All this talk about droopy knobs, broken woodies, shafts, nipples and bottoms is enough to make Kenneth Williams blush. Brilliant work Lee, and good to hear you're recovering 🙂
Have you ever thought about getting in talks with Albert over at AtariAge to see if you can get this mass manufactured? This is GLORIOUS, and an excellent alternative to the OG controller!
Lee you're an absolute genius of an engineer that was an amazing build! I could see the plastic flex being quite a challenge to overcome I hope you find something for it!
I also use PTFE lubricant when I have plastic on plastic contact for moving parts that need to be smooth. It acts like a grease, but doesn’t react with the plastic, doesn’t have a strong smell and the stuff I use if food grade.
The quickshot joysticks had a tightly coiled spring at the bottom of the stick pointing downward into a circular container in the bottom of the case. That might help with centralisation problem.
For a centering option, try making a ring that attaches to the center hole, then add 10ish fingers that stick out at 20 degs angles, making contact 360 degs. so when you lean it, the set of fingers slightly bend. Take the tamper ring off a milk jug or soda bottle to get a general idea.
For a smoother top, maybe try using adaptive layers, which will do smaller layer heights near the top, but not slow down your overall print like it would if you printed at a lower layer height.
Great job! I like the design very much! As for the joystick issue, how about printing a sleeve with TPU that goes between joystick opening and the knob? Tension should keep it centered and since TPU is flexible, it should allow it to be pushed in any direction. It will wear over time but worth a try.
Modern 3D modeling tools (even open source ones like Blender) offer neat ways of merging spherical and elliptical elements into planar ones that automatically clean up the uneven polygons. IMO it's really worth investing some time into learning hard surface modeling with Blender, it's got a bit of an odd interface to get used to, but when you do it becomes second nature.
Amazing engineering there Lee, so much to figure out there. I always think joysticks are deceptively simple but there's all these little details that need to work to have a fun experience. Completely agree it looks AMAZING! Only when side by side could I really tell the differences between yours and the original, the look is very faithful. Super practical and authentic way to play Atari games that need paddle controls. These things take time, I think it's best not to rush videos out when you're working on something cool like this and also considering your operation and recovery too. Bravo and hope people cleverer than me can help with the centring problem - will be fun to see part 2 when you've tried some of the ideas. Cheers!
A thought I had about improving based off the original design would to be embed thin metal clips to act as and renforce the spring mechanism of the tabs that press the switches. This could provide strength to the plastic preventing breakage, and solve the centering problem. Kind of like adding tie bars to concrete.
I was looking at spring clips that can be soldered to the PCB and I found a few, but I think they are for terminals, connecting grounds and battery's so not sure how they would do with the repeating joystick movements.
@@leesmithsworkshop I was thinking more of a piece to repair the plastic of the original model. I think maybe a couple pieces of nitinol wire embedded in each "flap" with a 90 degree bend into the column might do the trick. I have the Sears video arcade II, and just thinking outside the box to stiffen it up. I access to a few printers, I'll see what I come up with. You did an amazing job with what you came up with, and have inspired me.
@@leesmithsworkshop The problem you observed with your mockup also occurs on the original design. They become loose, and you have to press harder to get a response, until eventually it's too loose to work
For your centering problem, you could try 8 round flat neodymium magnets arranged in 4 pairs instead of the springs. Four of the magnets would be glued to the part you show at 26:10 . The other four would be glued to the case. They would be set up to repel each other. You would just have to try it to see if it felt right. It would all depend on the strength of the magnets and the distances between them. Nice video and the 3d printed parts look great.
