Can I get an opinion on some perimeter attachment issues?

[seedjar]

I've been having some issues with infill not connecting to perimeters, particularly on the top layer of prints. I'm wondering if I should fiddle with my belt tension or if something else is indicated. Here's a picture of the 20x20x10 test print I did after putting on a new HE280:

I'm running an older Orion (from early 2014 - I think v2?) with a RAMBo v1.1b. Yesterday I put on the new carriages, ball joint arms, EZR Struder, and HE280 (pre-assembled) with only a few hiccups and everything seems to be mostly in order. I ran through all the steps in the HE280 install guide (and the how-tos for the other parts) up to the advanced calibration. It's 3AM and I need a break, LOL.

Before the upgrades, I was occasionally having this same perimeter issue, but it seems to be more pronounced now. Also, the sides of the print look somewhat uneven and more rounded/bowed out than usual.

Any ideas as to what could be causing this? I set the belts on the replacement carriages by hand, so while I did take care to pull out the slack I didn't like yank on them with pliers or otherwise set them especially tight. One carriage seemed to want to move more easily than the others and that bugged the crap out of me but I wasn't really able to get it to tighten up so I had to give it up after a few tries.

Before the upgrades, like I said, I did have some similar problems with gaps between my perimeters and infill, especially on the top layers. I also had a pernicious bed leveling issue that I never managed to resolve adjusting the endstop screws. Sometimes the first several layers of a print (up to 2 or 3 mm I'd say) will be smooshed outward a little, like I'm pushing too much filament maybe?

Any help is much appreciated - thanks.

~Joe

[Xenocrates]

I think it may be a combination of calibration issues, a new hotend (And geometry, including of the nozzle), and a not particularly well set Infill Overlap (May be named something different, but it should be there) in the slicer. Lucky for you, the last one is easily fixed, and that should help a lot with the results.

[seedjar]

Thanks for the reply. Sorry not to get back sooner, I've been getting over a bout of tonsillitis.

I played with the Infill Overlap and it did fill the gaps but did not solve the wobble in the printhead. While raising it filled the gaps, it also resulted in quite a bit of slop on the top layer, so I think I had the value more or less correct before. Here's a comparison - 0.25 was the original setting. Sorry about the white-foreground-on-white-background in that last shot.

While the higher settings did close the gap somewhat, both at 0.275 and 0.3 there was about 0.5mm of extra Z height bulging above the top layer perimeters. I haven't tuned this setting before so I don't know what I'm looking for, but my initial impression is that I'm causing more new error than I'm correcting.

In hindsight, this was a poor choice of test prints to demonstrate the problem I'm having. I've got a better example I'll post below.

~Joe

[seedjar]

Hopefully it's easier to see what's going on with this part. Here's the part I'm printing, designed in Fusion 360:

Here's how it comes out:

Here's what the cross-section should look like, based on the first layer in MatterControl:

Screen Shot 2017-10-29 at 18.27.29.png (8.44 KiB) Viewed 10017 times

And here's what the first few layers look like on the plate:

So, I had some minor dip towards the back tower before, skewed to the left, but it's wildly worse after the upgrade. The loops from priming would've come out more or less even in my previous configuration - I could print to about a 100mm diameter before it got this bad.

The real problem is the skewing. The test square from my previous posts had more than 1mm +/- error in the X and Y axes. And sharp angles come out with curvy irregularities. I'm pretty sure that's contributing to the beading and layer misalignments in these other pics:

Any suggestions would be much appreciated, and apologies again for not getting back to this sooner. Would it be good to start a new thread? How long is considered thread necromancy around here?

Thanks much,

~Joe

[Khayrisill]

Hi Joe,

You should look after a several things, I think it's mechanical issue. Start by checking if the three tower are perpendicular to the base, adjust the tension of the belts and on the same times inspect the carriage, did you have the new injection molded? if not I know there is an adjustment on the old one, try to have the same tension everywhere.
And especially try looking for a problem like a loose pulley, inspect the fixation of the steppers.

