MakerJim's MP3DP Repeat build log (Will be slow...)

[MakerJim]

An initial post to start this topic, as promised.
The parts shipment from Ryan for the start of my Repeat build arrived today.
Parts were well packaged and promptly shipped- as expected. Off to a good start.

I started printing over the weekend, initially of the stepper mounts.
I started these on my Taz5, and am using a inland black PLA that is a first use of that particular filament for me. I’ve had a bit of struggle with these parts and I’ll post that below. I believe this is because the filament’s own temp recommendation is too high (and, this printer if anything always seemed to run a bit low.)

The teardrop part of the stepper mount prints great, but on the opposite side where there is a compound outer curve and also the overhang of the circle for the pulley- this is having pretty bad sagging and a bit of curling of the PLA on the edges. The TAZ has cooling from both sides but it’s probably inadequate along with the filament temp being too high still.

Aside that, the lower portion printed nicely with a 40% grid infill and 4 parameters.

I slice these with Cura 4.x and spool them to an Octoprint instance on a Raspberry Pi 3.

[MakerJim]

Here’s the first stepper mount, on the Taz build plate.

Lots of unused build plate area…

Here’s a closer view of the 2nd attempt at printing the stepper mount. I’d dropped the print temp a fair bit. Can see the sagging and bad bridging as well as the curling around the compound overhangs.
Ringing isn’t uncommon for the TAZ, I’m working on that.

Finally, here’s the 4th print. I’d lowered temps even more, and rotated it so the largest overhang is facing the along the X axis so the cooling fans on left and right are blowing straight on it. Actually came out worse as the sagging bridges/overhang caused a layer shift from nozzle impacts- which as fate would have it mostly recovered on a subsequent layer. This could probably be trimmed and still be usable (in fact all of these parts are likely usable- just a bit ugly).

[vicious1 (Ryan)]

My last batch of inland is not working at all for me either. Seems too hot and too cold at the same time. I have used hundreds and hundreds of rolls, the new stuff on the cardboard spools is not good.

I am doing fine with hatchbox, and I have some others coming this week to test.

[MakerJim]

This is the inland black PLA on the cardboard spool.
Interesting. I have a stockpile of older white natural inland PLA on plastic spools, I’ll take a look at that.
I was just asking myself “How could it be too hot and too cool at the same time?”

The thing I liked about this PLA is it reports a 215 to 230 temp range, which means it could be a bit more compatible with PETG (I’m playing with using PLA as a support for PETG on my QIDI Tech1- PLA and PETG do not like to stick to each other).

Tomorrow I may try a print with the other PLA.
I’m also debating one of my other tricks which is to print two copies at once as that has helped a bundh in some hard-to-print items- the extra cooling between layers as the extruder alternates parts may help this.

[vicious1 (Ryan)]

I print full beds and the sagging is still there and the stringing is out of control. I have even tried slowing down and printing at 50% normal speed, still has issues. Whatever they changed is a complete fail. This is actually part of the reason I have been holding off on the LR3, the prints look like garbage and I have to hand trim and flame polish them.

On top of that is hardly sticks to the bed. I have giant parts that work perfectly fine, with this junk I have little 1x1" parts that pop off.

[jeffeb3]

You buy enough filament. Do you have a direct line to their customer service?

[vicious1 (Ryan)]

No, They are not that great at it. I can’t even get the tax exemption for resale. Very poor customer service, I pay full price and taxes (twice). Luckily now, amazon and hatchbox seem to be far better and very similar price. Those to do honer tax exemption and have responsive customer service.

[turbinbjorn]

I bought a 2kg roll of generic PLA a few years back. (3dnet.no) it was super brittle and the whole roll basically broke in pieces. The vendor apologized and sent a new roll from a new factory, which was of great quality. I still get my filament from them, because of this customer service.

[RockinRiley (Riley)]

So there is a filament startup company here in michigan wonder if they could do anything for you. I keep wanting to try them but never get to order ahead of time to get the 8 bucks off for pickup.