I believe textured plastic is usually achieved by sandblasting the mould; they can polish a part of the mould to give it lacquer type finish and sandblast part of it to give it that textured finish. There is also a process of flocking to give items a different surface, but i believe you'd have a sticky mess on your hands by now if that had been used. The original joystick subassembly is probably moulded from PA6. You can have yours SLS printed from PA12 and then condition it in moist atmosphere to reach final pliability; as an extra advantage PA are self lubricating materials, low surface friction. You put it in a ziploc bag with half a teaspoon of water to condition it and wait a day or two and you'll have a slick and springy part; however it will also expand a little. It shouldn't break conditioned like that, and it should hold onto conditioning for about 20 years. Some alternatives i can think of for spring wings, you can use zipties as inserts that you somehow attach to the print, perhaps just with epoxy, with the moulded in teeth of the ziptie providing the key that epoxy can hold onto by interlocking, which is needed since it won't actually form a permanent surface bond with PA zipties. Other thing i can think of is spring steel thickness gauge foil (maybe 0.2mm or so), but i don't have a sample of the material at hand so i don't know if this is right. Odds are the original controller, as designed, has these wings a little more flexible than they are now, so they're now more susceptible to damage, due to loss of conditioning.
I was watching a video about injection molding and they mentioned the spray method so that is my only reference. The main problem with testing and fitting parts is the cost to have someone print parts in pa12 while I work iterations and it's just not worth the cost involved on a small scale. same for the spring steel you mention, i just don't have these things handy to test.
@@leesmithsworkshop GF variant is not for this purpose. It's for when you need extra stiff parts where you want as little flex in there as if it was cast metal, and also need to eliminate creep. Ultimate strength is of course lower than metals but so is weight and manufacturing cost at small scale. It can be a little abrasive as well, so that's not ideal. Very convenient sponsor for this sort of project; indeed i have something in mind and it's great that several SLS/MJF companies can cover actual small scale manufacturing of a custom part with this sort of material where it makes sense anywhere between 20 items being made once and to potentially hundreds a month. As a sponsor they might just bridge the cost of iteration for you and make it sensible as a one-off as well, after all it is possibly a perfect showcase of something a customer might want to do at actual scale. You can potentially FDM print PA at home as well, but i don't have experience with it. From what i've heard and seen, it's a bit of a chore due to just how hygroscopic it is. Also i don't know if PETG printed wings would necessarily break, they might just be fine. The material is also kind of low friction, so that's nice, but not quite as silent as PA.
@@SianaGearz I am not planning to sell these so I can't go back and ask for test parts especially when my videos get low views. At the moment I have an idea and testing some springs in pla, asa and petg, but they are printed on the side as separate objects and will make a new central part that can fit into.
@@leesmithsworkshop Ah don't sell yourself short. You never quite know how the marketing budget and accounting of these companies works out, no point in making decisions like that for them. There might be more use out of collab than straight up one way advertisement, they might put your vid to use as a showcase, advertising your vids in the process, and they're good vids, you explain and showcase things very nicely, there's value to that. They're likely to have excess capacity on slow days that they're looking to put to use which doesn't actually result in any substantial costs to them. They might be looking to host things like that as "Shared Projects", letting people order kits from them, this way generating sales as well. They might be willing to take some gambles with less popular creators knowing that perhaps a handful of these projects might just pop off and create a sufficient revenue to cover all of them, and you shouldn't think that this project has no chance to prove popular - lots of people have an Atari console, can put this project to use, and may be inherently curious. And it's specifically thanks to them supporting the classic computer scene in breadth that this company enjoys great reputation in this community, that's an incredible sort of network effect and image capital that transcends immediate monetary value.
not going to lie the past month was rough, I had an operation to have my polyps removed again and found out I am allergic to penicillin. everything is improving, but after 3 weeks my hands still get itchy. Still it's good to have this video finished and I am working on the next.
Honestly, I'm impressed with how good you are with Fusion360. I feel i could learn a lot if you do some tutorials on how you use it. If you ever wanted to create a series on it.
I think you could solve the droop by either moving your current microswitches over a bit to change the level point on the switch Or go for some more heavy duty switches with stronger internal springs? The other more advanced idea I had was to create a ring with 4 prongs that poke in diagonally inwards towards your joystick assembly and then you could have 4 springs mounted on the prongs pushing back against the joystick assembly And the springs should fit between the micro switches?
I tried springs before, but I think I just did it totally wrong, the springs held the joystick in place, but also pushed it up so it added friction when it moved. I was talking to someone else and we came to the idea like yours to try it like a trampoline .
