I ran out to Costco in between meetings Friday, where I found Clorox sanitizing wipes for the first time since March. A 5-pack of giant containers, of course, Costco being Costco, and it turned out that two of them had been crushed just enough that they’d be dried out by the time I wanted them, so I pulled out the wipes and double-bagged them in freezer bags.

Before I left, I started a six-hour print job. When I came back, it was finished. Or, more precisely, it wasn’t printing any more, and there was no error message on the display, but it was only maybe a third done. It looked fine, I had plenty of filament left, and a quick extruder test showed no issues of any kind. I wish I’d left my status script running, which would have captured any error messages.

Lesson learned. Reprinted the same file and it worked perfectly, of course.

I couldn’t understand why my print-in-place hinges were failing. They worked perfectly in the Squeaky Sunlu PETG, but they were so fused together with the Hatchbox PLA that trying to break them free broke the model instead.

Then I discovered that I’d somehow set the extrusion multiplier to 1.05 for that filament, when the correct (calculated and verified) value was 0.97. Oops.

ObAmazon:

If you want to print stackable boxes…

…this is a well-designed set, with free samplers. The only real limitation (besides the time and cost of 3d-printing boxes in the first place) is that stacking only works one way. For instance, you can stack 3 1x1 boxes/lids on a 3x1, but not the other way around. They’ll be almost stable, but not quite.

I’m going to reverse-engineer the design into a parametric OpenSCAD program. Not to undermine his sales, but to alter the basic layout (multiples of 60mm X/Y and 20mm Z) to better suit my needs. If I ever get to drive to a kumihimo conference again, I could do with some better organization for supplies. The incredibly cheap stackable pencil and crayon boxes at Michaels were the sensible choice for my large collection of tama of different weights and sizes, but there’s a bunch of fussy little stuff that still ends up in ziploc baggies and makeshift toolboxes.

I could also do something like the box inserts people are creating for board games (one of the refreshing non-printer-parts categories of object being posted), but I like the flexibility of the grid design.

+1 Bed Of Adhesion

The Dremel 3D45 has a removable glass print bed. I have two of them, so I can pull out a finished print and immediately start another one, allowing the first bed to cool and release the print.

Occasionally, objects with large flat bottom surfaces show a diagonal wave pattern that indicates the nozzle is too close to the bed, something I thought I’d sorted out weeks ago. It wasn’t until I tried printing various samples from the above box set back-to-back on freshly-cleaned beds that I caught on: my two beds have slightly different heights. I’ve now labeled the one that requires an extra +0.1mm Z offset.

(I really miss the tool-height sensor on my Nomad CNC; a lot of 3D printers have automatic bed-leveling systems, but you pretty much have to go to industrial models to get one that can verify the exact Z position of the nozzle tip(s))

Microfiber and a binder clip

I went with the simplest filament filter and loaded up the Sunlu PETG again. Much better, with only the expected minor defects of a filament I haven’t gotten the parameters quite right for. The only poor result came from printing something large enough that the skirt ran all the way to the edge of the bed, and lost adhesion in a few small spots. It didn’t harm the print at all, since the head never even passed over the skirt once it got the first layer down. Either I didn’t have enough hairspray at the edge of the bed, or it was just enough cooler at the edge to pull up the corners a bit; both are fixable, since I was at the low end of the temperature range and I have lots of hairspray.

After several quite successful prints with the Sunlu PETG, I started getting occasional defects on the vertical edges, accompanied by squeaking or clicking noises. After some basic troubleshooting and a few hits of the “purge” button, I watched the next print lay down a perfect first layer followed by a big glob of filament that stuck up high enough for the nozzle to plow right through it and create a loop.

I switched back to the Hatchbox PLA and kicked off a 5-hour print, and had no defects or unusual noises. I’ll have to inspect the Sunlu filament carefully; with luck it’s just surface cruft that can be removed with a filament filter.

…as soon as I find a filter design that is A) sensibly printable, B) doesn’t cause immense drag on the filament, and C) doesn’t occasionally shed foam onto the “cleaned” filament. Most promising idea I’ve seen so far is to double-fold a small piece of microfiber cloth and attach it around the filament with a standard binder clip.

(and, no, I won’t use it to “oil the filament”; that seemed like a bad idea even before I read up on it)

In other news, my two free spools of Dremel PLA arrived, and instead of the old half-kilo spools I was expecting, they’re the new 3/4-kilo size. Both blue.

Thingiverse is in a horribly broken state, and apparently has been for quite a while (1, 2, 3, 4, 5). The Customizer feature has been down for months, and the search engine still hasn’t registered the miter box cam pin I uploaded over a month ago. Despite being downloadable from the direct link, it doesn’t even show up on my profile page.

(a lot of people seem to be migrating to Cults3D, which is at least functional, although the ongoing infestation of garbage seems to be clobbering its performance; I mean, how many dozens of kit-bashed buddha-with-another-head variations does anyone really need to upload in one day? Also, you have to create an account before they’ll let you download anything, and they don’t have an equivalent to Thingy’s customizer at all)

So I haven’t bothered to upload the customizable spool adapter I made to fit Sunlu filament to the Dremel 3D45 external spool holder.