Polar Filament

[vicious1 (Ryan)]

Dang that local pickup is a deal! I will have to shoot them an email.

[RockinRiley (Riley)]

If you do tell them i recommended you maybe i will get a free one, lol

Or maybe i get local pickup and ship 2u

[MakerJim]

I’m not having any build plate adhesion problems, but that is an interesting thing about the TAZ.

In the pictures above, you can see that it’s got a glass bed, and I use a specific, really crappy hair spray as an adhesion agent.

Most parts adhere really well, including the black inland filament stuff that is giving me problems. They also release reasonably well when the bed cools back down (I heat to 60C for PLA).
Worst case, I loosen the mounting ears and put a build plate in the upright freezer in the garage for 5 minutes, and all the parts are off it.

A careful look reveals that isn’t a stock Taz 5 plate.
I bought the printer used with really low run time from an older gentleman who had messed up the factory PEI surface, and was improvising his own. The factory Lulzbot Taz5 bonds a silicone heater to a borosilicate glass bed and then bonds PEI to the top of the glass.

What the fellow I bought it from had been doing is to cut down picture frame glass (!!), bond the silicone heater to that, and then mount it in the stock mounting tabs.

The Taz has a horrible (in my opinion) mounting arrangement with four mounting tabs at the corners.

The previous owner had been regularly breaking his improvised build plates and was simply rotating three different silicone heaters and peeling off/re-mounting silicone heaters as he broke them. He was completely frustrated. He also had a collection of other interesting habits for ‘maintaining’ the extruder assembly and I suspect had mis-located the y axis (bed assembly) at some point. I lose about 20mm of print area on this printer as a result, but haven’t had it apart yet to fix that.

At any rate, when I bought it , as a temporary measure I took a 12 inch square, 3/16" aluminum sheet and bonded one of those silicone heaters to one side. I then put one of the previous owners’ spare cut glass sheets on top of that. I printed up some modified corner mounting tabs slightly enlarged to accommodate the 12x12 bed, and slightly thicker as the build plate is a bit thicker than stock.

I had assumed I’d break this in short order, and started working on a better build plate.

I’ve yet to have any problems with my ‘hack’ of a build plate.

In the meantime, I have another version, also based on a 3/16" aluminum sheet, but with an adhesive magnetic sheet and then a removable spring steel pei sheet on top of that. This is again slightly thicker, and I’ve been waiting for over a year to switch to it. No idea if it would be better- my first hack is working and since it isn’t broken I haven’t ‘fixed’ it.

I mention this in part also because I’ve been debating using that same spring steel build plate as the build plate for my repeat build (would need to add some mounting holes for the kinematic mount points.)

I still have one 24V silicone heater pad available, so maybe I’ll build up a fresh one for the repeat. Is anyone else doing similar? That would define my build if I went that route, for a 12"x12’ build plate size.

[MakerJim]

For what it’s worth, here are a few pictures of the build plate that I put together for the TAZ (The one not yet fit to the printer)

Those Lulzbot 24V silicone heater assemblies aren’t cheap. They do work well. The Taz5 never has any trouble heating up or holding temps.

I doubt that these thin aluminum sheets are super flat, and there are imperfections you can see in the magnetic surface.

My experience with the glass build plate and hairspray is that if I use a thick first layer (.3mm) and have done a good job leveling the plate, it ‘just works.’

When I put the glass plate hack together, I spent about 2 hours with an improvised mount and a cheap harbor freight dial indicator running back and forth getting it as level as I could. Haven’t had to touch it since. Just make sure the hairspray is fresh, submit a job, and let it run.

The one thing that is fiddly is the Z axis minimum sensor is an endstop paired with a thumbscrew looking thing with a spring tensioner on it. That walks a bit so once every two or three spools of filament I have to adjust it slightly.

[RootsMedia (Kyle M)]

I don’t mean to clutter your thread, Jim. But I gotta say… I was about to pull my hair out with this damn inland PLA on the cardboard spool. I am kind of glad to know I’m not the only one. This actually makes me feel a whole hell of a lot better about the actual functionality of the printer that I’ve been running it through.