This is brilliant! I have an original one of these I picked up at a garage sale years ago for $2. Still works (after a good cleaning of course). But I'd be interested in seeing the new and improved one. How feasible would it be to make the two buttons independent of each other for full Atari 7800 compatibility? Not sure if there are enough pins on the DB9 connector for this though.
I have this same console with several of these unique joysticks, but unfortunately most of them don't work well. I did manage to create a part to fix the joystick in Tinkercad. Still a work in progress though.
I saw your reply, but it's not showing up did you delete it ? I think the material would have to be nylon of some sort to have the flex it would need, nothing I know would work if you copied it as the original and just break.
All this talk about droopy knobs, broken woodies, shafts, nipples and bottoms is enough to make Kenneth Williams blush. Brilliant work Lee, and good to hear you're recovering 🙂
all totally innocent I swear !!!!!! thanks it was a rough month, but I had my checkup yesterday and they are happy with my recovery.
@@leesmithsworkshop excellent news
I've just seen Rees walk past my house. He's carrying a VCS Woody and muttering about sticking it somewhere. I'd watch out.
right up !!!!!
Have you ever thought about getting in talks with Albert over at AtariAge to see if you can get this mass manufactured? This is GLORIOUS, and an excellent alternative to the OG controller!
Lee you're an absolute genius of an engineer that was an amazing build! I could see the plastic flex being quite a challenge to overcome I hope you find something for it!
I am Pro AM haha no even close to Genius . Lots of good idea in the comments and on twitter so I am sure I will find a good solution.
I also use PTFE lubricant when I have plastic on plastic contact for moving parts that need to be smooth. It acts like a grease, but doesn’t react with the plastic, doesn’t have a strong smell and the stuff I use if food grade.
The quickshot joysticks had a tightly coiled spring at the bottom of the stick pointing downward into a circular container in the bottom of the case. That might help with centralisation problem.
For a centering option, try making a ring that attaches to the center hole, then add 10ish fingers that stick out at 20 degs angles, making contact 360 degs. so when you lean it, the set of fingers slightly bend. Take the tamper ring off a milk jug or soda bottle to get a general idea.
it's all sorted and I will be making part two soon.
Amazing result. that sparkle black should be used on All things. it suits it so well
I amusing it for the mechtrum case too, it just looks so good.
Kudos bro! You really keep the ATARI alive :)
Injection moulds can be acid etched to achieve a textured finish but it is longer lasting to use spark erosion.
Thanks, i just know it's a process added to the metal after.
For a smoother top, maybe try using adaptive layers, which will do smaller layer heights near the top, but not slow down your overall print like it would if you printed at a lower layer height.
All I had to do was sand the ASA and it would be perfect, but I wanted to say something good about the resin prints from PCBWAY.
Great craftsmanship and ingenuity!
Great job! I like the design very much!
As for the joystick issue, how about printing a sleeve with TPU that goes between joystick opening and the knob?
Tension should keep it centered and since TPU is flexible, it should allow it to be pushed in any direction.
It will wear over time but worth a try.
i had thought about that, just need to find some time to play with the TPU, but work must come first.
Modern 3D modeling tools (even open source ones like Blender) offer neat ways of merging spherical and elliptical elements into planar ones that automatically clean up the uneven polygons. IMO it's really worth investing some time into learning hard surface modeling with Blender, it's got a bit of an odd interface to get used to, but when you do it becomes second nature.
I tried to learn onshape, but I always get frustrated when I can't work something out and I just go back to 123 design and do it it an instant.
Amazing engineering there Lee, so much to figure out there. I always think joysticks are deceptively simple but there's all these little details that need to work to have a fun experience. Completely agree it looks AMAZING! Only when side by side could I really tell the differences between yours and the original, the look is very faithful. Super practical and authentic way to play Atari games that need paddle controls. These things take time, I think it's best not to rush videos out when you're working on something cool like this and also considering your operation and recovery too. Bravo and hope people cleverer than me can help with the centring problem - will be fun to see part 2 when you've tried some of the ideas. Cheers!
you can make something look fancy , but if it is not enjoyable to use it's a fail.
