|
I printed up one last swatch. The "START" text came almost immediately. Everything looks great on it except the "A", which isn't that bad either. Probably just stick a little extrusion of material to widen the center hole on the real thing and hope that works. This reminds me of Mario 64. Weird. Also, I've been notified by my friend that this thing looks like a bomb.  Good luck taking this through the airport, fucker  More Enclosure Work  Wireframe makes everything look really cool. I whipped up a second iteration of the left controller cover with most of the tweaks I mentioned before.  Looks better, and fits even better. It's playable now that the shoulder buttons aren't stuck. There's still some more issues: * Back-off the top screen bezel by maybe 0.5mm. It's covering too much now. * Not sure what the lifting around the "S" in the SELECT was all about. It was identical to the last font swatch. Probably need to wipe down my bed again. * Shoulder button holes are widened so the bridged area is wider and droops more. Also, since the holes are wider, that area of the case is even more brittle than before. * I used a 35 degree chamfer on the Select button hole and it doesn't seem to make a difference, but has worse print quality. Gonna put this back to 45 degrees. Also I remove the speaker holes and just put 3 holes in a row. I think this looks fine, but something interesting I'm finding out is how tough an task this is for a 3d printer. I feel like making enclosures with a lot of details on them is not really a strong suit of FDM printers, ironically.  This iteration also has a mockup of the inner lip and back enclosure attachment screw holes. Now it looks like a real toy.
|
#
?
Oct 27, 2021 06:36
|
|
|
|
| # ? Apr 25, 2024 16:17 |
|
|
Cory Parsnipson posted:This doesn't really seem like something I should be using the 3D printer to do. I'm gonna try shrinking the latest "SELECT" with Tahoma and Arial to 3.5mm and see what happens. If not, I might just use "SE" or "SEL" instead of the whole word. Please take the advice of someone who went all-in on 3d printing a while ago: The printer is one tool in your toolbox, and you have others. This points up your second point: post-process your print. A little bit of sandpaper goes a VERY long way. If you finally get your design to a place you love, print your final in ABS and then solvent vapor smooth it. For your text, you can throw a cheapo diode laser module on your print head and use it to laser-etch text. Or stick-on labels. Or paint. Or a word spitter. Or metal stamps heated to emboss text. Or any other more suitable tool.
|
|
#
?
Oct 27, 2021 09:12
|
|
|
babyeatingpsychopath posted:I've had to do text pretty big to get it to work. I agree, this isn't something you should do solely in the printer unless you've got resin. Or, for your Final Version, get a .2 or .15mm nozzle and just let the print take a week. Probably not going to get a small nozzle for this. The longer print time just isn't worth it for me. I think having interchangeable cases like the 3ds does is a must. I think it's so much fun to have the different characters and stuff on them, especially if you could 3d print one and put an image on it yourself. Also, I definitely see the appeal of the custom GBA modding community that makes their own neon and pastel colored cases and buttons. Experimenting with filler, sanding, and paint is on the TODO list at some point. Also I want to look into seeing if that fancy hydro dip techinque can put a printed image onto the case. Oooh boy that would open up a whole can of worms.
|
|
#
?
Oct 29, 2021 22:45
|
|
|
Front Enclosure Plate modelled I stared and clicked repeatedly on this rectangle until it resembled technology! I was trying to think about how I should approach the enclosure modeling next, when I decided that I should settle on the interface (and by that I mean screw holes), print the whole front plate, and then experiment with pieces of the back that conform to what I have with the front.  Man, the wireframe of the top plate makes it look really cool. Keep in mind, this is actually about the size of a Game Gear but longer, instead of the svelte Anbernic looking device that it looks like in the picture. This would be so loving cool if the entire device could fit in the front plate. Alas, until I make another one, I can only meme about it.  dis me Here's an in progress shot of the inside of the front enclosure piece:  When I place it into the chassis assembly, it fits around everything with room to spare.  The reason for this is because the plate needs to go on shoulder buttons first, diagonally. So there's space on the bottom to push the plate upwards while putting it on. I also wasted a lot of space on the sides because of that one area on the left that juts outwards. I can focus on thinning this thing down the next time around.
|
|
#
?
Oct 30, 2021 09:31
|
|
|
Technical Difficulties, please stand by Just popping in here real quick to say that there might be a delay with progress on the enclosure because I've found that one of my printer fans is failing and I need to buy a replacement and take apart the printer to swap it out. I can see sections on the last few parts that look like the fan stopped spinning intermittently and I have at least two 9 hour prints to do next, so it looks like the maintenance has to be done now... Also I think FreeCAD is giving me carpal tunnel, so maybe a few days away from the computer would help with that.
|
|
#
?
Oct 31, 2021 23:39
|
|
|
 Incident Report Stupid chimp-brained moron (henceforth known as User) has slammed the device in a door. The circumstances of the situation involved trying to pass through an open portal with User's hands full and using the device-holding hand to grab the door knob and pull the door closed behind User. A lapse of judgment was responsible for User not realizing that the device is long enough to extend past User's hand and get caught between the edge of the door and a piece of the door frame. Collateral damage includes cracking the plastic molded 3DS "B" button and some minor cosmetic damage to the 18650 cell battery and holder.  wee woo WEEE WOOO  Now I understand the fabled "risk of puncture" that people keep talking about. Luckily, the B button still works flawlessly although it is an infinitesimally slightly less clicky than before. Probably more lucky is that the battery is unscathed and not shooting an unending stream of fiery sparks. User is quite surprised to find that the device's inner frame and components are more sturdy than User thought. Suggested action: Unknown. It is recommended to finish the external pieces for extra protection. Waaay down the line, I want to get into doing some destructive testing, but that's a fairytale at this point. Quiescent Power consumption update Honestly, this data is quite disappointing. I'm seeing brown-outs of non-game related activity (such as compiling or using makefile commands) at around 28% to even up to 35% of the battery. After a few charge/discharge cycles, the fuel gauge seems to have settled into reading 3.8ish volts as 100% and 3.4ish volts as 0%. The voltage range being small is exacerbated by using a boost converter. I'm not sure why the charger doesn't go past 3.8. This could be due to a misconfiguration, or maybe what happens when the cell stops being new (it's nominal 3.7 volts), or who knows what. It really feels like it drops a percentage point like every 30 seconds, so I wouldn't be surprised if the total battery life is like around 40 minutes or something. I'm gonna reset the fuel gauge some time in the future and try again, just to make sure. My first few data points were collected with the cable connected, which always causes the voltage to read around 4.12. Also I should charge up to 100% and then completely discharge until a brown out for the first cycle. Maybe the accuracy will improve. Oh well. This just means that there's room for massive improvement when I switch to a dedicated power supply. CAD work continues The replacement fan for the 3d printer should be here Friday. Until then I realized that I can still print throwaway parts if I don't care about perfect first layer adhesion or surface finish. So I guess I can actually get started modelling pieces of the back enclosure piece that I know I will end up throwing away.
|
|
#
?
