3D Printing Engineer Reacts To Raspberry Pi Cases | Design for Mass Production 3D Printing

[@slant3d]

Which case or feature was your favorite?

[@ImWillyHeem]

I loved the sleeve case and the remixed mounted version. Very simple and sleek design, which I prefer. Also looked like an easy print.

[@frankearl9285]

Clockspring's entry is like you said...looks great. The main problem is that he's trying to make the ventilation look good. On a Pi4, it generates enough thermals it kind of needs a passive heatsink, so you have to get that OUT of the case if you can. So, in order to do the cool screw knob lockdowns, you have to get the airflow you're talking to. I doubt that Clockspring was working on the premise of manufacturability...it was looks and performance. Now, what's my go-to? That Sleeve case- I've cranked out a bunch of them over time. That being said, the edge cap snap fit has issues for either the sleeve or the rackmount sleeve. For manufacturability purposes, they work, but the end user use is so-so at times. What would you say for this one: www.thingiverse.com/thing:2471005 For a Pi4, it seems to be one of the better designs. Prints simply in the vertical orientation with what I think is sufficiently low first layer interface issues (This is the only at-question thing in my mind, hence my asking y'all...), and does so nicely. It is a largely press-fit assembly with only one M3 screw and square nut holding the whole assembly together. The sleeve case was passed over by myself due to the issues I mentioned previously with the snap-on edge cap for the case. It's not nearly as robust as I would like for it.

[@ImWillyHeem]

Oh and it also looked like it did the best at limiting features that could potentially wear over time.

[@3DPrinting_]

Vesa Mounts by 0110-M-P, I like the variability of the bottom plate and the fact that it just looks good. This is the one I would use.

[@YourArmsGone]

Most of these cases are designed for one off prints for people with their own printers where bed adhesion is a huge variable. They are great for a person to print one for themselves, they aren't made with mass production 3d printing in mind. Great video, I like seeing the different considerations between mass production/batch printing, and single hobbyist printing.

[@NotParadoxish]

Huge bed contact areas aren't a good solution for hobbyists with questionable bed adhesion. The problem is that if you have poor bed adhesion, those big areas are just as likely to lead to warping and curling. Your print might stay glued down (if you're lucky) but you'll end up with different problems. If you're unlucky, you'll end up wasting a lot more material because the print will just fail much later after the warping causes the head to collide with a previous layer or something equally catastrophic. I would (strongly lol) disagree with the idea that there's anything wrong with Clockspring's RPi case, but the advantage of lying flat is that it's an easy one-shot print-in-place design, NOT that it helps with bed adhesion. In fact, you'd probably want to have a pretty good first layer for that print to really work out correctly.

[@Shraedar]

@@NotParadoxish If someone is having issues with adhesion they will likely have more issues with a smaller contact patch on the print bed. I also never understood why people just give up trying to hunt down bed adhesion issues. Sure, there are a few more variables in the diagnostics, but it's one of the most important aspects of 3d printing.

[@ydoucare55]

@@ShraedarI have about 10 years of experience on all types of printers and bed surfaces. Water washable glue stick solves adhesions problems that aren't caused by major z offset or levelling issues on almost all build surface and all types of filaments. I think it's funny that people are still fighting such an easily solvable problem.

[@willofthemaker]

Interesting to hear the perspective from automated print farm manufacturing. One suggestion is to keep the design on the screen while you are talking about it instead of just for a few seconds. Do a picture in picture

[@eslmatt811]

Christmas ornaments would be interesting to see discussed.

[@nigelhungerford-symes5059]

It would be interesting to understand more about filament choice and how that plays into designs for enclosures of electronics that get somewhat warm. A Raspi4 can get hot enough to soften PLA if it touches the SOC chip for example

[@izzieb]

Some of these cases have had a lot of thought put into them, but I think the folding case is my favourite from them. It seems to me like it had 3D printing factored into the design from the get go.

[@slant3d]

Walter seems to be a great designer. Always designing with the end-process in mind.

[@dennisolsson3119]

With PEI and heated bed I have a lot of adhesion, but when the bed cools the parts are not adhered at all. As in I can not tell the difference between a printed and placed part on the bed. It takes time to cool the bed though (5 min or so, less with fab), I don't know if it wears and I don't know if it contributes to the failure rate in mass production. Just saying that if you haven't tried it seriously in a while, maybe you should do some new tests. It would make auto eject work with everything more than a mm thick. Biggest part I have tried so far was a 80mm circle and it was totally unattached afterwards. I have never needed a spatula or similar.

[@WaffleStaffel]

If it has vents on the bottom, it needs to have vents on the top! With few exceptions, convective cooling needs a vertical pathway for airflow.

[@s.sradon9782]

What do you think of designed-in custom supports? ie instead of slicer adding support for this large overhang one can make a snap-off pillar that makes printer bridge the overhang instead.

