Page MenuHomeAleph Objects Inc

Cluster feedback for redgum
Closed, ResolvedPublic

Description

This ticket to collate the feedback from Cluster for the Redgum machines they will be getting to test.
We need, primarily, longevity testing. Please keep the machines running as often as possible. To that end @EricNugent is modifying some TAZ 6 cluster slices to work on redgum.
Please note in this ticket:

  • Wear on the printers in cluster environment
  • Problems with operating the printer
  • Print issues relating to tool head
  • Print issues relating to bed
  • Persistent errors in printing
  • Cracked, broken, or strained printed parts on the machine after a period of use

These printers will need to be checked over periodically to check for wear and/or broken parts that may not be obvious from within the cabinet.

Event Timeline

There are a very large number of changes, so older changes are hidden. Show Older Changes

It appears that the PP-GP0438's have an issue with the file. The parts are coming out with a "honeycomb" appearance (will attach photos as soon as possible. Also a layer separation on one side. The PP-GP0236 appear to be cosmetically and dimensionally sound. The PP-GP0407's also appear to be good.
Please get additional test files for us to print on WE, and let us know when MER looks at the file of the 438 (I will bring some examples down to MER), so that we can get that part correct.

Shawn added a comment.Jun 17 2019, 6:41 AM

The y-belt mount screws on M1 have fallen out twice since Thursday. @daustin tightened the screws after the first occurrence on 7/13, only for them to fall back out 3 days later. I am wondering if using counter-sunk screws instead of the BHSC screws with washers could be causing this issue.

west added a subscriber: west.Jun 17 2019, 8:23 AM

@Shawn which screws are the ones falling out? the belt mount uses 2x M3x6 FHCS and the belt clamp uses 2x M3x10 BHCS

Shawn added a comment.Jun 17 2019, 8:27 AM

@west The belt mount screws.

Shawn added a comment.Jun 17 2019, 8:29 AM

The 2x M3x6 FHCS

The 2x M3x6 FHCS fell out of M2's belt mount again.

Shawn added a comment.Jun 19 2019, 1:41 PM

The percentage of probe fails is greater than 25%.

karrad added a subscriber: karrad.Jun 19 2019, 2:36 PM

@Shawn We have not completed redgum profiles for ABS in 3.6.11, have you made any adjustments to wiping/probing temperature settings to help with probing issues?

Shawn added a comment.Jun 19 2019, 2:55 PM

@karrad No I have not made any adjustments to temp settings. We are running gcodes for a few parts that were sliced by MER.

@karrad if there are any changes to be made in testing files for Redgum, they will need to be made in MER and then the file pushed and the original removed, so that there is only one file in the Cluster repo for the cluster techs to use. We will want that to go through MER always. Thanks

Shawn added a comment.Jun 20 2019, 6:42 AM

M4 - One y belt mount screw fell out.

@west I know there is a LOT going on. Have we made any changes to the screw falling out issue? Just checking to see if we need to install anything different on the printers in Cluster?

difficulty in printing PP-GP0438. @EricNugent is working on the slice for it

Loctite has been applied to the y belt mount screws of all 4 TazWE's.

adam added a comment.Jun 25 2019, 8:32 AM

@Shawn if you get a chance, please flash all of your WE betas to 2.0.0.135 and run the auto-calibrate through the LCD (under movement) and let me know if any fail due to being too far out of position.

Changed wiper pads for all 4 redgum machines.

Shawn added a comment.Jun 25 2019, 4:10 PM

The TazWE's have been flashed with 2.0.0.135 firmware and auto calibrated. 425 loctite has been applied to the y belt mount screws. I tightened them to 5 in.lb. and allowed the loctite to cure for over an hour.

The blower shrouds on the TAZ WE's are warping downward on the back right corner. M2's shroud is now low enough to sometimes make contact with the part that is being printed.

Interesting, we don't see this on any of the Mini2's in the cluster?

