Can't get Center Assembly Square XY direction

I have searched for other posts regarding this, but been unable to find any…

My printer is very square. I print a lot, and stuff match together (small gears etc)
I have tried measuring my parts and even turned them on the buildplate and reprinted… all without any changes…

I AM using the correct files for the correct conduits, which are the correct size as well. Same for bearings etc.

The photo I attached is exaggerated to show what I mean.

I have tried adjusting using the relevant screws/nuts, but there is no give in them to adjust anything… I drilled some of the holes larger, and I could adjust it, but then I adjusted it away from all bearings touching the conduit.

It seams as if F-XYZ should be opened up a tad more, but there isn’t any way to adjust that way…

… any ideas?

Perhaps your printer is printing a little small, what slicer are you using? maybe you’re overextruding… are you using pla? what size conduit? were you able to insert the bolts without having to drill them out?

I believe it’s simply down to the F525 pieces beeing too snug regarding the 8mm holes
I’m using the F525 so conduits are 25mm OD :slight_smile:

Measurements and overextruding
I just measured holes in STL file and all ø8 holes are 8.0~8.2 which is the root cause really. They should have been 8.2~8.4.
Since they are meant to give room for adjustment they really should be even larger. At least the ones meant to give some adjustment.

My measurements are spot on as far as my Multimeter can measure +/- 0.01 … hard to be exact though, as the tiniest pressure or wrong angle on my part is going to mess up the measurements.
Don’t think i’m overextruding… not to any extent where it should matter in any case.

Material
I’ve used PLA for the center assembly, but also used ABS for other pieces… I have had to drill up ALL holes for ALL ø8 bolts. I have not had to drill up holes for m3 for motors and bolts to fasten top to lower roller.
Other than drilling the ø8 holes everything fits very nicely together.

While measuring the F525 STL files I noticed the correct height to ensure good ø8 bolt holes is 0.2 or 0.25 and not 0.26… might mean something or not… I still think it’s the SIZE of the holes in the STL files that causes it.

Edit: I should say that the entire assembly runs very smoothily and all bearings touch the rods :slight_smile:

I have tried printing the F-XYZ in different PLA but same result.

The “adjustment” bolts don’t slide around inside the hole for adjustment, they actually flex the parts and squeeze the bearings against the pipe, if need be. I think you’ll be okay after a few runs, once everything has worn in. just remember not to tighten down any of the bolts yet. your bolts should be tight enough to snug up the bearing but still be loose enough that they turn freely… if that makes any sense.

I agree with Walter! They adjust slightly by squeezing the part. Leave everything loose and use it a bit. Tighten it up only if necessary.

If it is still not square it should only be by a tiny bit. Measure from the roller to the corner block on all 4 sides, this little adjustment happens before all of my cuts. Because of the geometry of this machine you can make it not square by moving the head around. You will always have to take the 30 seconds to verify matching axis roller distances or you will cut parallelograms and ovals.

Walter

The “adjustment” bolts don’t slide around inside the hole for adjustment, they actually flex the parts and squeeze the bearings against the pipe, if need be. I think you’ll be okay after a few runs, once everything has worn in. just remember not to tighten down any of the bolts yet. your bolts should be tight enough to snug up the bearing but still be loose enough that they turn freely… if that makes any sense.

Tight, snug loose:
I’ve tried different ways… both very loose and very tight… my bearings always turned fine regardless of how much I’ve tightened it.
In no way can the bearings be move away from touching the Conduits - After I enlarged some holes a lot I could do this… I know some of them are not straight, which makes the issue worse, so had to enlarge them even more to make some wiggling room.

I hear what you say. And it might work. I’m just not sure what you say regarding adjustment is spot on… Or at least the instruction is VERY unclear to me… Viscious1 is in this thread, which is nice, but we seem to have different views on what is clear or not; I personally find the entire instruction very unclear with no clear correspondence between text and images :
I lack some visual references on the photos, as to the bolts we talk about, and better description of function of said bolts.

I started out with all bolts (A, B, C) very loose, but it really made almost no difference between being loose and tightened up. All my parts fits snugly together without bolts… Not so tight as they stay put without bolts, but enough to not leave any wiggling room.