[@slant3d]

Amazing idea

[@Hilmi12]

I produce flat ABS parts on textured PEI sheet. I print them flat and soon as plate cools up they literally fall off. For production I take out the hot plate and install a cold one so production of the new part can start while the other plate cools off. I get maximum layer strength compared to printing it upright as layer adhesion issues would make for a low strength part

[@Derp3D]

I really love the 3D engineering inside, your doing awesome work pushing this industry forward, 👍

[@christianbureau6732]

Learning each time, thanks !

[@slant3d]

Glad to hear it. Thanks for watching!

[@802Garage]

Sleeved case was my fav. We just printed a very similar case for a vehicle ECU board. I printed it out of transparent PETG. Only issue was warpage on the long bottom edge, the back of the case. It turned into a bit of a banana, but still functional.

[@uiopuiop3472]

chunky non z-dominant prints can also be made ejectable by proper nozzle height and glass bed that it cooled down to near room temp

[@electricalychalanged4911]

The foldable one is just an elegant solution maybe not the most longlived

[@SmashingBricksAU]

😆 That reaction on the last one...classic! Authentic. Another great video!

[@dasdasdadsa5910]

some of the parts you can mass print with SLA printers you need to manual remove but if you take like a elegoo Jupiter you can put like all 50-100 top plates on one build plate have one worker work on this like 1-2 houes and then FDM print all bottom chassis

[@richdobbs6595]

It seems like you are assuming that folks actually print the parts as they are presented in the STL files. Given designs that are individually printed, I would guess that a lot of the STL files show the part as it is intended to be used, and that folks use the slicer with rotations to actually come up with the best orientation for printing. I use OpenSCAD for my modelling, and as I iterate, I often don't update the rotations and translations necessary to allow the slicer to print the part as presented.

[@sublimationman]

Pi's get hot, the holes and patterns that you don't want to go all the way through are needed or some other form of cooling is needed, some even use fans inside blowing on the board and some have heat syncs on certain chips.

[@lucase764]

Hi. I may have missed it in your earlier videos, but did you guys decide on a spool for your filament? I'm trying to find filament on plastic spools, but everyone is going to cardboard. Thanks.

[@wbeel]

I believe they made a vid called "our worst branding mistake" or smth. Prob a podcast They are reusing other brand spools they used and they just put their sticker on it I think? At least for now

[@lucase764]

Thank you much! I'll try to locate the video you mentioned.

[@wbeel]

@@lucase764 kzbin.info/www/bejne/Z3TIhH2smb-lj8Usi=Ct6QfUeOFLI0FARn

[@s.sradon9782]

This "reduce complexity for mass production" idea makes me want to add in custon logos and cutouts just so nobody can ripoff my builds and sell em profitably (like a form of DRM)

[@theebadge]

If you’re mass producing something and ejecting the same part over and over, are there issues with bed adhesion in that spot? Or how is that addressed?

[@Rampamrampapapapam]

The only "Reaction" vids worth watching.

[@jtproduction_]

Love your vids Nice to see the progress of your chanel 😊

[@joell439]

Awesome review. Looking forward to your future enclosure design tips. Thank you

[@TheNadOby]

Enjoy your videous, learn from them a lot and you are going pretty strong with finding new and not gimmicky themes to speak about. I would suggest to look on an open source harwware products that incorporate 3d printing. One that is close to my hart is P.U.M.A microscope, but anything would go. Not just manufacturabilty but overall design. Cheers and thank you.

[@richdobbs6595]

How do you print two parts with a 0.1 mm gap and not have them fuse together? I need about a 0.4 mm gap for the two parts to be completely separate. Apparently, with your proposal to print the two halves and having them support each other, you are hoping for a partial fusing, but that is still separable. But I don't understand how you could hope to do so with only 0.1 mm between the two parts!

[@Achmedsander]

You have probably mentioned this in your other videos, but what is the benefit of printing the case standing vs flat?

[@linkeroniw]

Printing them standing up reduces the surface area so they can be auto-ejected easier rather than having employees take parts of of the beds

[@YourArmsGone]

basically vertical and minimal bed contact makes them easy for a printer to auto eject. Whereas a large bed contact makes them easier for hobbyists all with different machines to print, but removing prints can take longer or require a person.

[@sprobertson]

I thought it was interesting you focused on ejectability to avoid manually removing from the plate, but recommended supports you have to manually remove... would be interesting to see real world tradeoff in a production run

[@GuillaumeLT]

Idea: Taking the first exemple, do hole really save material? Hole vs infill, what is most efficient (cost wise)

[@slant3d]

Infill is always preferred. Adding holes creates more surface area and complexity, which adds print time/cost. We have a video that covers this more in-depth: kzbin.info/www/bejne/d4a1dKeXh9eIfa8

[@TheButchersbLock]

An episode of designed in support would be awesome, great video as always mate 👍🇦🇺😊

[@TS_Mind_Swept]

That last one was ballin Keepo

[@3DPrinting_]

Could you do something like this but for arduino?