Nov 4, 2021 04:53
|
|
|
Making a 3D printer shrine I got depressed waiting for the new fan, so I bought a table for my printer. It's not perfect, but I guess it's a good start. Gotta figure out how to get Wayfair to stop emailing me now.  Status overlay mods Also I added a screenshot function to the long press power button: code: Things are getting real home-made software-y now. I don't think I'll keep it around, because I too have a Nintendo Switch and accidentally take screenshots way too often. But this'll make getting playlog content easier for the time being. Metroid Zero Mission (for real this time)  Call on meeeee/Call me Watch the battery go down in real time on the gif above.  The last session, I ended up exploring the main two vertical columns of Brinstar and stopped there.  Noo! My only weakness: bombs! I was stopped by some bomb blocks and I don't have those yet.  Found the map, though.  And quickly found the bombs after that.  Still, I am stopped by my other only weakness. Jumping high. Drat.  Bombs have opened up a hole in the floor leading to the next chozo. (I could have sworn I got a screenshot of it...)  Time to go to Norfair.  Also, here's an extra room with a missile tank just teasing me. There's bees in this room that make it extra annoying to jump up to that area. BEES  Haha! You won't stop me, MACHINE. Bombing the red stripe causes it to raise up.  gently caress. My other other only weakness: ledge grabbing.  I wandered around some more and got lost. Then came across a giant enemy centipede. Playing this part was a little tricky because I had to press the R1 button with my right hand to get the missiles out, move, and aim at the same time. That's a little difficult right now because I'm pinching the bottom corner of the console trying to avoid touching anything on the back. Once you get used to it, it's doable, though.  This thing was hiding a charge beam!    Norfair is very purple and extremely on fire.  Before heading to the little glowing dot, I went in the opposite direction and found some chozo ruins. I remember playing this game a long time ago that the ledge grab is here.  SpoooOOooOOoooOOOooky  Noice.  Noice.  I found my ship from the opening cutscene. Crashed here for a while.  I went back into the chozo ruins to explore the other half. Alas! My other other other only weakness: superbombs! Also, I forgot how gorgeous this game looks. It's especially apparent from the comic shaded background here. It looks even better than it does on the original gameboy, since I have a bigger, backlit screen, albeit a little pixelated since it's scaled 2x.  Finally, I'm getting the rewards I deserve.  Upon heading to the glowing dot in Norfair, I'm presented with this conundrum. And I slowly realize that I was actually supposed to get the ledge grab before coming here. I thought I was being clever by visiting the chozo ruins first, but it looks like the game developers intended for me to do that from the very beginning! They played me like a goddamn fiddle!!  My retreat here is cut off by a super missile door. But I got the ice beam, so I guess it's a fair trade.  Next destination: Kraid. In contrast to Super Metroid, Kraid loving rules in this game, while Norfair was pretty awesome in the older game.  AAHHHH IT BURNS  I went back to the BEES room and got the missile.  Goddamnit. I also went back to the ledge grab over lava, but I'm blocked by a standing chozo statue. I have to come in from the other direction, but this requires you to freeze 3 flying beetles and have high jump boots. I'll have to come back for this later.  That super missile is just sitting there and gloating.  Can you hear this screenshot? This music freaking slaps.  I got lost again and came across a giant enemy bobbit worm. This involves jumping and then landing and shooting missiles straight across. Interestingly, it was easier on the controls than the centipede. I remember dying to this thing like a million times when I was playing this as a kid.  I explored a bunch of Kraid. The top right corner has a couple missile tanks and some platforming puzzles. Whelp, time to go get the dot. Also, this is where I had to take a piss and then I remembered that I made a goddamn portable machine. So I kept playing as I went to take a piss.  This is where I got the low battery warning when it hit 10%. Unfortunately, at this point, my controls stopped responding and I couldn't do anything. Jstest showed no response from the Arduino, so either it died from undervoltage, or it crashed somehow when the low battery warning fired. I think the latter is more likely since it happened right after the low battery warning showed up, but I still need to investigate. I had to hold the power button to do a hard reset to get out of it, since I couldn't even exit the emulator. Playtime stats I'm very proud to say that the playtime was longer than 1 hour. According to collected data, I started playing around 12:02 AM and ended at 1:21 AM. 1 hour, 19 minutes! And I was still at 10% ish, though I did get cut off by the controls issue. I actually kind of got bored and wanted to stop playing slightly before this point, which was something I wasn't expecting to happen. It looks like running an emulator is much less power hungry than running shell commands. I suppose the RetroArch team did a lot of optimizations and stuff during the emulator runtime. Very nice. Showstopper When I plugged it into the outlet and rebooted, thankfully the controls were back. The bad news is that retroarch doesn't save the data until the emulator exits and if it doesn't exit gracefully, the data is gone... So I lost the last hour and 20 minutes of playing.  I'll need to go through all that stuff again. And also, I found this page to change the setting so it doesn't do that.https://www.reddit.com/r/RetroPie/comments/60a53t/dont_lose_saves_set_this_option_kind_piece_of/ Seriously what the gently caress, this should be the default. Cory Parsnipson fucked around with this message at 22:05 on Nov 4, 2021 |
|
#
?