[seedjar]

Thanks. Yes, I have the new injection molded carriages. Can you suggest a guide on tensioning the belts? So far I haven't come across a procedure for tensioning or a method for gauging the correct tension.

Also, what am I looking for when it comes to loose pulleys/steppers? I'll check the towers with a square in the morning. I really hope that's not it. The platform was moving pretty well before I did the upgrades, and I don't think I pulled hard enough on the belts to have dislodged anything.

~Joe

[geneb]

I don't think the issue is in the towers unless they're actually loose.

To check for a loose pulley without having to actually get your hands on it, get a fine tip Sharpie and draw a line across the face of the pulley and the shaft. Start a print job and watch the line - if the line on the shaft face diverges at all from the line on the pulley face, you need to tighten the grub screws on that pulley.

Also check to make sure that the hot end assembly is firmly attached to the effector platform.

g.

[Khayrisill]

Yes true, I dont think the issue is in the tower too, but that's my routine checkup if I have a mechanical problem.
Good point about the hot end assembly specially after an upgrade.

[seedjar]

Thanks Gene. The hot end came pre-assembled and feels quire solid. When I poke at it, the only looseness I feel seems to be coming from the motion of the arms/platform. I grabbed the top of the hot end and tried to wiggle it - no motion apparent except in the arms. Then I took a metal probe and pushed gently on the heater block while still holding the top of the hot end. Also no movement.

I'm afraid I don't understand what you mean when you reference the pulley and shaft. By shaft you don't mean the tower, do you? Do I need to remove the translucent plastic covers? At the top or the bottom? Are the grub screws the exposed screws holding the shaft in place? (I'm not sure which shaft you're referring to but I can imagine.)

~Joe

[Khayrisill]

Yes Joe, behind the Acrylic cover, at the bottom of the 3 Tower,

And looking for this assembly, the pulley on the steppers at the three tower (bottom), The shaft is the "rod" that comes from the stepper, inspect the area.

$_35.JPG (12.41 KiB) Viewed 9951 times

[seedjar]

Ah gotcha. I'll open it up in a bit and have a look.

In other news, the towers aren't square. The two in the front skew pretty significantly. I'm not sure how I didn't notice sooner. Is there a way to adjust that in place or am I going to have to tear it down? o.o

Thanks again,

~Joe

[Khayrisill]

The towers is supposed to be square IF everything is seat correctly. You need to check first on the top and bottom, on the RostockMax there is a screw head to position the aluminum extrusion there should be something similar on your Orion.

[seedjar]

The towers do appear to be fully seated at the bottom. It seems like this version of the Orion does not have screws on the top, though.

As for the pulleys, they seem to be tight. I ran some test prints last night with kind of sloppy marks, and didn't see any drift but wasn't entirely sure. Today I found a pen with a finer point and checked again while moving the print head with the console. Again I didn't see any wiggle.

Went to start another test print and discovered that my extruder now won't heat up. I guess I need to pull out the board and check the connections. It was working fine after the upgrade, but to get at the pulleys I've had to move the wire bundles quite a bit so maybe I pulled something loose. The whip for my HE280 has a lot of slack on the end and I wasn't able to tuck it away above the power supply as suggested in the guide - the thicker wires kept poking out and prevented me from getting the board back in place. Just how much slack do I need?

~Joe

[seedjar]

Somehow my hotend ground came loose. I was afraid I'd sheared it off or something. Back to normal now.

Ran a print and watched the pulleys again. They didn't seem to be loose. I took some high-speed video on my phone. Looked good in slo-mo too.

So what's next? It seems like a common answer for alignment issues is to just push it back in place by hand - is that, uh... safe? I really appreciate your help with this, thanks again.

~Joe

PS - Re: the platform being firmly attached; my hotend feels firmly affixed to the platform, but the platform itself feels like there is a bit of twist in it that I never experienced with the U-joints. The ball joints on both the platform and the carriages feel firmly attached... somehow though the arms seem to want to twist out of parallel with each other.