How to square axis in assembly - solved

[jhinkle 1]

I feel really stupid because the directions makes no sense to me and there is nothing posted in the forums.

So it looks like I'm the only one not understanding.

The assembly instructions say to use wood part 3D as a gauge between frame and the sliding block.

If I measure Left and Right sliding block to frame at the BACK - there is an difference of about 1/16 inch.

If I set the distance Back Left and then force Right back to be the same size (makes the extruder left/right axis parallel to back frame.

Set both set screws (left/right on back axis) -- move extruder to the FRONT and measure gaps --- expecting to find no difference --- now I have about 1/8 difference.

So --- I'm have a brain fart in that I just don't understand the directions and the theory being used to square the axis.

Any help would appreciated.

Thanks in advance.

Joe

 

[IRobertI 428]

I thought the new kits came with special tools for this. The use of the "3B" part is from way back (and sucks). Look around in your pile of pieces and see if you find two long pieces with a series of half circles cut out along one side.

What you do is loosen all the pullies.

- Then position the head in the center of the machine

- Move the head half way or so towards the left

- Now take those long pieces and put them between the rod in the side of the case, and the rod that goes through the print head. One tool closest to you and one on the opposite side.

- Now push the head towards the left again so that the rods are pushed firmly into the two tools.

- Provided you used the same notches on both tools the rods should now be parallel and you can go ahead and tighten the corresponding pullies.

- Rinse repeat for the other axis.

 

[IRobertI 428]

Couldn't find a picture of the wooden tool but here's a printed version:

https://www.youmagine.com/designs/adjustable-axis-alignment-tool

And here you can see another variant being used:

http://www.thingiverse.com/thing:38300

 

[jhinkle 1]

Thank you.

I found the tool and its a lot better than 3B.

My issue is I set the screw on the Back left pulley. Move the extruder to expose the set screw on the Back right pulley.

Position left tool to set left position --- right side is now off.

All pully screws except the Back left are loose BUT the system is VERY tight. It is very difficult to move the right sliding block to align it with the tool.

Can you comment/ explain what might be done to loosen things up to be able to permit alignment.

Thanks.

Joe

 

[IRobertI 428]

Are you sure it's not simply because things are out of whack and causing it to bind up? You might have to nudge the blocks on the other axis a bit to free up the axis you're trying to move.

Also check that the belts aren't running really crooked. You want the belts to run directly above and parallel to the rods.

Also, you'll want to have both tools on there at the same time to make sure both ends are in the correct position before tightening the set screws. If the screws aren't lining up (check if you can reach the screws from weird angles as well ) move in or out a notch on the tool and there's a good chance you'll find a spot where you're able to get at both screws at the same time. "Worst" case simply loosen the belts so that you can rotate the pullies so that they match.

 

[jhinkle 1]

Movement of the extruder is tight.

I associated that with the resistance associated with the stepper motors.

I can (with some effort) move the extruder to all corners of the printer.

I believe all is as you suggested.

The first pic two pics show the difference in alignment.

Third is shot from front to back, forth is back to front.

The right slider block does not want to move forward or backward. Moving the front and back moving blocks do not helps as they appear to be locked into position and not able to move relative to one another also.

Comments/help -- all welcome.

First

Thanks.

Joe

 

[IRobertI 428]

Looking at that last picture it's clear that the rods are pretty badly misaligned, you need to fix that. Personally I'd just start over by loosening all of the pullies again and re-doing it. With that amount of misalignment the head will indeed be very hard to move around.

 

[jhinkle 1]

I thought I had the solution by going back and releasing all of locks on the belts --- but that did not work.

This whole axis alignment is under the premise that the two rods coming out of the extruder are at 90 degrees.

I saw no adjustments to make sure the rods are exactly at 90. If they are off --- I would be experiencing the issue I have.

I've checked the extruder box and all appears well there.

Any ideas on how to move one of the rods in the extruder by a sub-angle amount?

Thanks.

Joe

 

[owen 17]

Any ideas on how to move one of the rods in the extruder by a sub-angle amount?

Thanks.

Joe

You can't do that Joe and don't need to as that is set at 90 degrees.

When your axis aren't square it means that the extruder is twisted so that both axis are out.