Nov 4, 2021 21:59
|
|
|
Cory Parsnipson posted:It's kind of glitchy though. When an emulator is running, the notification is unable to display. Also I think it drops screenshots every so often, which is really annoying. I fixed half of this issue. The screenshot dropping was happening if I take them too often. The filename used the date, hour, and minute to create a unique identifier, but if I took screenshots too fast, they would be named the same thing and overwrite the old ones. I changed the hour from 12 hr format to 24 hour format and added seconds. I think it should be impossible to take screenshots more than once every two seconds because you need to long press the start button. Though, I should add code to add a numbered suffix or something instead because if the functionality changes, for example if it was tied to short press or being done programmatically, it could definitely happen faster than once per second. https://github.com/CoryParsnipson/rpi-proto-1/commit/152ee266085e4e40e722b349b420cedc3b7b3a4b?diff=unified  So dumb...The notification display issue is something that might be harder to fix. Basically, I seem to be running out of memory when displaying a certain number of simultaneous png's. This isn't a set number that I can plan around, it just happens when some internal program (pngview or raspidmx or something lower) runs out of memory. pngview isn't meant to serve as the base for an entire UI so that's just the price paid for hacking together a program from existing stuff. ...I'll fix it later. Maybe...
|
|
#
?
Nov 5, 2021 08:33
|
|
|
Steam Deck?? More like... STEAM DICK AHAHA GOT EM Hey guys, my new dev kit arrived today. It's greyer than I thought and kind of smaller than I thought.  It's quite different to see the whole face plate in real life rather than on a screen or trying to imagine it. It's larger than you would think and also looks kind of boxy. I think that type of issue is less apparent for a device that's smaller and has less open space on the sides. I think I can rectify that with the PCB-utilizing version.  Not done yet. Almost there, though. I would like to get it done this week, but that depends on how many test fittings I need and the time it takes to print stuff.  The device right now Printer maintenance: replacing the part cooling fan I bought a part cooling fan from PartsBuilt and it came in on Friday. Shout out to insta from the 3d printing thread who told me that the fan was failing and I should replace it with a dual ball bearing version. When I opened the box--  wha...? Why do I keep getting toys and candy in my big manly machine parts boxes for big manly men???  I--... I... don't know what to say. Do you think she likes me? Y/n???? Gee wilikers, I ain't never met a girl who cared so much about my 3d printer. Hannah, maybe??!?!?  So anyway, I took my printer apart. There were instructions on the prusa website so it wasn't so bad. But having to take off the back plate of the extruder, and undo the cable bundle and wiring on the Einsy board just to replace a fan is kind of a pain.   This is part of what I was scared of... Though it wasn't too bad. I shouldn't be so afraid of taking stuff apart. I've been spoiled by Prusa too much. Everything you never knew you needed to know about muffin fans So reading up, it looks like there are "sleeve" fans and fans that use "dual ball bearings". The ball bearing ones are more expensive and slightly louder, but they should last for like 5 or 6 years while the sleeve ones only last about 1 year. The ball bearings in the fan are like ones you would find in a skateboard or light machinery. Good stuff. I found this video that tells you how to tell them apart. I wanted to be sure it was my fan going bad, was a sleeve fan, and that the new one was a ball bearing fan as expected.  As you can see on the original OEM fan part, the sleeve is made of some bronze looking material (probably bronze). Also I could jiggle the shaft with a pair of plyers. It has a lot of play just like they said in the video.  The new one from PartsBuilt3D is silver and barely moves when I try and grab it with the pliers. Excellent. I won't know for sure which is which unless I take them apart, but I am 95% confident I know what they are. Muffin fans, people!  Front Plate Enclosure I spent the day trying to dial in the first layer offset and printed the front plate the next day.   It came out fine. No oil spots and areas of intermittent cooling failure, so that's great! At least the fan replacement was worth it.  Unfortunately, I didn't double check my model enough and FreeCAD decided to move the bottom screw holes down into the bottom right corner when I wasn't looking. Very annoyed at that.  Also, I'm miffed that I had to give up my perfectly calibrated first layer adjustment because it's not perfect anymore and it shows a tiny bit. Thankfully, the text came out GREAT and I'm stoked that it did. I don't need you. I don't need no bottom screw holes. I don't need nobody! If Tony Montana drops his gameboy it blows into a million pieces. That's how Tony Montana does things. Fitting Front Plate  The frame fastening screw holes on the left side are trickier to get at than the right side ones. They're obscured by the USB hub cable and some other stuff. This one on the bottom basically has no visibility.  And once more for posterity. Looks great in this picture! The 0.5mm thick bezel actually looks a lot better than I thought it would. I assumed an adhesive LCD screen cover would be necessary for a premium feel but maybe that's not true, which would be great because the glue would make swapping face plates much more difficult.
|
|
#
?
Nov 8, 2021 23:29
|
|
|
I've been cleaning around the house because of some stuff that is happening this week. Unfortunately, that doesn't leave much time to be working on the back of the case. However, the table came in early and I spent a couple hours before bed putting it together. I was going to wait until after the back plate was done to work on the 3D printer "shrine" but I need to move it back and forth a few times this weekend and I think the table will make it easier. I need my own place.  I thought this would be better than an Ikea Lack but it still kind of sucks. It uses the same construction methods and parts of the legs are pretty loose. At least the particle board is an upgrade over hollow core stuff. I'm hoping it's a good enough start but I would have liked to be able to mod some retractable wheels on it. Gonna pick up a concrete paver and put the foam pads underneath it. I'm also thinking of getting a wooden dowel and printing tent pole style connectors and maybe making an enclosure using cling wrap or something. (I actually don't want to use an enclosure, I just want it to keep the dust out when the printer is off.) And thus is the start of my descent into madness.
|
|
#
?