No. Not to my knowledge.

west added a comment.Jun 26 2019, 11:51 AM

Its the same exact toolhead except with hardened steel nozzle and hob

adam added a comment.Jun 26 2019, 11:52 AM

T8313 for sagging issue

adam added a comment.Jun 26 2019, 11:52 AM

@david.hall speculated it is the higher printing temp and larger bed (more ambient heat) that is contributing to the sag

If there is sag on the mini 2's, it's very minor.

west added a comment.Jun 26 2019, 12:29 PM

@adam it would be the same ambient heat as a mini 2 110 on the bed is the same for both a mini plate and a taz plate, the only thing that changed is the nozzle, but the shroud has a structure inside that would prevent warping. @logan can you take a look at this toolhead and see whats happening

@west I believe it is happening on all of the redgums in cluster.

@west what exactly do you want me to look at?
I think the idea that @david.hall was trying to get at is that the area of increased ambient temperature is larger on WE than on Mini 2. My understanding of the hypothesis is that when printing larger parts on Mini 2, the tool head is more likely to be outside the area of greatest ambient temperature for a larger period of the print than when you're printing the same size part on a larger bed. Does that make sense? @david.hall am I understanding correctly?

Shawn added a comment.Jun 26 2019, 1:00 PM

Shall I put a Redgum tool head, with a new shroud, on a mini 2 and see if it also sags?

logan added a subscriber: anolen.Jun 26 2019, 1:10 PM

@Shawn Both shrouds should have been printed with the same gcode. If we're not seeing those sagging to this extent already I doubt it would show us much. It may not behave properly given the hardened steel components if using already generated gcodes.
Speaking of which, @EricNugent how do the extrusion/bed temperatures that are being used for redgum compare to that of Mini2 (without hardened steel nozzles)?
My understanding is that we've found in general through profile development that the hardened steel nozzles require higher temperatures to achieve the same performance. @anolen is that understanding valid?

@logan From what I have seen the layer adhesion is approximately the same using the same temperature settings. I have not done any insert pull out validation to get anything quantifiable, but can do so if needed. I am using the same bed temperatures as the taz6 and mini2 slices that we have been using for some time now.

HS nozzle have shown to need a temp bump but I havent seen anything like this on the Redgum's @hanna and I are working with and we have used ABS on them. We have not tested in an enclosure the quality and ambient temp of Cluster's tho. I would be interested to see, in cluster cabinets, one machine hold a high bed temp with the nozzle cool and low to the bed and another machine hold a high nozzle temp high off the bed for a while and see if either present with drooping.

I adjusted the bed temperature and @Shawn replaced the blower shroud. The workhorse edition slices may need to have a max bed temp for printing in enclosures due to the size of the bed and enclosure conditions.

@logan, You expressed what I was thinking well! I noticed when looking at the Taz6 next to the WE that the space between the fan shroud and the bed surface seems to be less on the WE. Meaning that the fan shroud is closer to the bed surface on the WE than the Taz 6. This may lead to the concept expressed that the ambient temperature is higher than on the Mini and this is contributing to the droop.
I was thinking(may be problematic) that we printed the fan shrouds on our Mini's (in Cluster) from Village Plastics ABS. We have since moved to Chromastrand ABS. Is it possible that the filament change could be an effect here? We could print some fan shrouds out of IC3D ABS and see if we have the same issue. Just a thought.

@david.hall as far as I know blower shrouds for mini2 were all printed in chromastrand. We switched from VP to chromastrand before beginning production on mini2.

adam added a comment.Jun 27 2019, 10:50 AM

@Shawn have you tried @anolen's suggestion with holding the tool head close to bed with bed at temp and nozzle cool on one machine and holding the nozzle at temp with the bed cool on another?

@adam I have not, but I can set that up right now.