A thought I had about improving based off the original design would to be embed thin metal clips to act as and renforce the spring mechanism of the tabs that press the switches. This could provide strength to the plastic preventing breakage, and solve the centering problem. Kind of like adding tie bars to concrete.
I was looking at spring clips that can be soldered to the PCB and I found a few, but I think they are for terminals, connecting grounds and battery's so not sure how they would do with the repeating joystick movements.
@@leesmithsworkshop I was thinking more of a piece to repair the plastic of the original model. I think maybe a couple pieces of nitinol wire embedded in each "flap" with a 90 degree bend into the column might do the trick. I have the Sears video arcade II, and just thinking outside the box to stiffen it up. I access to a few printers, I'll see what I come up with. You did an amazing job with what you came up with, and have inspired me.
@@davidmalia9057 are they known for damage on that part ?
@@leesmithsworkshop The problem you observed with your mockup also occurs on the original design. They become loose, and you have to press harder to get a response, until eventually it's too loose to work
For your centering problem, you could try 8 round flat neodymium magnets arranged in 4 pairs instead of the springs. Four of the magnets would be glued to the part you show at 26:10 . The other four would be glued to the case. They would be set up to repel each other. You would just have to try it to see if it felt right. It would all depend on the strength of the magnets and the distances between them. Nice video and the 3d printed parts look great.
Very good! I have played around with small cylindrical neodymium magnets and yeah they feel like they might just be it.
magnets is not a bad idea so I will test this.
I believe textured plastic is usually achieved by sandblasting the mould; they can polish a part of the mould to give it lacquer type finish and sandblast part of it to give it that textured finish. There is also a process of flocking to give items a different surface, but i believe you'd have a sticky mess on your hands by now if that had been used.
The original joystick subassembly is probably moulded from PA6. You can have yours SLS printed from PA12 and then condition it in moist atmosphere to reach final pliability; as an extra advantage PA are self lubricating materials, low surface friction. You put it in a ziploc bag with half a teaspoon of water to condition it and wait a day or two and you'll have a slick and springy part; however it will also expand a little. It shouldn't break conditioned like that, and it should hold onto conditioning for about 20 years. Some alternatives i can think of for spring wings, you can use zipties as inserts that you somehow attach to the print, perhaps just with epoxy, with the moulded in teeth of the ziptie providing the key that epoxy can hold onto by interlocking, which is needed since it won't actually form a permanent surface bond with PA zipties. Other thing i can think of is spring steel thickness gauge foil (maybe 0.2mm or so), but i don't have a sample of the material at hand so i don't know if this is right.
Odds are the original controller, as designed, has these wings a little more flexible than they are now, so they're now more susceptible to damage, due to loss of conditioning.
I was watching a video about injection molding and they mentioned the spray method so that is my only reference. The main problem with testing and fitting parts is the cost to have someone print parts in pa12 while I work iterations and it's just not worth the cost involved on a small scale. same for the spring steel you mention, i just don't have these things handy to test.
PCBWAY do have PA12 and PA12-GF as options, I wonder if I can ask them to print a test item and check it for me.
@@leesmithsworkshop GF variant is not for this purpose. It's for when you need extra stiff parts where you want as little flex in there as if it was cast metal, and also need to eliminate creep. Ultimate strength is of course lower than metals but so is weight and manufacturing cost at small scale. It can be a little abrasive as well, so that's not ideal.
Very convenient sponsor for this sort of project; indeed i have something in mind and it's great that several SLS/MJF companies can cover actual small scale manufacturing of a custom part with this sort of material where it makes sense anywhere between 20 items being made once and to potentially hundreds a month. As a sponsor they might just bridge the cost of iteration for you and make it sensible as a one-off as well, after all it is possibly a perfect showcase of something a customer might want to do at actual scale.
You can potentially FDM print PA at home as well, but i don't have experience with it. From what i've heard and seen, it's a bit of a chore due to just how hygroscopic it is.
Also i don't know if PETG printed wings would necessarily break, they might just be fine. The material is also kind of low friction, so that's nice, but not quite as silent as PA.
