The point you are missing "Mr. Factual Engineer" is that in the field, when your product isn't marked right - these mistakes will happen.  Neither arm is marked, the instructions "Point" to the 4 Screws, this is the only way to know.  Instead of "Defending" the simplistic fix of marking the parts properly (as most Engineers do) blaming the user is easier.  Mark the parts properly and this issue goes away - us field guys do things differently.

To the fine gentlemen up north I did the same process, I use a pair of thick tweezers with a light magnetic grasp, it is tedious but your "Surgeon" comment is spot on.  Take a break after each side, its way less time then de-constructing the unit. 

Take care.  

RE: Z axis on wrong side

Posted by: @maximum-overdrive

The point you are missing "Mr. Factual Engineer" is that in the field, when your product isn't marked right - these mistakes will happen.  Neither arm is marked, the instructions "Point" to the 4 Screws, this is the only way to know.  Instead of "Defending" the simplistic fix of marking the parts properly (as most Engineers do) blaming the user is easier.  Mark the parts properly and this issue goes away - us field guys do things differently.

To the fine gentlemen up north I did the same process, I use a pair of thick tweezers with a light magnetic grasp, it is tedious but your "Surgeon" comment is spot on.  Take a break after each side, its way less time then de-constructing the unit. 

Take care.  

I assume your referring to my comments.  I still stand by my original interpretation, if one cannot follow the very specific instructions, with pictures, then why do you think a sticker will work? 

The only good fix here would be to "idiot proof" it as we "Factual Engineers" say. This is achieved by making it only possible to mount it 1 way, by moving the location on 1 screw or something along those lines.  Aka making it impossible to mount in the wrong spot because it won't fit. Short of that someone will figure out how to do it wrong and blame the engineers. 

RE: Z axis on wrong side

Yeah, I’ve dealt with engineers like you all my life, once pressed you finally start giving solutions.  Making your product idiot proof, while calling us idiots is just fine.  Make it fit only 1 way is a simple fix, or putting stickers on it would be even easier- I know it’s hard but now your thinking.  Since you're not an applications engineer I would assume R&D, dissing your clients is natural, this is while you should never be customer facing - since your thoughts are of course superior.  Let me give you a sales and marketing tip my friend - the easier and more “idiot” proof the products are the better brand recognition you are going to receive for ease of use -- which ultimately leads to sales so you can get paid to build a better mouse trap….  Think like your customer so you can understand our idiotic minds - then with your superior mind you can engineer a “simple” fix to the product. 

Have a blessed day.

RE: Z axis on wrong side

Just a general reminder to keep things civil guys.

RE: Z axis on wrong side

Agreed - thanks 3Delight!

Even though I agree that a better indication of the part would have helped here a lot (maybe something to improve on), I also have to say that it was explained in the online manual. (Apparently I have overlooked that as I was new to the Prusa way of assembing things.)

If an improvement idea comes up, it can be suggested and maybe adapted, but there is no need to bash.

(Re engineering: In the end we have chosen the self assembly version and some engineering is required there?).

RE: Z axis on wrong side

I am reading quite a few posts and this one is the first to state the rotation arm goes on the LEFT. If you look at the manual I have used online, the manual and the photos show the arm that rotates on the right hand side of the unit. RH is my right hand if the front of the machine is toward me. I am in a 30 minute wait for chat, so I am trying to find an online resource. So far I am seeing spacers, loosening everything, and very much condescending comments. Is anyone out there able to show a picture of the unit assembled with the rotation arm on the LH side? For anyone that reads this, “Go back and read the instructions” is not helpful. Not all of us come here before we triple check.