On another note, good luck with your build. Mine has been quite a slow process as well.

[MakerJim]

I’m a guest here, and if this thread helps others out then it’s a good thing. No clutter at all.

As far as the PLA, I switched back to the older Inland PLA and printed two stepper mounts at a time.

The print still was a struggle as the compound overhangs and that bridge around the pully opening.
Up till then the print was great, and after it was still OK.

Two more usable- but not great- prints.
I think I’ve got a bit of work to do working on tuning Cura and the filament.
I also need to find a different PLA supply and have started looking.
It’s a bummer as it was nice to run to Microcenter and just pick up materials on a whim- but that new PLA just isn’t working for me.

I’m considering a mild remix to the stepper mount to add a feature like the shape that Prusa uses on its’ rod holders, sort of a modified teardrop, above the pulley opening. I don’t see a downside and that part just doesn’t print well for me.

But I think I’ll get on with printing more of the parts for my repeat in the meantime and keep working on tuning things.

You can see there’s still a layer shift from all the banging around on those rough bridges.

[SupraGuy (Dan)]

The printer that I did my Repeat with isn’t good at bridges and overhangs, but still didn’t do too badly. A couple of parts got reprints after I had the repeat up and running though.

[vicious1 (Ryan)]

https://teachingtechyt.github.io/calibration.html
That has to be the best printer calibration I have ever used. I highly recommend trying it out. It is fascinating. Some setting have a huge variation window and some need tight control. The only thing he does not really touch on is dimensional accuracy, at least not to the degree I like to tune (probably because most printers can not be adjusted).

[MakerJim]

That’s a very polite way of saying “Dude, fix your $h1t”.

I’ll go study that calibration page.
The Taz has always done well for dimensional accuracy. A couple of weeks ago I made a jig in OpenSCAD to drill 20mm holes on a 96mm grid, laid diagonally across the bed. Came out right on the money.
It’s been temps, flow rates, and recently this cooling and bridging issue that have been the bane of my use of the Taz.

I see in that guide they use Simplify3D. I may just fire that back up. I have 2 licenses (4 seats). Five years ago it was way ahead, but recently it’s nearly abandonware (license servers go down randomly, promising a V5 release real soon now for over a year, etc.). I stopped using it because I hate having tools that suddenly stop working- but maybe it’s time to give it another go on some calibrations.

S3D always did great support structures even with single material prints- maybe I’ll try another print with the inland black filaments that have given me trouble, with supports just for the big bridge in the center. It also seems to produce significantly smaller .gcode files than say Cura or PrusaSlicer/SuperSlicer.

Meanwhile, I did do a print of two Truck Triggers at once.

I did this with the older white inland PLA, slowed the movements down, and dropped layer height to .15mm. These came out pretty good- but took 12 hours to print (!!).
The natural PLA looks way worse in the picture than it actually is, you can see internal features taht look like surface finish issues, but aside from ringing in the print it’s actually pretty good.

[MakerJim]

As a part of staring to work on printer calibration, I fired up Simplify3D.
Using the S3D Taz 5 S3D factory default profile and process, I changed the filament size to 1.75, and the temp to 215. (My Taz has a clone V6 hotend, and has a modified Wades extruder mount instead of the factory 2.85)

I then added supports and removed all but those for the big opening in the stepper mount.

I printed it expecting a failure, and got a usable print. This is with the new Inland black PLA that wasn’t working for me previously (Sliced with Cura)
There is some stringing, but I realize that I had retraction turned off.
No layer shifts, less curling during print.

So, I’ll be continuing to work on calibration and printing various parts for my repeat as time allows.
Also comparing setting-vs-setting between Cura and S3D.

(Tried to enhance an image taken in bad lighting)

This evening I’m going to flash the firmware with DrunkenOctopus (Marlin 2.0 based, developer is the former LulzBot dev who now does DO firmware) This is the same firmware in the SynDaver Axi printer which is a fork of the Lulzbot design.

[gpagnozzi (GP)]

Take a look at SuperSlicer it has some calibration tools built-in