Nov 11, 2021 10:02
|
|
|
HDMI port test plate I decided to start with what I think is the most complicated part (whyyyy), which is the part with the HDMI port. Getting everything lined up was a pain, but again, the assembly saves my rear end.   BOOM, baby! Also printing on the table seems good. It's not shaking. The noise is slightly louder in the general vicinity because it's being amplified by the table, but one room over is virtually identical to before. I think the ringing is worse, but I haven't gotten a concrete paver yet. I'm eager to get one and compare. I'm testing several things with this piece: * HDMI port hole size fit * Audio jack hole positioning * Print quality of vertical hole with chamfer * Depth requirement of back enclosure piece * Height fit for both the top and bottom sides of the back enclosure piece  The hole fits pretty good first time. Man I'm good.  The chamfering on the vertical audio jack hole is pretty good too, though I saw it was a bit elliptical and flat on the top.Both the audio jack and HDMI holes need to be moved upwards about 0.8 - 1mm. The depth of the back enclosure needs to be increased by 2 - 5 mm because it's pressing on the end of the wires of the RPi GPIO and flexing. And speaking of flexing, the bottom side-wall and back are much more flexible than I thought they would be. I either should up the infill or add some raised lines on them to make it stiffer.  Saving this for future reference...  I'm pressing this down because it's flexing due to the wires. I kind of wish I had the bottom screw holes now because I'm not sure if the case will fit together nicely without them now.  Also the HDMI cable fits. I thought I messed up by making the sloped section start too close to the hole, but the cable leaves quite a bit of room sticking out of the connector so everything fits fine. As you can see, the HDMI port is in a really weird spot, so hopefully these clearance issues won't be a problem for any new designs I make. Ok, so next I'm probably going to make a second HDMI test plate to try out all the changes I mentioned above and then start working on fitting the holes for all the other connectors on the top. After that, it's pretty much just fleshing out the sides and then going for the final print. Cory Parsnipson fucked around with this message at 23:28 on Nov 15, 2021 |
|
#
?
Nov 15, 2021 22:20
|
|
|
HDMI Port Test Plate 2 I went to bed in the middle of this 4.5 hour print and I have to say it's way more convenient when you don't have to worry about putting the filament back in the bag at the end. I'm at the last couple meters of my first roll and not having to worry about keeping the filament dry makes 3d printing 100% more convenient. I want a dry box with PTFE tubing now!   Here's the new piece with all the changes I needed to make. Also I added in a 2mm chamfer on the bottom edges that would be in the final print. This will match the edge chamfer on the front piece.  That did the trick. As you can see the front lip is flush with the edge when the connectors are in the holes. If you go back to the pictures of the first version, there's an 0.8mm gap there.  The sides are stiffer now with the raised lines. I might have overdone it actually. I could probably spread them out a bit and make them slightly smaller while keeping them the same strength. I don't think it's worth testing out right now, though. Unfortunately, this print warps imperceptibly, but since it's so tall it amplifies the imperfection until you get to the mating surface and the back edge is way off its mark.  If I push it together with my fingers, you can see that everything does fit together nicely, but not without some force keeping them stuck together. I'm hoping this will go away when the whole piece is printed and there's 4 walls keeping everything in check. If not, then there'll be an ugly gap on the bottom. I won't be able to fix that without printing another front cover and I don't feel like doing that. I'd rather just get it right in the next version, especially since that one is supposed to be smaller so I won't use as much time and material printing it out. (And besides, if the next one is thin enough, this won't even be a problem) Cory Parsnipson fucked around with this message at 22:06 on Nov 16, 2021 |
|
#
?
Nov 16, 2021 21:47
|
|
|
Fitting connector holes Man, this back plate is so big, there so many pieces I need to work on and some of them are boring as poo poo. This piece was kind of hard to get off the ground because it's mostly lining up holes, and I don't have STEP models of the components, so I have to take measurements in real life that aren't very accurate and there's a lot of trial and error. I can tell it's boring because I start daydreaming about how to solve problems for the next iteration while I'm doing this.  I pulled out this screw hole test plate I made a while back, because this is what I'm using on the back plate. It's a long 3.25 diameter hole going all the way down where it meets a 45 degree 1mm countersink into a M1.6 screw hole that goes through 2mm of solid material to mate the enclosure halves together. The dimensions on this aren't ideal because the screw holes were an afterthought and I'm working with the leftover space around the inner components. For instance, this column is only one 0.4mm nozzle width thick.   Aand I forgot that I want to give the edges a 2mm chamfer and the screw hole goes right through it. Looks cool, but complicates the printing. The slicer looks like it prints a couple layers in mid air, but I wanted to see how it would turn out so I just printed it like this.  Huh, not bad actually. I'm gonna just do this. But between the unsupported curved overhang and the single thickness wall, it's not something I'd want to purposely do again. On the 3DS, I noticed that the screw holes are inset more around 5-10 mm from the edge. I think that would be good to copy.  The inside of this test plate looks like some kind of D&D miniature of a factory or something. Heh. Lots of stringing here I think due to a combination of wetter filament and I bumped the temp up to 209 C from 207 C. Also I tried a different version of the reinforcement lines that are 1 mm thick, spread farther apart, and bigger fins in the corner. This seems stronger than the last one actually, mostly due to the huge-rear end fins in the corner. (And it looks better too!)  It fits pretty well. The USB ports are almost spot on. I should nudge the power button hole a bit to the right and 0.4mm downwards. The charging port is off by a whole 1-1.5mm, though the size is acceptable. drat. The screw hole clearance is too long for my screwdriver. I had to go and buy a 5 dollar screwdriver off Grainger just for the one temporary part!  Who said size doesn't matter??? Take this, participation award apologists! Oh and by the way, this marks the end of my very first roll of filament.  I had to do a mid-print filament change 80% of the way through. The line is visible. I bought a second roll of grey Prusa filament because I just had to finish this whole project in one color. Also, since the shipping for 2 rolls is the same as 1, I bought a roll of Gentlemen's grey for future prototyping parts. One day I will buy a real color. 
|
|
#
?
Nov 20, 2021 08:17
|
|
|
Buy a screwdriver?? Surely the most reasonable option would be to print one?
|
|
#
?