I replaced the shrouds on M4 and M7. M4 has a bed temp of 115 with the nozzle close to the surface. M7 has the nozzle heated to 250 in the home position.

jmielke added a subscriber: jmielke.EditedJun 29 2019, 3:28 AM

M1 is printing uneven-squished parts. Was printing mostly pp-gp0236s. Also tested with pp-gp0407 and it still printed squished. Tried level x-axis function to no avail.
Left some of the squished parts in front of the machine to showcase behavior.

7/1/19 Edit: After leaving the machine (m1) sitting for a day tried some more parts; and had no more squish issues. So it randomly fixed itself.

logan added a comment.Jul 1 2019, 8:56 AM

@jmielke Please check to ensure M1's Z-Pulley set screws are tight, with one engaging the flat of the motor shaft. We have dental mirrors for inspecting this, if you need one let me know.

jhan added a subscriber: jhan.Jul 1 2019, 2:06 PM

the y belt mount screws came loose again on M4.

jhan added a comment.Jul 1 2019, 2:11 PM

Apparently, there was loctite on the screws

logan added a subscriber: tutley.Jul 1 2019, 2:26 PM

I'm very surprised by that, and uncertain of another solution. @west @tutley @adam ^ ideas?

Do we need to do a BHCS with lock washer? Would it even work on ABS?

karrad added a comment.Jul 1 2019, 2:30 PM

What has changed in the design with the way we are attaching the belts? I don't recall this being an issue in TAZ 3-6, can we look at what is different or maybe change back to that mount style?

tutley added a comment.Jul 1 2019, 2:44 PM

just need to confirm, was the loctite applied to the threads, or just the bottom of the screw head?

what is the ambient temp of the cabinets in the cluster? The loctite 425 should be good up to 85C

this issue has only been observed in the cluster, which makes me think it is heat and/or resonance related.
The cabinets are in a constant state of vibration due to the other printers being in motion. This vibration is what i believe is causing the screws to back out, and the heat is assisting with this as the hole in the aluminum plate gets larger as the material expands

anolen added a comment.Jul 1 2019, 2:48 PM

@Shawn any droop yet on M4 and M7?

logan added a comment.Jul 1 2019, 3:01 PM

@karrad For Quiver/Redgum we are using continuous belts, so the clamp is only a clamp and doesn't double as the tension mechanism. The mount has greatly decreased in footprint with the endstop switches moving to the Y Motor Mount, and is using 2 fasteners instead of 4. We're using M3x6 FHCS, for TAZ 6 it was button heads from below, TAZ 5 and prior were socket heads from above into inserts below. This belt mount/clamp has changed many times from TAZ 3 to 6. The initial Y Belt Mount for TAZ 6 was awful and had a high fallout rate. Also because it was the tensioning mechanism it would frequently lose belt tension in the cluster and resulted in lots of circular holes arriving to production not being circular. Revisiting those mount styles should not be pursued. If the following doesn't resolve the issue I think our only other option is to redesign the current mount to use a different fastener, likely button heads with washers.

@david.hall @Shawn
Lets try this again with a higher torque spec and a longer cure time. Use 8in*lbs torque and allow a full 24hrs to cure before returning them to use. Also please ensure a brand new bottle of Loctite is used, just in case storage/shelf life has reduced the effectiveness of the Loctite. And fresh fasteners as before.

@Shawn will not be here till next week. The loctite was applied to the threads and then screwed in. The Loctite should be fine at 90f.
@anolen I will have to look tomorrow and see if we have any droop on the M4 and M7, I assume you are talking about the fan shrouds?

tutley added a comment.Jul 1 2019, 3:09 PM

maybe not enough was applied?

karrad added a comment.Jul 1 2019, 3:11 PM

@logan Ahh, thank you for the thorough explanation I appreciate it! (Absolutely makes sense in my mind)

@logan, we will see if I have time to get to the printers tomorrow to install new hardware Loctite and torque the screws to 8 in lbs. We can get a new bottle out tomorrow and see if that helps.

