Lift Between Towers

[dcfc_sandy]

Hello all,

I've just come back from a term in Italy and after settling in to my new place I thought it was high time to get the 3D printing going again and get Rusty all set up. He seemed to fair well in the 3 hour car transit so I'm pretty happy.

I had an issue before I left for Italy regarding lift between the Z-X and Z-Y towers, maybe 2mm each way, which to me points to an issue regarding the Z tower. I remember reading a thread months ago regarding a similar issue and they were advised to straighten the affected tower.

I'm going to get a set square from B&Q tomorrow to confirm the tower is not straight, but this is currently my best guess.

So I would like to know how one goes about straightening the towers and realigning. I would have thought from the T-nuts the tower would have to be straight as the towers is aligned by them. Any advice would be muchly appreciated, so hopefully soon I can venture out from my safe bubble that is dead centre of my print bed

Grazie!

[Xenocrates]

The T-nuts provided help with alignment, but do not totally fill the groove in the extrusion, meaning there is slop for alignment. (I remember this well because it took me 15 minutes to line one back up with it's hole while I was installing the brackets for my chamber) I would loosen but not remove the middle T-nuts to start with, and slowly push it into position, If you have a second person available, have them torque down the bolts when it's aligned, while you hold it in position.

[dcfc_sandy]

Righty, I'll give it a wack in the morning.

The square edge confirmed the theory.

Thank you

[dcfc_sandy]

It's was a job as I live alone and have no friends .

But it worked, the tower was straightened print calibrated and the results are incredible! I had maybe a 20cm diamter worth of print radius before now I have the whole bed!

I tip my hat to kind sir

[geneb]

\o/

[dcfc_sandy]

Okay so after going through the 24v upgrade the issue has resurfaced, although thinking back I don't believe the issue 100% had gone. In the above photo I'm printing a gear cube for a house mate and hadn't done the bed levelling print after the tower square just some simple G code and it looked better so away I went. This print isn't very taxing on the Y axis.

Since the 24v upgrade I tried to get the printer as physically sound as possible square the towers screw by screw as I went round and levelling on the onyx to within +/-0.02mm with a calliper and spirit level. Visually when dismantled the snowflake and onyx looked fine. I recalibrated using endstops Z height and horizontal radius, started to print bed levelling aid 200mm on my nice now possible to hit 105C bed and around Y = 70 I started to lose adhesion, it returned around 2cm to the right of the Z tower and lost again 2cm after and returned around Y = 70.

I've since rechecked tower squaring, bed levelling, belt tension, cheapskate tension (if that's the correct way to describe it?), the hobbed gear on each motor an individually watched each motor to ensure no skipping is taking place. The more to the extreme X is worse than the centre and around Y 70 - 80 is the worst. Visually (and with callipers) as I don't have feeler gauges the print head is around 0.5-0.75mm off the bed at Y = 70 X=44.

I am really at a loss to what else I can try, all this information to me points the the Z tower but I've no idea what else it could be?

I had a tantrum over it and have bought and FSR kit (in the hope that OpenDACT will solve my issues) , even if it did I would like to know what the issue is.

Any smart ideas?

[dcfc_sandy]

Now I've calmed down a bit and I feel Rusty isn't in danger of being thrown out of a window... Like it would fit! I decided to tackle this as a scientist should and logically change a variable at a time.

I'll be swapping out parts with another tower, more in the hope that the problem follows. I'll do one at a time with easiest to change parts first, probably: arms, belt, cheapskate, hobbed gear (I didn't threadlock it... Please don't kill me Gene), motor, rotate the bed and snowflake 120, maybe the effector (but at this point I'm just avoding the inevitable), long and agonising cry about why life hates me and then the tower.

Do we think I've missed anything glaringly obvious?

[dcfc_sandy]

In case I wasn't clear. After I've swapped a part I'll 3 point calibrate and run the bed level print.

I though about endstops, but around the tower is fine so I ruled those out. And of course endstop screws will move with the cheapskate

[dcfc_sandy]

Because I am lazy I didn't do what i said i would... I tried the arms and calibration was pretty much the same, calibrating after each one sounded like a ball ache so I cheated and swapped the X and Z stepper, hobbed gear and belt.

I am please to report that the issue now manifests itself around the X tower instead of the Z.

After showing face at uni tomorrow (before they stop paying me, I've not been back since Easter! ) I'll come home and swap the belt of the X with the Y axis calibrate, check if problem manifests in Y, then repeat until I've nailed this pesky problem.

It seems and odd problem to be cause by what is probably missed steps somewhere down the line as it only manifests around the problem tower... but hay ho!

Hopefully good news will be present tomorrow! That's all folks!

[dcfc_sandy]

After swapping the belt of the X tower with that of the Y, the issue has instead returned to the Z tower...? This to me must show that something is seriously mechanically wrong...

[dcfc_sandy]

I may have found the culprit... The U joint axle on each of my cheapskates has only one screw holding it down...Pg 120 of the Manual Fig. 8-13.

I have checked my spares box and there are no more of those screws, and unsurprisingly nowhere locally... or in the country as far as I can see sells them (stop using imperial screws!). I may just bite the bullet and make that Trick Laser order I've been umming about.

To any gurus that may stumble across this (or people who have had similar issues), is the play from the U-joint axle likely the reason for this lifting I've been getting around the Z tower? There is play on each of the Axles and probably more so on the Z then the others. If you believe it is, why is this only happening at the extremities of the bed (because in the middle is close to a calibration point maybe?) finally and most curiously to me, when I swapped the Arms, belt and motor of the Z and X tower why did the problem follow before returning to the Z once the X and Y belts had been swapped...?

A lot of strange goings on

[dcfc_sandy]

Right I've managed to source some #4 screws... I tell you they appear to be gold dust over here. Hopefully delivered before the weekend. Also investing in a digital angle finder just to be 100% on tower squareness, but I don't believe that's the biggest issue.

FSR kit should be here tomorrow, then I need to print the parts and get the fixings; will hold off on the install until I've nailed down this issue now as I think I'm zeroing in on it!

I think this thread has turned more into a blog (or a diary to remind me of what I've actually tried...)! More updates as they happen

[dcfc_sandy]

Thought I'd try the Matter Control 13 point bed leveling just to quadruple confirm its not the bed being off and its not... However the problem area moved from being either side of the Z tower between X and Y to being directly on the Z tower...

Someone out there must know what this means... I certainly don't :'(

[dcfc_sandy]

Latest update...

Swapped and flipped the belt of the Z tower (originally the X) and the X tower (originally the Y), as well as swapping the tower motor inputs on the RAMBO to see if driver issue. So X,Y,Z have been plugged into Z,Y,X as the endstops to account for the new position.

This appears to have solved the issue, the nozzle doesn't appear to noticeably lift at any point. I'm currently contemplating whether to put the towers back in the correct position or just leave the printer 120 degrees out of whack (basically I am scared the printer will return to its old ways). However, as the issue hasn't followed the motor to the new tower I can rule out a motor driver issue and probably put this down to either the belt of the missing screw for the axle bar.

Although the screws haven't been installed on the u joint axle bars as they haven't arrive, I have screwed in each side of the joint using the one screw i have for each and they appear to be better seated. While on the subject, when I changed the motors input ports, I forgot to account for that in the firmware (as the Y axis motor is reversed) so my X and Y towers went down while homing, the u joint axle bar on the Y tower cheapskate popped off.

To avoid confusion when I say X/Y/Z tower, I am referring to them as they were in the original build not to where I have moved them! I was confusing myself for a minute.

I'll muster up to courage to pop the motors back in there respective places soon and report back.