Nov 20, 2021 12:31
|
|
|
Slugworth posted:Buy a screwdriver?? Surely the most reasonable option would be to print one? gently caress!  This might be hard to print because it's a reallly long, thin print and the phillips head has really small details that have to stand up to relatively high forces. But now I want to try it and see what happens.  e.   I tried printing this upright, like a rocket, but the slicer wouldn't even print most of the tip. It would end in a nub just after where the phillips head fins started. I thought cutting it in half and adding holes for "biscuits" to go inside afterwards would be more doable...  Even this was kind of difficult due to first layer adhesion issues. You can see that the tip of the screwdriver is again laid down as a rounded nub. I decided to kill it after one layer because it didn't look like it would get that far.  So the thing is, I need the shaft to be less than 3.3mm in diameter and it has to be able to drive a M1.6 screw. This brings it into "precision" screwdriver territory, somewhere between a #0 or #00 size, and it looks like it's just too much for my poor little FDM printer with a 0.4mm nozzle.If you're curious about this type of thing, there's some *people* in the 3D printing, electronics, and CAD threads that can make this. I think normally they'd use a lathe and a CNC mill to make one out of some soft metal. If you don't want to buy heavy machines then a resin printer, "lost-wax-resin" casting molds, and a crucible/kiln sounds like it would do the trick. Or something like that I dunno. Cory Parsnipson fucked around with this message at 00:39 on Nov 21, 2021 |
|
#
?
Nov 20, 2021 19:21
|
|
|
Back Cover WIP  The back cover continues to drag on and on... I had to learn a few new things such as sweeping a shape along a curve and probably will need to learn how to send a linear pattern around in a loop too. I put in the connector holes, but now I feel like I still need another test print to determine how much chamfer I can put on the holes.  It's mostly done, but missing a lot of the stuff I need to put in the inside. I saved the most "controversial" features for last, so moving them around will break less stuff.  This thing looks like an alarm clock...
|
|
#
?
Nov 24, 2021 06:58
|
|
|
Cory Parsnipson posted:This thing looks like an alarm clock... Try to use the words "unique" and "custom-built" and "hand-designed and fabricated." "Bespoke," even. It rings better than "cobbled together because I have no idea what I'm doing and the next one will be better I promise  ." Speaking from experience.You're doing great, by the way. Do you still intend to make a version 2 using what you've learned? babyeatingpsychopath fucked around with this message at 00:59 on Nov 25, 2021 |
|
#
?
Nov 25, 2021 00:56
|
|
|
babyeatingpsychopath posted:Try to use the words "unique" and "custom-built" and "hand-designed and fabricated." "Bespoke," even. Yeah, i built this in a cave with a box of scraps I mean, the purpose of this was to playtest and figure out things that I like and dislike and also to serve as a baseline to move design elements around, so mission accomplished (or will be soon after I finish the back plate). My main regret is that it took slightly longer than a year to make this, but I had to learn so much so I guess it was worth it. babyeatingpsychopath posted:You're doing great, by the way. Do you still intend to make a version 2 using what you've learned? You bet. I think I'm going to go all the way to make a second one that I'd like to call the "real" one and then call it a thread. This doesn't mean I'll stop working on it after that but further development will be dependent on who is still interested in this (myself included). I'll re-assess when we get there. I'd like to get it done in less than a year, but the realities of what I need to accomplish and my past performance all suggest otherwise. All the things I need to do to go from this prototype to the "real" one are bigger, tougher, and harder than anything I've worked on so far. One of things that are bumming me out right now is that whenever I seem to overcome an obstacle, the ones behind it are bigger in scope and more complicated than ever. I'm trying to think of ways to speed things up, so I'm realistically reserved about estimates but still hopeful to see what happens from here on out. I'll write more about this after the back plate is done.
|
|
#
?
Nov 25, 2021 03:10
|
|
|
Cory Parsnipson posted:You bet. I think I'm going to go all the way to make a second one that I'd like to call the "real" one and then call it a thread. This doesn't mean I'll stop working on it after that but further development will be dependent on who is still interested in this (myself included). Great! I can't wait to hear your insights. I have thoughts about this as well. Mostly around battery and power stuff. You've done AMAZING work with the raspi and the UI.
|
|
#
?
Nov 25, 2021 04:18
|
|
|
babyeatingpsychopath posted:Great! I can't wait to hear your insights. I have thoughts about this as well. Mostly around battery and power stuff. You've done AMAZING work with the raspi and the UI. Ok, feel free to share after I do a small write-up after the case is put together. I'm curious to know what you think. The power system could use a lot of work. I think it's probably the deciding factor in whether or not the next version will be nice to use. I'm already feeling some pains related to the battery in the current one. Adding a Concrete Paver to the 3D Printing Shrine This is a common thing, apparently. This cheap concrete brick can supposedly dampen almost all printer noise and increase quality of the print. I went out and got one from my local Lowe's home improvement store (TM). Lemme tell you, I never thought something as innocuous seeming as carrying around one driveway brick would draw so many suspicious stares and disrespect from the local boomer population. It's a single driveway tile! What you want from me??  My very own personal concrete paver after washing with dish soap, sanding the edges down, and sun drying over two days.  I bought 20 dollars worth of EVA foam, though I read that you can buy the slightly more expensive sorbothane sheet and cut it into a 16x16 square for a lower profile damper.  Boom! Lemme tell you, I was pleased to say that everyone's accounts of its dampening properties are not exaggerated one bit. It's quieter than when it was on the floor, even. Previously, I could hear fast long travel moves through the wall and now I can't. Occasionally, I can hear a faint, low-frequency hum when I take my headphones off and that's it. I mainly bought the paver for reducing the ringing effects on print surfaces though.  For instance, the right most piece is the printer directly on the table with the table legs directly on vinyl flooring. The ringing is pretty noticeable--you can see big, visible ripples coming off the edge right there. In the middle is the 3d printer on the table with felt pads that I stuck on there. On the left is the 3d printer on a concrete paver + EVA blocks and felt pads on the feet. The main jump in quality comes from adding the felt pads on the feet of the table, though I suspect that without the felt pads and with the concrete paver I'd get identical results. The important part was that I had a flexible damper somewhere in the system. Having both keeps me covered for many situations but I don't see any noticeable difference in quality between the middle and the left pieces.  Here's the other side, but the story is the same. There's many ripples between the two support lines on the wall. The table with felt pads, with or without the concrete paver, have almost imperceptible differences where some places the concrete paver helps slightly and in other places the paver seems to make it slightly worse. I wouldn't worry about such a small difference, though, and with the paver, I get superior noise dampening. Cory Parsnipson fucked around with this message at 22:34 on Nov 26, 2021 |
|
#
?