In the assembly instructions it says:

First check that the X and Y rails are equally tensioned in the XY blocks, all the bearings should make contact. If you need to add any tension start with “Tension Bolt A”, at this point it is best to leave “Tension bolt B” loose (it locks the assembly in place as well as rail tension).

Second step is to make sure the Z rails have equal tension and all the bearing are making contact. If you need to add any tension use the “Tension bolts C”

How I understand it, and please correct me If I misunderstood it, the bolts do 3 things:
  1. Tighten the parts together (obvisouly)
  2. Adjust bearings if they do not touch the rails - So there must be some movements in the mountholes? - Mine has none, and in fact had to drill up the bolt holes to even get bolts through it. (holes are 8~8.2mm when measured in STL file, which means they has to be drilled in most cases)
  3. Make adjustments by tightening the parts in special sequence - what sequence?
    1. Worn in after a few runs
      How do you mean it should be worn in? Problem really is that the top and bottom F-XYZ parts are pushing the F-ZY parts a bit too far apart (outer edges), meaning the XY cross-conduits are not square.

      When I turn on my controller it locks the drivers, which means I could turn it off, make the Assembly square and then turn controller on… it would be square now… I could run it like this, but I do not see how it should wear F-XYZ in any way to fix the problem? (problem: open up F-XYZ)

      Pre-checking before each run:
      Yea, I can see how it is needed, but as it is now there is a 1-2mm gap from inner to outer end of my rather small Square… that is a lot in the big picture.
      My primary 3D printers are UM2 clones, so I do know the how the mechanics works.

      Squaring tool:+
      What is the distance between center to center of XY rods (the 4 outer rods)
      Asking as I’ve wanted to make some squaring tools as I use for my Ultimaker machines, but correct distance would be nice :slight_smile:

To answer your last three questions.

1-Most of the bolts just hold bearings in place, only a few actually hold things together.
2-If during assembly all your bearings do not make contact with the rails something is wrong. The only adjustments that gt made are tension, very small adjustments. The bolt holes are meant to be very tight to the bolt, if you drill these out you have ruined the piece and it will most likely be crooked.
3-Not really.

You are looking for giant adjustments like eccentric bolts or something. This is not how this works. If you watch what happens for every piece when you insert a rail it might be more clear. Every single piece should flex open to accommodate the rail. This is why the middle bolts on every single rial part should be extremely loose when assembling. You can even go as far as not adding a nut to them.

The adjustments happen only on the bolt with plastic on both sides, I call these the middle bolt/bearing.

When you tighten the middle bolts it is bringing in not moving the middle bearing, it is bringing the other two bearings slightly closer.

I am sorry you are having difficulties but as I have said in the emails there is no secret tightening sequence to get this right. Most people’s assemblies come out really square, If you are having issues the only adjustments possible are the 8 middle bolts on the middle assembly. The down side is most change 2 axis at the same time. The only thing I can suggest is loosen or tighten a bolt and check how it effects your assembly. Put the square on your rails and watch it move. When I need to do any adjustments I spend ten minutes doing just this. Obviously it is much easier not on the rollers.

You ask for a dimension at the end of you post, I am unclear which one you want. I think you are looking for 27.5mm. Squaring tools are not a good solution, you should check your roller to corner distance before you start the machine, takes 1 minute.

Hi,

Just confirming that when I put this together it was just like Ryan said… assemble, and put in the rails, and it’s square.
I have the 23.5mm kit for 3/4 EMT conduit. In building the center piece, the bolts fit in smoothly and the bearings generally just touched the conduit when I slid it in. At that point, I checked it with a 30cm square that I have and found that it’s perfect within my ability to measure (sub 1/2mm over 30cm). I placed the square on one of the pipes and then looked at where it sat on the other pipe to see that it was indeed square. I think this is what you are doing, but it’s hard to tell in the picture. I might set it up so you can see the square touch one pipe and not the other all in the same picture. like this:https://www.v1engineering.com/wp-content/uploads/2015/07/IMG_20160527_13131101.jpg

For the first few runs on the machine, it was a bit rough because there was the coating on the conduit. After a few runs though, the coating started to flake off and wear down and the movement settled right in and became very smooth.

