HewlettHackard

Or too thin (much less than one layer height)?

It’s not necessarily fluff; features do matter, but which ones matter and how much they matter is complicated. Do you want to tinker with a printer, or do you just want a tool that works?

If you want a tool, I recommend a Prusa or maybe a Bambu. My Prusa just does what I ask it to; I’ve done zero calibration, optimization, tweaking, etc. Cheaper printers often require understanding bed leveling, figuring out how to adjust them best, etc. or vary more copy-to-copy (so one Ender 3 might work great, while another might be a source of frustration) or require upgrades to really become enjoyable.

Do you have a budget? Or is it just about making sure you’re getting your money’s worth (so $2000 is fine if it’s 10x better than a $200 printer)?

Yup. My kid found the idea hilarious when I explained why I’d swiped a washable glue stick from the arts and crafts box.

It hangs on well, but also makes removal easier.

You could always try it…

I use PEI instead of glass and find that a thin layer of glue stick works great for helping sticky materials release.

Could you share some examples of what you’ve been able to create, and approximately how long they took?

Wow, thanks for the incredibly broad write up! That’s a crazy number of tools to try.

FreeCAD:

giving up as soon as any two edges’ fillets touch

I haven’t quite had that issue with FreeCAD; an ordinary corner works fine. Where I’ve had that issue is when the fillet is large relative to the edge sizes though, so there are definitely failure cases.

not working well if you don’t learn its definition of good design

This is so incredibly true. I’ve had to restart projects because I didn’t realize how I’d need to approach them to get FreeCAD to allow me to do what I wanted to do. That’s probably the biggest issue I have with it: you really have to know how it wants things done, and often you won’t learn that til you’ve tried, failed, and watched a couple YouTube videos from more experienced users. Fortunately, after a few projects I’ve gotten to the point where I can usually get relatively simple designs right the first time…but I have no doubt I’ll still sometimes have to start over or take big steps backward to change my approach to a problem.

Others have mentioned OpenSCAD and that’s what I started with years ago; it’s fine for simple things (or fractals/programmatically specified forms) but when designing something with more complexity I find it very difficult to manage, compared to just directly drawing what I want and then constraining it appropriately.

You’re right, 50%@70f (let’s say Texas indoor conditions) is 19%@100f, so you’re right that just heating is probably perfectly good.

You’d need a very sensitive filament to bother doing more. That said, if you did have a peltier with the cold side at 32f, you’d get 25%@70f or 9%@100f…which actually isn’t amazing since you still need to warm it up to 100f to get that 9%! Peltiers are inexpensive and I might still experiment with it one day, but…it doesn’t seem like a huge win.

I see. You have a huge delta-T there, which is unnecessary for keeping a filament box dry.

A couple sources say PA6 (a random moisture-sensitive filament) should be stored below 20% , or at 15% relative humidity. In a 66f/19c room temperature, that 15% RH would be a dew point of 17f/-8c according to dpcalc.org. In a 74f/23c room, 15% is a dew point of 23f/-5c; as you warm up the temperature further the dew point to achieve a particular RH increases (which is why we heat air to dry things).

Huh? Running it backward as a thermoelectric generator is low efficiency (Carnot limit) but the efficiency pumping heat from e.g. freezing to room temperature isn’t that awful. Picking some completely random model from DigiKey, if the hot side was at 27c/80f, and you wanted the cold side at 0c, 2.4 amps at about 7V (16.8W input power) would let you remove about 15W from the cold side with a deltaT of 30c (i.e. the cold side at -3c/27f) per the graphs on page 5. That’s a COP of about 0.9 for cooling. You’d need to remove about 30W from the hot side.

In practice, cooling the hot side is difficult so you might instead use the Th=50c graph, at which point 2.4A @ 8v (19.2W input power) would give 3-4W cooling power with the cold side at 0c, for a COP of about 0.18. But nowhere near your numbers.

If you put some decent thermal mass on the hot side and operated intermittently, as long as you drained the condensation, you could get closer to the Th=27c performance.

I don’t think the efficiency is a big deal; you’d only have to run the peltier intermittently (e.g. each time you open the box) as long as you can drain the condensation somehow and it’s otherwise well-sealed. I’ve idly pondered OPs idea myself and I think it’s not a terrible way to dry a box. You avoid the consumables or need to recharge your dessicant materials.

One other thing that occurred to me is squish. Just like your first layer issues, the plastic on every layer needs something beneath it to squeeze against so it converts from a circular tube to a flat, wide oval. Maybe your slicing didn’t have enough support or infill in the main body as compare to the final top surface? You could try increasing the horizontal shells.

By the way, I assume you did not turn on ironing, because that could also affect the surfaces differently if one got ironed and the other didn’t for some reason.