Nov 26, 2021 22:14
|
|
|
The Last Test Plate I decided to make another test plate to figure out how tall I needed to make internal supports and to play around with different chamfer values on the side holes.   The good news is that the supports were measured out pretty well.  The bad is that none of the new values I tried worked that well. The problem is that the chamfered area leaves behind a portion of the wall that isn't in the chamfer and I wanted to see how far I could push it. Turns out it wasn't very far. The wall needs to be at least thicker than 0.45mm beyond the chamfer to prevent the slicer from telling the printer to print in mid-air. I miiight use the power switch hole settings on the left side and just clean it up because for that hole in particular, I need it to taper to a thin ring around the power button. The USB ports look like crap and have holes in the corner and the charging ports look terrible too.
|
|
#
?
Nov 26, 2021 22:18
|
|
|
The Final Print (maybe) I got the whole back cover modeled with all the greeble and stuff on the inside. There's columns that are 39mm tall to screw into holes I put in the second tier holder in the hope to make the back less flexible. There's also columns stuck to the top wall where the shoulder buttons are. They are there to press down the inner frame so when you press on the Start or Select button, the frame won't deform and make it a little harder to press them down. If you recall, way back when I made the controller holders, I cut corners on the start and select buttons and the frame stiffness wasn't that great in those areas.  TECHNOLOGY I feel like I must have spent something like 20 hours working away at this thing. Most of the time and effort spend can be attributed to closing loopholes and shortcuts that I made earlier in the design. Everything had to be carefully measured and tested to make sure the cover fits all the stuff I haphazardly placed to get everything to line up. Alright, let's see what the damage is--  34 HOURS?!?  34 hours. That's from Sunday all the way to Tuesday. Poggers. drat, and 48 meters of filament too. Better get started, I guess. So I scrubbed the build plate in preparation for The Long Print (TM) and set it off around noon today. And then I realized that it would have to go a whole night in this house that literally has a gaping hole to the outside cut in the living room wall which is covered loosely with a piece of wood (I poo poo you not), and I got anxious and creative at the same time...  This being November, it's getting kind of cold at night so I hope this will help smooth out the temperature changes. Screwdriver  Also the screwdriver came in early, so I tried it out and it actually fits, so that's good. The hold is very weak because the screw hole column is very thin. There's almost no material holding everything together. That makes me think that the attachment of the back cover will be precarious. I hope the screw holes in the middle of the back cover going into the second tier piece will add some strength to it.
|
|
#
?
Nov 28, 2021 05:04
|
|
|
Phase 1 Complete Whelp, here it is. I didn't forget that the left dpad button still doesn't work and I need to cut the led on the power supply to see if that increases my quiescent battery life. There's still a lot of small changes to make and data to gather. I'm thinking now that babyeatingpsycopath's patches might have shuffled the linker addresses or something and maybe that is how the left dpad button worked for a while.   The screws hold it on just fine and it feels mostly sturdy. The crease where the front and back halves meet is clean and almost all areas that I was worried would have flexing issues don't have any. I have no faith that it would stay in one piece if I dropped it on the floor, though. What is this, a cell phone from the eighties???  mfw I accidentally a whole game console  Ahahaha. Ahhh. Final Print Complete  The back cover came out just fine. Well, I mean there were some minor extrusion-related issues with thin walls and other design flaws, but it's mostly cosmetic. Funny story, actually, I slept terribly last night because I kept having one long nightmare that people were breaking into my house and loving with the printer. And it kept going in one long narrative even after I repeatedly woke up and went back to sleep! I got nothing to worry about. My brain needs to calm down.  NOITAZILAUSIV Cory Parsnipson posted:VISUALIZATION   Huh, well there was quite a bit of size creep. Here's the final product next to the original paper mock-up. Getting the face down to the correct size would be a feat, but would also be very nice in terms of useability. Step 158: Crack a cold one open with the boyz to celebrate  Oh man, this was quite a long endeavor. I think I'm going to take a (short!) break to do some smaller projects and finish stuff I've put off before coming back to do the mk. II. TO BE CONTINUED I want to write up a post-mortem of things I learned doing Phase 1. And also make a post about the direction of research/improvements I want to do next. Coming shortly.
|
|
#
?
Nov 29, 2021 11:24
|
|
|
Extremely cool! You did it!!
|
|
#
?
Nov 29, 2021 16:27
|
|
|
Seat Safety Switch posted:Extremely cool! You did it!! Thank you kindly, goon sire!!  Another round of guiness fo r Seat Safety Switch 
Cory Parsnipson fucked around with this message at 06:34 on Nov 30, 2021 |
|
#
?
Nov 29, 2021 22:03
|
|
|
Great work! I wholeheartedly and unreservedly applaud your efforts into getting it done! I am especially in awe of your ability to say "this isn't working and I'm fully done dicking with it even though I think I'm close to getting it working" and then moving on to something easier to get it working.
|
|
#
?
Dec 1, 2021 06:29
|
|
|
I don't get why you didn't make it thinner. It seems like it would be better if it was thinner. No, this is impressive as poo poo and has been a wild ride from "This seems like an ambitious idea for a project that won't finish" to "Holy poo poo, he's making it".
|
|
#
?
Dec 1, 2021 19:40
|
|
|
babyeatingpsychopath posted:Great work! I wholeheartedly and unreservedly applaud your efforts into getting it done! I am especially in awe of your ability to say "this isn't working and I'm fully done dicking with it even though I think I'm close to getting it working" and then moving on to something easier to get it working. Thanks, I appreciate your continued support. The people in this thread have really helped me out when times got tough. I completely underestimated how hard it is to pull out of the sunk-cost fallacy sink pit but I find having a bit of dunning-kruger helps. Something I also realized is that working on different parts simultaneously can really help when things get monotonous to change things up and boost motivation. Also, I don't know what I was thinking before. Totally interested in hearing what you have to say about the power system, ignore what I said before about waiting for me to write something up. Slugworth posted:I don't get why you didn't make it thinner. It seems like it would be better if it was thinner. Uh excuse me, please don't fat shame my Raspberry Pi  It has been completely wild. I thought it would be a short side project but I ended up having to fill a lot of gaps in my knowledge and now I think I'm getting sucked down a rabbit hole. I think the best part is that I haven't really had this much fun doing a tech related project in a really long time. It's much more engaging to me than stuff I'd normally do for work *coughcodingandspreadsheetscough* and really satisfying to be able to see and touch what I'm working on. Also thanks, dude. I'm counting on you to tell me if this project gets to the point where you'd want to have one. 
|
|
#
?