[@kokodin5895]

i found out the hard way that 3d printedstructural parts and enclosures really don't apriciate sharp edges, generally 3d printers don't like corners so even though i downloaded a cool fast 3d printed chonky frame design i will probably have to redesign it to faster printing and much less sharp edges because i already snaped 30x30mm profile printed in petg with 0,8mm nozzle and 3 side walls on a joint just because aditional part created a levar and support point fun thing is how it break, catastrophical failure with inside riped out and not really along the layer lines

[@slant3d]

We have a playlist of design considerations for electrical enclosures that may be useful for you: kzbin.info/aero/PLkUv8_afCbJ_k5e-ItHp1KTA8oqqA-9wq

[@richdobbs6595]

It seems like your assessment of design for mass production 3D printing, is really mass production for 3D printing at Slant 3D, given Slant's limited ability to remove flat printed parts from the print surface. Maybe there is a good correlation between these two issue. But it seems suspicious, considering in iterative 3D printing, I hate tall parts that fail often and take forever to print. OTOH, large parts with a lot of bed contact often warp at the corners...

[@slant3d]

Hobby vs factory

[@richdobbs6595]

@@slant3d Yeah, but it is your factory. And your limitations. I think a more honest title would be Design for Slant 3D Mass Production

[@grimsdagger]

Headphone stands and desk organizers would be cool

[@legocreator768]

hello Slant, could you look at SlimeVR Full Body Tracker community cases sometime?

[@plazmax]

i like Raspberry Pi Zero 2 W Case designed by Squirrelbrain but modular one good too

[@gerolori]

do home design pieces! Like knobs, lights, organizers

[@tacticalskiffs8134]

I like the Barch designs for R Pi. There, the 3D print was used as a prototyping exercise for something that was ultimately CNC machined in Aluminum, which is a far better solution. The only enclosure I make is a pencil box for artists that is also an easel. It takes forever to make the lids so I make them out of 1/8" plywood. Makes a really nice product, however, obviously a nightmare if one is looking for something that can be printed and just let rip. However, compared to the all wood version, all the complexity is in the print, and it looks really good. One issue is how to glue wood to PLA. I know my solution works, but whether it works no mater how stupid the consumer is, is something I can't say.

[@smalik..]

to the collection of ideas for videos, housings for Arduino, especially those that use an LCD1602 screen

[@holski77]

How could you not review the Bit Drafted raspberry pi case?!?!? it's the best pi case in the world. And it keeps mine ice cold compared to the other coffins.

[@Adisken]

aaa, what happened to audio at 8 minute :c

[@ziomalZparafii]

No idea why you want printing such cases vertical and saying that flat covers are bad. This seems exactly opposite to what sounds to be reasonable (to me). Printing flat horizontal means we have smooth surface and no need to support. This review sounds totally backwards to me O_o 13:04 what? why such hexagon cutout cannot be printed vertically? A week ago I've designed my custom phone holder with exactly such hexagon cutouts and printed it vertically - no issues at all. This whole video makes no sense to me, all is backward. -1 from for huge amount of misleading info.

[@Vlogger_s]

Mobile accessories

[@slant3d]

Great idea. Any types of accessories specifically?

[@TaseenK]

@@slant3d I have one, phone stands!

[@slant3d]

@taseenk You're in luck: kzbin.info/www/bejne/mmfFpWCHqb-bp6s

[@NoPancakeMixLeft]

Phone gimbals/mount? With how prevalent content creation is, a stable and ergonomic and maybe collapsible mount to hold would be pretty popular i would imagine. maybe even with magsafe compatibility

[@mbunds]

You don't like vents or multi-part assemblies? Sorry, but not everything to be 3D printed can be reduced to a cube.

[@_droid]

That sleeved case is actually not very good. It's in the details. That is, there are none. No work was done to make it a real world usable design. The cap/top is not secure and basically just falls off. Also the Pi is flapping in the breeze inside there with nothing to hold it in place. There is no room for a heatsink... etc, etc The VESA case is pretty good but as mentioned the screws are 100% unnecessary and make it a PITA to work on. Also, the top and bottom are way too overkill thick.

[@willofthemaker]

I think one thing to note in this video is his comments are based off how good a design is for automated print farm manufacturing. Not necessarily how good the design actually works

[@Artemonim]

I would say any case is meaningless unless it is a heatsink. My Orange Pi's run in metal radiator-cases

[@EdRopple]

Radiator cases are by no means bad, but most SBCs just don't get hot enough to worry about it. More modern ones can with something like the RK3588, but something like the Pi 4 or a RK3328 just won't care much under most workloads.

[@crashmaxx1987]

Good video, but I really don't like your phrasing. Most of these weren't meant to print vertical like you prefer, but they aren't bad designs because of that. You can say they don't work for your preferred workflow without making it sound like the creators did a bad job designing them. I get what you are going for, I just find the phrasing a little rude.