@jhan will be looking at the fan shroud and making an entry soon.

logan added a comment.Jul 1 2019, 3:18 PM

@david.hall with @Shawn out if you need any assistance with this let me know

jhan added a comment.Jul 1 2019, 3:26 PM

@anolen
M4 has shown plenty of droop, and M7 has shown some, but not the the same extent.
@tutley i do not know. when shawn gets back , he should be able to better answer your questions. i'm not actually the one currently doing repair.

logan added a comment.Jul 1 2019, 3:35 PM

Talked with @david.hall in person just now, we are going to reinstall the one fastener of these machines that has an issue using the method above and see if the others loosen as they were (5in*lbs + 1hr cure)

M4 is grinding on the X axis when it goes to the left, so we shut it off.

tutley added a comment.Jul 8 2019, 7:19 AM

@jlively is the grinding coming from the idler bearings? the x motor itself?

@tutley , let me get with @Shawn and have him start the printer back up. @jlively is the weekend shift and did not make good notes on what was happening. i will see what Shawn finds on the printer and we will document it.

tutley added a comment.Jul 8 2019, 7:41 AM

@david.hall sounds good! thanks!

Shawn added a comment.Jul 8 2019, 2:24 PM

I haven't been able to reproduce the x axis grinding. The reported grinding might have happened because somebody mounted the spool arm so it pointed towards the door of the cabinet. The feed tube holder was also relocated to the right channel of the front-right aluminum rail. I did not test the machine before I moved the arm and the feed tube holder to their default locations.

We just got 2 brand new Taz WE's and one of them extruder can not maintain temperature

adam added a comment.Jul 11 2019, 12:58 PM

what firmware are they running?

Shawn added a comment.EditedJul 11 2019, 1:38 PM

the machine that can't maintain temp has 2.0.0.139. The other machine has 2.0.0.135

adam added a comment.Jul 11 2019, 2:27 PM

First thing we need to do is flash all redgum to 2.0.0.135 - if they still can't maintain heat correctly I think we should have R&D take a look at them to diagnose.

adam added a comment.Jul 11 2019, 2:28 PM

you will need to flash the aerov2 firmware as there is an issue with the hardenedextruder firmware.

Shawn added a comment.EditedJul 11 2019, 2:54 PM

The machine in M4 won't print due to bed heating fail.

Shawn added a comment.Jul 11 2019, 2:55 PM

I don't know if I will be able to flash the firmware today. If not, David will have weekend shift do it tomorrow.

Shawn added a comment.Jul 16 2019, 1:17 PM

M2's extruder motor stopped functioning. Replacing the Rambo fixed the issue for now.

Shawn added a comment.Jul 16 2019, 1:18 PM

T8802 - ticket for Rambo replacement

M2 and M4 have uneven bead height, were it squishes one corner of the part every time. When probing, both printers press down really hard on the back right corner, which is where we're getting the squish. We've cleaned the nozzles, leveled the gantry, and this problem is still consistent.

M1 is now having print fails. Part starts off fine, but the top half becomes ragged. We left the parts by the printer.

Shawn added a comment.Jul 22 2019, 7:52 AM

M1 failure caused by screws falling out of the y belt mount. Applied loctite and will cure for 24 hrs.

Shawn added a comment.Jul 22 2019, 8:30 AM

Uneven bead height on M2 and M4 has been resolved by tightening the screw that secures the z min wire to the toolhead, and cleaning the bed corners.

The shrouds on the Red Gum's are sagging again.

When running a TazWE with the V3 installed, extruder 1 retracts twice and extruder 2 does not retract before wiping. Ticket can be found here - T9015

jmalley added a subscriber: jmalley.Aug 4 2019, 4:32 PM

M4 is having prints fail at the same places every time. I left a part on the bed. We are not sure what could be causing it.

adam closed this task as Resolved.Fri, Nov 22, 11:42 AM