When you can't get to your screw on the other side after doing one screw you remove your alignment tools and move the axis to be able to get to the screw. Then put on your alignment tools again. The axis will now be positioned into a different half hole on the alignment tools (but should be the same on each one)

If you loosen everything first you can make it so your screws on both sides are adjustable at the same spot before you start tightening them.

 

[jhinkle 1]

Not quite following you.

The premise of this whole squaring process is that the drive rods and slide rods in the frame are 90 degrees to each other.

Using the tool -- you are attempting to set the extruder rods parallel to their associated drive rods.

If they can not be set parallel the something is off.

The only flexible attachment points are where the small extruder rods enter and are secured to the sliding block. Those are almost a press fit which then locks the extruder rods to the "slide" rods which are expected to be 90 degrees and parallel to the drive rods since they are all positioned in the frame with no movement allowed.

So ... if I have not misunderstood the complete process --- the ONLY place that could be OFF and would need some sort of adjustment would be the angle between the rods in the extruder.

The extruder angle is fixed ... so how can you twist one axis as the assembly instructions suggest to make everything square.

What is the squareing process really trying to accomplish as I see it accomplishing little but to add a binding force.

Please tell me where I wrong so I can get this unit squared

Thanks for any comments -- even ones that tell me I have my head up my ass.

Thanks.

Joe

 

[IRobertI 428]

You have your head up your ass!

C'mon, I had to

The rods aren't held as rigidly in the slide blocks as you think, they can flex a bit in there. Have you loosened ALL the pullies like I said in the first post? Once you have done that the rods can move around quite a bit.

And just for future reference, the "extruder" is the part that hangs off the back of the machine with the big gear. The thing that moves around at the top of the printer is the "print head". Just to avoid confusion in the future.

 

[jhinkle 1]

Robert:

Yes --- I did you what you suggested. -- Still issue.

I have now taken the print head (sorry for the wrong reference- I knew I had my head where things smelled funny) -- completely off and was going to go through the step-up/assembly again.

Problem is ... if I did make a mistake ... and don't understand its a mistake -- my reassembly will probably end up with the same issue ..... that is why I'm trying to get a clear understanding of the process.

Joe

 

[jhinkle 1]

Robert:

FYI ... as to why I was calling it an extruder ... taken from the assembly instructions. :-P :-P :-P :-P

 

[owen 17]

Joe:

The squaring process involves sliding and tightening the slider blocks along the outside shafts so that they are the same distance from the other outside shaft which will make the head axis parallel to the outer axis.

That is, adjusting both X slider blocks to be same distance away from the Y outside shaft.

This is achieved by by loosening the screws on the pulleys.

You should start by having all 8 pulleys loose.

Then do X axis, align with tools, tighten up the 2 pulley screws at one end then the 2 pulley screws at the other end.

Then repeat for the Y axis.

The only way they can not be square then is if your your frame/outer axis aren't square(probably extremely rare for a UM) or the inside axis were bent (not very likely)

 

[jhinkle 1]

I totally understand and agree with what you say.

Explain if you can, ....

All frame located rods are square.

Assume rods in print head are 90 degrees.

If all of that is true --- then securing the print rods to the sliding blocks should also produce square.

If its not ... then the only logical piece that could be off would be the print head.

Assume for a minute that they were 89/91 degrees to each other.

You will set the first print head rod parallel to its related frame rods.

The other print head rod would not be 90 degrees and would produce a difference when using the alignment tool.

What I see the squaring process saying to do is to "force" the distance to be correct --- thereby squaring it --- but in doing so -- you are actual applying a binding force to the bushing in the sliding block.

That is what I'm trying to understand ... how to FORCE square without adding the additional binding force.

Joe

 

[jhinkle 1]

I just confirmed that the rods in the print head are not 90 degrees to each other.

I drew a line on a piece of paper and then, using drafting tools, -- drew a line perpendicular to it.

I placed the rods in the print head and placed it on the drawing.

Getting my eye over the middle of the print head, I aligned one rod with the horizontal line in my drawing.

I then expected the other print head rod to align itself with the perpendicular line --- but is is off by 1 to 2 degrees.

So .... THAT is my issue ... Now ... how to adjust the rods in the print head to get them close to 90.

Comments?

Joe

 

[owen 17]

So .... THAT is my issue ... Now ... how to adjust the rods in the print head to get them close to 90.