@@SianaGearz I am not planning to sell these so I can't go back and ask for test parts especially when my videos get low views. At the moment I have an idea and testing some springs in pla, asa and petg, but they are printed on the side as separate objects and will make a new central part that can fit into.
@@leesmithsworkshop Ah don't sell yourself short. You never quite know how the marketing budget and accounting of these companies works out, no point in making decisions like that for them. There might be more use out of collab than straight up one way advertisement, they might put your vid to use as a showcase, advertising your vids in the process, and they're good vids, you explain and showcase things very nicely, there's value to that. They're likely to have excess capacity on slow days that they're looking to put to use which doesn't actually result in any substantial costs to them. They might be looking to host things like that as "Shared Projects", letting people order kits from them, this way generating sales as well. They might be willing to take some gambles with less popular creators knowing that perhaps a handful of these projects might just pop off and create a sufficient revenue to cover all of them, and you shouldn't think that this project has no chance to prove popular - lots of people have an Atari console, can put this project to use, and may be inherently curious. And it's specifically thanks to them supporting the classic computer scene in breadth that this company enjoys great reputation in this community, that's an incredible sort of network effect and image capital that transcends immediate monetary value.
Fascinating AND entertaining. Well done, sir!
thank you for watching and just being awesome
So cool and inspiring to watch your process.
hey lee, nice to see you back, hope you are well..
these would be great for playing arkanoid on it on the mister
not going to lie the past month was rough, I had an operation to have my polyps removed again and found out I am allergic to penicillin. everything is improving, but after 3 weeks my hands still get itchy. Still it's good to have this video finished and I am working on the next.
@@leesmithsworkshop take it easy lee mate.. don't go pushing yourself too hard.
Honestly, I'm impressed with how good you are with Fusion360. I feel i could learn a lot if you do some tutorials on how you use it. If you ever wanted to create a series on it.
it's not fusion I actually don't know how to use it. This is he older program 123design.
Well done on the video!
Thanks for watching :)
I think you could solve the droop by either moving your current microswitches over a bit to change the level point on the switch Or go for some more heavy duty switches with stronger internal springs? The other more advanced idea I had was to create a ring with 4 prongs that poke in diagonally inwards towards your joystick assembly and then you could have 4 springs mounted on the prongs pushing back against the joystick assembly And the springs should fit between the micro switches?
take a look at my recent community post, I have one solution that I have tested and worked so just waiting for some springs to try out.
I already bought the strongest sub micro switches I could find and it was not strong enough.
I wonder if you could mount 4 horizontal, opposing springs. They can’t be too strong obviously. I’d probably start with springs from a Biro.
I tried springs before, but I think I just did it totally wrong, the springs held the joystick in place, but also pushed it up so it added friction when it moved. I was talking to someone else and we came to the idea like yours to try it like a trampoline .
This is brilliant! I have an original one of these I picked up at a garage sale years ago for $2. Still works (after a good cleaning of course). But I'd be interested in seeing the new and improved one. How feasible would it be to make the two buttons independent of each other for full Atari 7800 compatibility? Not sure if there are enough pins on the DB9 connector for this though.
I think it would need a 2nd switch for 1 - 2 buttons depending on how the 7800 is wired.
I have this same console with several of these unique joysticks, but unfortunately most of them don't work well. I did manage to create a part to fix the joystick in Tinkercad. Still a work in progress though.
people have said the pots need some deoxit and that the flappy wing parts I mentioned do break over time too. What are your faults ?
I saw your reply, but it's not showing up did you delete it ? I think the material would have to be nylon of some sort to have the flex it would need, nothing I know would work if you copied it as the original and just break.
@collingall of Laser Bear Industries maybe you best bet for help. I wonder if a much higher force switches would help.
I was not having much luck finding them that are small enough to fit inside.
@@leesmithsworkshop What about low profile KB switches or OMRON Mouse Micro Switch both come in multiple Nuton levels
@@leesmithsworkshop Also what about Seimitsu LS-58, LS-58-01 springs?
@@flibble666 what about them ?
@@leesmithsworkshop would they work to keep the stick central
Have you seen the gemini atari joystick?
I just did a search, is that the one with the yellow button ?
Oi!
Oi saveloy