Dec 1, 2021 21:16
|
|
|
Cory Parsnipson posted:Also thanks, dude. I'm counting on you to tell me if this project gets to the point where you'd want to have one.
|
|
#
?
Dec 2, 2021 02:43
|
|
|
Slugworth posted:Y'all got Breath Of The Wild 2 on that thing?? Always have 
|
|
#
?
Dec 2, 2021 05:03
|
|
|
drat, that's so exciting that I'd buy 2, but I just don't have the space in my house for a second one.
|
|
#
?
Dec 2, 2021 05:24
|
|
|
Slugworth posted:drat, that's so exciting that I'd buy 2, but I just don't have the space in my house for a second one. 
|
|
#
?
Dec 2, 2021 08:01
|
|
|
Great work, the grey brick look makes it seem like a retro-future 80s stab at the GP2X. The internal support frame integration was especially impressive, halfway to a watch. Throw on some mirror shades and play some Tetris.
|
|
#
?
Dec 5, 2021 15:24
|
|
|
Decoy Badger posted:Great work, the grey brick look makes it seem like a retro-future 80s stab at the GP2X. The internal support frame integration was especially impressive, halfway to a watch. Throw on some mirror shades and play some Tetris. Thanks!! I think the inner frame would be really convenient if I could print it in one piece and for that to happen it would mean that the whole thing should be 1 level. The main purpose is to keep all the pieces attached to each other while making it extremely easy to replace the outer casing. I think sometimes that it might not be necessary if I end up using one big PCB in the center, so that's something I have to explore next. Also, wow that's interesting, I didn't know about the GP2X. I can't believe how long ago they did that. Also looks great, those guys had good taste.  Recently I saw some people play Tetris effect on the Switch and it looks really fun. I like how they worked in multiplayer into it. Now to figure out what kinds of Tetris will run on this thing...
|
|
#
?
Dec 5, 2021 22:16
|
|
|
That's an impressive beast you have there. I'm gone back to beating my head against the wall on a similar project. Powering the thing still seems like the biggest issue. I'm really leaning towards a 2s2p setup with protected 18650s, and just have the batteries swappable like I'm using a Game Gear or something. I already have an external charger for them and it would eliminate a lot of headaches. I think all I would need would be the pololu buck/boost board that I already bought when I saw you grabbed one. I'm hoping for my uses it wouldn't be too much of a pain in the rear end.
|
|
#
?
Dec 6, 2021 15:21
|
|
|
A Proper Uppercut posted:That's an impressive beast you have there. Hey man, good to hear from you again! Yeah, the power supply was a huge pain. When I read about other people's projects usually it's a single line mention of dropping in a charger or something. I don't know what I did wrong. At least 2s would be good I think because even at their lowest charge they'd be at 6V, giving you enough room to power 5V with a good LDO. When mine gets low on battery it just browns out. I'm gonna try and find a power supply design textbook or something. I think it'll help.
|
|
#
?
Dec 6, 2021 22:31
|
|
|
Cory Parsnipson posted:Hey man, good to hear from you again! Yeah, the power supply was a huge pain. When I read about other people's projects usually it's a single line mention of dropping in a charger or something. I don't know what I did wrong. Something I think I've seen on more than once is people just pull the guts out of a power bank. They have all the circuitry for charging and balancing the cells built in. I tried asking around to see if that kind of thing would be suitable for this but I couldn't seem to find a straight answer.
|
|
#
?
Dec 6, 2021 23:44
|
|
|
A Proper Uppercut posted:Something I think I've seen on more than once is people just pull the guts out of a power bank. They have all the circuitry for charging and balancing the cells built in. I tried asking around to see if that kind of thing would be suitable for this but I couldn't seem to find a straight answer. Now that you bring it up again, my initial reaction was to say that I don't know enough about power banks to re-use one in my designs, but I've been thinking it over some more and that's not true. Given what I've learned the first time around, the feature-set of a power bank does line up exactly to our power supply requirements, as you've said. I think if you want to attach a 2s+2p+ cell to a power bank, you will need a separate balancing module, though I guess a way to work around this is to find a power bank with multiple ports and just hook one cell to each port to get extra capacity. (Don't quote me on that, though. I'd look into this more if I was going to do this seriously). And then the other stuff that I bolted on, like the soft power switch can still go on unmodified. The only thing we'd lose is the integrated fuel gauge, which is kind of a lot to ask for anyway. I'll keep a look out for power banks with the correct power rating, that are compact, and look easy to open.
|
|
#
?