It MIGHT be worth taking the whole center piece apart again and then reassembling it methodically just to see if you can figure out what’s wrong with it. In my case, everything fit together very nicely and with appropriate tightness without changing any of the holes.

My machine is 2’x2’ and it really doesn’t need the steps that Ryan takes to square his up each time. it’s so close when I’ve checked, that I don’t do anything to it. and again, it just came together like this, no real work at all to square it.

Good luck… it’s got to be very frustrating.

Viscious
Might be the drilling, as I HAD to drill up the metric holes some… Have you even tried these yourself?

All bearings touched all rails though and everything ran smoothly - after drilling.

I figure it might be due to layerheight of 0.26 and holes of 8.2~8.4… 8mm bolts just do not fit.

When I draw my own pieces 0.2 is snug and 8.4 is loose.

Anyway… I’ll have to tinker, as we have reached a point where it’s just mud throwing which neither of us benefits from :slight_smile:

John
You did not confirm my configuration… you used a different set of pieces and bolts…
It doesn’t matter when you test on other pieces… problem with METRIC parts is how the holes are 8.0~8.2 and layerheigh is 0.26… It just doesn’t work

My problem propably is my printer is more accurate than most really (from experience)… when I print 0.2 my stuff is snug… 0.4 it is loose… when I print 0.2 and wrong layerheight I can’t use it…

It might be as simple as the slicer you use. Slic3r makes smaller holes for me than simplify3D does. The parts i sell have been sliced with s3d, I don’t remember how they turn out with cura/matterslice.

First I would like to thank Vicious for an excellent build, and for helping out in this forum. My comments below are as much for getting help, as for helping improve the build. And if you need help from any of us with metric builds, let us know.

I have the same issues as described above with the 25mm build.
I currently have two suggestions I might try for aligning the center assembly.

One is making the holes in the image a little bigger. They are the only two holes that require force on order to make a fit. And forcing them into place, twists the assembly in the opposite direction of what I need. (01.jpg)

Second is filing down the part in the marked area, to allow the big X and Y parts to close in on each other, correcting the misalignment. (02.jpg)

Thanks,

(I also needed to drill out the 8mm holes)

the holes you reference in photo 2 can be misaligned if your print had a problem with lifting.

Thanks for the reply,

There were no lifting, and they are perfectly aligned.
I just need to squeeze the assembly, and that might be possible if I remove a bit of material.

Cheers,

Can you verify your z travel is correct with a 100mm move. Before I edit the model.

I had this problem on one of my printers, the z steps were set to 800, and actually needed 788. It was only off by like 2mm or something like that but screwed everything up.

Kronbjorn: First I would like to thank Vicious for an excellent build, and for helping out in this forum. My comments below are as much for getting help, as for helping improve the build. And if you need help from any of us with metric builds, let us know.

I have the same issues as described above with the 25mm build.
I currently have two suggestions I might try for aligning the center assembly.

One is making the holes in the image a little bigger. They are the only two holes that require force on order to make a fit. And forcing them into place, twists the assembly in the opposite direction of what I need. (01.jpg)

Second is filing down the part in the marked area, to allow the big X and Y parts to close in on each other, correcting the misalignment. (02.jpg)

Thanks,

(I also needed to drill out the 8mm holes)


I concur with the analyses, and I was about to do the filing myself…

But then I printed with 0.25mm layerheigh instead of 0.26 and everything fit much better.

Thank you for the replies,

I have attached an image showing the height of the tall Z part.

Does that answer your question ?

Regards,

(I’m almost sure I sliced with 0.26, but i may have used 0.25 as I changed it at some point during the build)

What he meant was to use the printer you’re printing these parts with and do a 100mm z axis move and measure whether or not it’s actually moving 100mm. now I’ll have to check when I get home, but I think you’re spot on if that is the tool holder mount you were showing off in that chopped up picture

The F-toolmount should be 118mm. They usually print a touch under from a squished first layer. You printer is pretty good but I think your z axis is moving too far. You should verify by doing the 100mm test.

Thanks for the reply, that would explain a lot.
(I will see if I can remember how to measure my Z-travel)

Just a thought though. In simplify3D the height is 118mm, but in Repetier (Mac) it says 11.89. Could it be that converting from inches to mm, there are some rounding issues ?

Regards,