Printed at 0.3mm layer height, all four pieces only use up 17 grams of filament, and take less than 90 minutes to print. OpenSCAD source after the jump.

more...

The external spool holder works well, although it’s a tight fit on the 3D45’s built-in spool-holder cylinder. The creator originally designed it for a 3D40, which has a slightly different diameter, and apparently used someone’s measurements for the 3D45 without personally trying it. I opened it up a bit with some sandpaper wrapped around a dowel, and it went into place without force.

Does it work? Yes; I anticipate no issues with reliable handling of third-party (and most Dremel) filament. The only obvious difference is that the open side reduces the chamber temperature by ~5°C. My makeshift dust cover, a 12-quart square Cambro propped up on its side over the spool and opening, cut that to ~1°. Instead of building a dust cover, I might just end up making a better stand for the Cambro and put a small tray of dessicant inside to keep the filament dry.

Fun Dremel firmware fact: if you print the same filename twice in a row with different contents, the printer will remember the total time of the first one, and report bogus completion percentages for the second one on the display and the API.

Having used up most of the purple PLA, I switched to the roll of translucent PETG that came with the printer. This had to go inside, because despite being one of the new 0.75 KG spools, it has a slightly smaller hole. Since this is my first time working with PETG, and I’m using PrusaSlicer’s generic profile modified with Dremel’s temperature settings, I started small, printing a known design that didn’t require supports.

It worked perfectly, and gyroid infill looks pretty cool showing through the translucent surface. Not an effect I’d always want, but it can add some visual interest to an otherwise boring object. The second print (one of the ubiquitous “thumb book holder” designs) at first looked like it had adhesion issues, until I re-checked the model and found that the long, thin ends of the arms were curved up so they didn’t touch the bed.

I’ve got a few kilos of colored PETG coming tomorrow, so I think I’ll save the rest of this one for things that would benefit from the frosted appearance, and switch back to PLA for today; I promised to make a batch of babydai koma for someone.

File under ‘baffling’…

In one of my recent Amazon orders, I included a 3-pack of Aqua-Net unscented super extra hold hairspray, since it’s been working out so well for print adhesion. Amazon reports that it shipped yesterday FedEX Ground from the vendor “challways”, and will arrive sometime after January 8th. A 4-pack arrived today via OnTrac in a walmart.com box with a jet.com return address on the shipping label. It’s the right stuff, though, so I’m kind of wondering if more will show up. The 3-pack was already going to be enough to last me a very long time, so I hope not.

Below, Mamako solves an adhesion problem…

In other news…

Valentine’s Day store displays are going up already.

Most Useful Thing

Thingiverse has a clunky and unreliable indexing system and search engine, to the point that it can be hard to find the same thing twice. So here’s a link to a quick, useful tool: Angle taker for tight spaces.

A while back I made a stab at designing a removable drip tray for the water/ice dispenser on my Samsung fridge, which was inexplicably designed with an oddly-shaped depression that will inevitably become etched by the residue of evaporation. I basically had to take rubbings to get measurements, and my first try got the angled sides wrong.

This little tool captured the angle precisely so I could transfer it to a protractor and correct my OpenSCAD code.

Other Useful Things

Bag Clips

While I generally frown on the idea of spending hours 3d-printing something that can be purchased for pennies at any local store, 2020 and Benito Newsom have made the shopping experience so cumbersome that I’m willing to make a few exceptions. I generally have at least three opened bags of something or other that need to be resealed cleanly, and this set of 3 clips works nicely. There are quite a few remixes and redesigns, but they don’t solve the only problem I’ve had with them a few times, which is the ends lifting off the bed slightly during printing.

It doesn’t hurt them functionally, and it can be avoided by keeping the infill percentage and number of walls down, to reduce shrinkage-induced stress. And we should all reduce shrinkage-induced stress.

2x MicroSD in SD case

When there are once again places to go and things to see, I will take pictures with my cameras. My pocket-sized travel camera fits nicely in this padded pouch, with only the security strap sticking out. The pouch has just enough room for an extra battery and some thin card storage.

Unfortunately, most of the printable MicroSD holders overcompensate for their tinyness by adding substantial bulk. This very simple design is the notable exception, just fitting two of them into a standard SD snap case, of which I have a lifetime supply.

Not a tool…

Apple’s iCloud service crumbled under load on Christmas day, and took 36 hours to recover. The richest company in the world is still struggling to build an online service that can scale.

Remember the days when you could unbox a gift on Christmas morning and it would actually work? Clearly no one at Apple does.

Unrelated

Got Black Ships? Japan has banned foreigners again. Well, at least until the end of January.

Pete, I’m getting great results with a very thin layer of Aqua Net Unscented Extra Super Hold. More reliable than the glue sticks with much better bottom surface quality. Best applied with a microfiber cloth that’s been wetted with a replacement cap like this one.

PLA (what I’m mostly sticking to right now) adheres very firmly to a heated bed, but pretty much pops off as soon as it cools.

(note on ABS, since I belatedly noticed that’s what you’re printing with)

Unrelated, plastic calipers. 😁