Dec 7, 2021 09:30
|
|
|
Phase 1 Post Mortem This post is about various topics related to the process of getting here that I want to ruminate about. Pre-emptive, “sorry” if it sounds like stuff I talked about a million times before. Reigning in the scope and timelines I have a huuuuge problem with picking manageable projects. This one was no different. Either I know everything that needs to be done and I get bored, or the idea is exciting but filled with things I need to discover and I have no idea how much work it is. Think of the smallest possible idea that still interests you and then halve it. And then take that idea and halve it again, and then you’ll probably get somewhere in the vicinity of a realistic goal.  I recall saying at the beginning of this thread that I was pretty far into the project, having already hooked up an LCD screen and speakers to the raspberry pi. I was hoping maybe I could get Phase 1 done in like, 3 months.  \Well, that aged pretty poorly. Looking back it makes me cringe so hard, bro. I had no idea what I was getting into. One of the things I’m pretty self conscious about is how slowly this project is coming along. I spent a whole year doing this drat thing and, to be honest, it’s pretty basic. I think an electronics project of this level (i.e. a raspberry pi with a battery, a screen, and some buttons attached to it) usually seems to take other people about a month. I know it’s not really healthy to compare yourself to other people, but I’ve been getting into the Hackaday blog feed and reading about interesting projects. When I do this, I like to dissect how the creator did things and read what the author writes about the creator’s particular motivation behind their design decisions. I look with envy when I come across an article or video with a fun, clever idea that composes basic components in a simple but really effective way. On the other hand, I learned a bunch of new skills like 3d printing, CAD modeling, and some circuit designs and that takes a lot of time. I also complicated things by deciding to make an inner frame and trying to fit as many features as I could into the prototype. But there’s one thing that I can’t ignore: Even though I might be having fun learning, I’m also impatient and I don’t want it to take years! Speeding up has to come from somewhere, whether it’s compromising on features, working smarter, or just working longer. Not sure what I’m going to do about this. Smaller development cycles When I’m making something, I usually think of a “loop”. The loop begins with an idea, some small tests if necessary, then proceeds with me making some changes, and ends with a result. If the result is bad (i.e. the 3d printed part doesn’t fit, or the code is buggy), then I go around the loop again. I wasn’t used to doing this with hardware. I really noticed this pain when I was wiring together all the buttons in the controller to the Arduino protoboard. Everything was permanently soldered in and the wires wasted a toooooon of space. This made it really hard to make changes. I should have used some right angle male pin header on the protoboard and dupont connectors on the wires to make the buttons detachable from the Arduino. Unfortunately, I didn’t think to do this or have any crimping tools until after I finished the power supply, so I have to chalk this up to the normal learning process. I’d like to make more stuff in a “modular” fashion so that I can experiment without feeling like it will take too much effort to disassemble what I already have. I think this should be a consideration for all designs right from the beginning. The cool thing about open source projects is that you can take them apart to mod or repair them more easily than a regular consumer product. Allegedly. Sunk cost fallacy I can’t do a post-mortem without mentioning the whole “cardboard debacle”. Initially I was going to stuff this into a…  cardboard box …and deal with the form factor later. Then I was wondering how fancy I could make the cardboard box and soon I was obsessed with trying to make a 1:1 model out of paper, copper tape, and corrugated card stock. I feel like going down this rabbit hole added an extra 2-3 months onto the total development time. I received kind of a lot of push back in this thread for doing this but I went ahead because it seemed cool and why not? There wasn’t any real reason. But okay, looking back on this, it wasn’t the right decision.  Baka! I- …I was wrong, okay? Here’s why cardboard didn’t work Cardboard didn’t end up becoming a money or time saving shortcut, nor was it a good option to make the project more accessible to people who want to replicate it. I didn’t have a chance in hell to make a Google Cardboard and I’m sure the people who did didn’t just pull it out of their asses the first try. I wanted to use conductive ink to draw a giant circuit on a piece of paper and then glue that to a cardboard backing holding all the electronic components in place but the real world implementation did not live up to the original idea. The conductive ink isn’t very conductive. I couldn’t get the ink pen to make even a 10 cm long trace that would pass a signal through it. Not only that, but fastening wires/pins to contact the ink circuit was so inconvenient I couldn’t even figure out a solution to this. To add insult to injury, the ink traces I made ended up oxidizing after a couple months and don’t even work anymore. Anyone with some electronics experience probably could have told you these things would happen, but alas, I am but a monké.  I ended up replacing the ink with copper tape manually cut into thin traces with a hobby knife. I solved the interfacing issue by using a soldering iron to bind wires to the copper tape. However, I still gave up because making the traces was so time consuming and a huge pain in the rear end to make. I suppose you could try and develop techniques to make paper circuits easier. For instance, you could get a Cricut machine to cut copper traces and cardboard shapes out and then solder everything together. But then that’d mean you would need to invest in a $200+ machine and you’d still have the severe lifespan and structural issues of the cardboard. And if the goal is to make things more accessible and save time/money, what’s the point in that? Speaking of accessibility and saving time/money, unlike software or 3d printed parts, you can’t just send someone a bunch of files and have them push a button to recreate what you made. The cardboard has to be cut out and measured and then glued together like some sort of monstrous artisanal kindergarten arts and crafts project. This entire operation becomes a bespoke creation. If you want to make something out of cardboard, the research to figure out how to make it happen would add so much more work that it becomes a defining goal of the project. And I think even then the final product would be a work of art and a novelty that is meant for being displayed behind a humidity controlled case in a museum instead of being used for its intended purpose. What do people want There seems to be very little overlap between a) people who like to play video games and b) people who like putting things together or taking them apart. There’s nothing wrong with that, but it means I need to rethink my assumptions about what is considered “simple” to (dis)assemble. I’ve learned that soldering is a high barrier to entry and going from something you can snap fit and build with screws to something that requires even a little bit of soldering is really intimidating. Obviously some of that is from the fact that you need to invest in the proper equipment but I’ve also seen a lot of hesitation when it comes to the manual procedure of doing the soldering. 3D printing is another high barrier to entry, but depending on the person, it could be more intimidating than soldering, or, if they already have a printer, could be the reason why someone takes on a project in the first place. Ok done I’m still taking a break. Mainly I’ve been working on getting my 3d printing station to be self-contained. I also want to figure out how to make an enclosure for it and then figure out how to print with PETG filament. Maybe do a series of smaller projects as well before coming back for version 2.
|
|
#
?
Dec 10, 2021 04:21
|
|
|
|
| # ? Apr 25, 2024 16:17 |
|
|
You absolutely should not be beating yourself up over mistakes and things taking a while. I'm a firm believer that any mistake is a learning experience as long as you analyze why it was a mistake in the first place. You gained a lot of niche knowledge that will definitely help you in future projects, and the fact that you got it made should give you confidence that you could do similar things in the future. Do you have any idea how many projects and hobbies I have left abandoned around my house and in the trash and/or sold? You should be proud of what you did. Also, do you mind me posting in this thread about my loving around with my portable build? I've decided to go in a slightly different direction with it.
|
|
#
?
Dec 10, 2021 11:19
|
|