Comments?

Joe

 

There is no adjustment their Joe. You could only put 2 shims diagonally across one of the linear bearings. Or you could loosen the head screws off a little bit. Maybe if you loosened and re-tightened the screws in the head in a certain order you may get it more square.

It's more important to have the axis square so maybe you could loosen off the screws holding the head a small amount so you can twist the shafts to square and align the shafts and then see if you can tighten up the head screws without adding too much friction to the head movement.

Slowly tightening the 4 screws a bit at a time and then going to the next screw rather tightening each one right up at a time should help.

 

[jhinkle 1]

I believe the current UM process of squaring the axises of the printer is severely flawed.

If every thing was built to precisely 90 degree angle or rods parallel to one another -- then there would not be a squaring requirement.

The proper way is to have alignment adjustments -- but that would add cost and complexity.

Forcing a sliding block so that it becomes square and holding that square by tightening the pulley set screw--- DOES produce a square system but adds a twisting force to the bushing associated with the sliding block. The induced side forces produce friction which requires more torque from the stepper motor to move the print head and a hotter running stepper motor.

In my case -- the print head assembly has a 1 to 2 degree error in the two rods being perpendicular. The block is designed with no adjustments so I've got some pondering to do to solve my issue.

This all assumes that the rods attached to the frames ARE 90 degrees and parallel. If not --- there are no adjustments there also --- just a bigger compounded issue.

I could fix all of my issues by manufacturing parts that would implement adjustments ... but I need my printer to do that,

Talk about a catch-22.

joe

 

[illuminarti 16]

Loosen the screws holding the print head together so it can flex a bit. Then adjust the sliding blocks so that the opposing pairs are exactly opposite one another - that's basically what your trying to do when squaring things - just getting the two blocks opposite one another.

Check also where the cross rods go into the sliding blocks. There can be quite a bit of wiggle room there - if one isn't fully inserted, it can push the opposite sliding block so it's no longer parallel to z, twisting the belts, and driving everything askew.

[jhinkle 1]

It squared nicely once I got the rods in the print head to be 90 degrees.

I loosened the print head's clamping screws and the adjoining bolts. The print head was still a very solid and no flexing anywhere was perceived.

Using my drawing of perpendicular lines and the rods as lever arms, I attempted to get to 90 degrees.

I felt nothing move at all. but after a couple of times on the rods applying force in the correct direction, the rods measured 90 degrees. I felt no movement -- but it moved.

With the print head rods now at 90 degrees I inserted the print head into the printer.

Set belt tension and placed the alignment squaring tools.

Perfect square!!!

It only took 8 hours of pain and anguish figuring out root-cause and addressing it.

Now that everything is square WITHOUT having side forces applied to the bushing in the sliding blocks -- the head moves smoothly.

I hope this helps anyone else with a squaring issue. Be careful NOT to just force square by forcing and setting the pulley (as UM suggests) or you will not run smooth and your stepper motor will run hotter than required.

Joe

 

[gr5 897]

With the print head rods now at 90 degrees

 

Yay!!

I just read all these notes at once progressing through your 8 hours in my 2 minutes. At each step I thought the same thoughts as other people and then you replied with why their advice was wrong and I saw that you were right but eventually you got to the root problem (the print head). I think that most people were just lucky and their print head was square when assembled. It's been so long since I assembled mine I really don't remember having squaring issues.

 

[aviphysics 9]

I wonder if the assembly guide shouldn't be adjusted to put tightening the printhead after squaring the rods.

 

[IRobertI 428]

I've (re)-assembled my print head probably 6-7 times and with two different sets of laser cut parts over the years without issues so it never even entered my mind that the head would be off-square. Sorry about that.

 

[jhinkle 1]

I appreciated everyone's help and input.

The hardest part in solving an issue is finding what the root-cause is. Once that is known - the rest is usually easy.

I would have thought UM would have updated their assembly instructions on line about the new tool they provide.

If not for this forum, I'd still be using 3B or the several other tools I was coming up with to do the job.

Again --- thanks to all in helping resolve my issue.

Joe

 

[aviphysics 9]

It is kind of passing the buck, but it is a wiki, so anyone could choose to fix it.

UM has said that they plan to update it, but who knows when that will happen.