Worklog The Wii SPii

[StonedEdge]

Have you tried original gman's pic16 programming codes v2, because you change some codes in the programming, it looks like BQ goes to shipping mode when changer plugs in and you also can try 5v charger without type c instead of 12v

I already tried all of the above, including removing the mosfets and tried to charge it with a 12V 1A AC adapter, but the same shutdown was happening. I appreciate all of your advice though. I double checked all of what you mentioned but I don’t think that is the issue.

Hopefully the mods can move this back to a normal worklog soon, but I have been reworking the board entirely. As you can see, there is a lot more breathing room for all of the components thanks to the TPS82085 regulator modules, which have an inbuilt inductor module. They work with two resistors and two caps, which makes them an ideal solution for my board.

I also have a call with Dmcke5 this weekend, who may be helping me with some design tips on how to get the SPii potentially machined and anodized! I know I failed for the 2020 contest (sorry everyone) but I want to deliver something special for this portable. Since I now have more time, I plan on reworking the shoulder buttons and making some other refinements too. The first PCB was a disaster, but I plan on taking my time with all of the polygons, routing, via placement and everything else for this new Rev 1.3 PCB. Watch this space!

New V Old (very old): Battery Management/Regulator Layout

 

[thedrew]

What you have achieved is amazing in itself. I'm sorry things didn't work out exactly as you planned, however all of this hard work will make the end result of this portable so rewarding and totally worth it. Will be keeping an eye on this one!

 

[xsping]

From my point of view， you already made a fantastic work, charging issue will be solved by yourself soon because you have all the ecad file

 

[StonedEdge]

I ended up re-doing a lot of the Wii SPii PCB, including adding my own SD reader that will convert the SD signals into USB 2.0. The little Micro SD port there will be spring loaded, so that the SD card can also be easily accessed from the side of the portable as well. It is a little redundant with the internal data switching, but still nice to have. As you can see, the board has a lot more breathing room now for routing and polygons, so it'll be much more enjoyable to route this time compared to the previous abomination! I'd like to include something up the top right part of the board since I don't really like wasting space, but given the weird shape of the board, maybe I can include some other nice silkscreens there. I'd like to thank Yveltal for suggesting to use SVGs instead of bitmaps. The polgyons look really nice in comparison to bitmaps, which are a little more jagged and pixelated for silkscreens.

I am working on re-doing the entire CAD model so that it can be CNC'd out of aluminum rather than 3D printed. Looking forward to showing you more when I can, hopefully soon! The shoulder buttons also will be completely re-done, along with the case becoming slightly thinner thanks to the removal of the USB-A connector port. I have toyed with the idea of removing the switch sticks and replacing them with 3DS sliders, but I haven't decided yet.

 

[StonedEdge]

Updates, updates, updates! I am currently working hard on finishing the second revision of the PCB!
Today was mostly spent using tRestrict, setting the grid to a very small value and having fun neatening up all of the messy polygons. Thanks to YveltalGriffin for suggesting this idea! Although it does not really impact the functionality of the board, it gives it a nice personal touch and makes the board look a lot neater than the auto-generated ground/power fills EAGLE generates for you.

But, actually, speaking in truth, nothing much has changed from the previous revision in terms of what components are included on the board. I have included an SD reader IC that will allow me to seamlessly read the onboard Micro SD card. The SD card will be accessible and removable easily by pressing on the SD card (i.e a spring loaded card) so that I can easily update the Micro SD whenever I feel like that be necessary. Sometimes, in my portables, I have put a smaller 128GB card and wish I was able to easily update it to a larger card to hold more games. Well, this will now be achievable thanks to the IC! The SSOP package is designed in such a way that the peripheral components can be placed close to the IC pins in a large capacitor/resistor "bank" so to speak. It is a little cramped since I am using 0603 packages, but this will make it easier to assemble of course. It also came in a smaller QFN package as well which might have saved me some space, but the SSOP package works nicely for me I guess.

I also ended up replacing that horrific potentiometer from Digikey (yeah, they don't have many good volume wheels to be honest) that was in the previous revision with something with a much bigger radius but thinner profile. This hopefully will make it easier to turn. I am aiming for the feeling of a GameBoy color feeling volume wheel that is really smooth and pleasurable to turn... uhh... on... and off

Here are some more pictures of the PCB so far. I want to wait a good month before ordering this time to triple check the schematics again and make sure there is no mistakes like last time. As you can see, the routing/planes are a lot neater than the previous revision, and everything has more breathing room compared to before which is a great improvement. The TPS regulator modules are so great, hopefully they are not too difficult to solder into place since they are not QFN but rather a uSIP package.

Here is the top half of the board. This houses the USB-C (still using the STUSB4500) and Gmans power profile code. I plan on porting this over to Jeff's code once I get an opportunity, though. As you can see, there are a lot of vias, with the VBUS signal being connected entirely with a large power plane and several stitching vias. This made it a lot easier to fit the components into a small area of board. The USB data switching and card reader dankness is on the left hand side there (look at all those caps/resistors...) Luckily, they are as close as possible to the chips input/output pins. The GL827L has an onboard 3v3 stepdown regulator and is powered entirely from 5v, so I didn't need to change anything with my switching circuitry/power the card on a separate 3v3 LDO regulator or something. This allowed me to easily incorporate it into the design, without changing much at all (the previous rev used a generic USB-A port which was huge and added a lot more height). I will be able to shave off a couple mm of case thickness now, too!

Here is the RVL-PMS, running the BQ25895M controlled over i2c in combination with the newer (to me, at least) fucking awesome TPS regulators. More stitching vias were used here to connect everything. I made sure to pull 3v3 and 1v8 signals for the U-AMP/GC+ from the bottom of the board (i.e routing them downwards) so as to not cut into the VSYS plane. It is important so that everything can breathe well on the board and there be no bottlenecks for the current to flow to where its needed. I think it looks really good for a noob like me! Without Yveltal, this wouldn't be looking as nice as it is now - a big thanks to him for always teaching me more about PCB layout and giving me little hints along the way (still have a long way to go, though... ). For now, the BQ25895 is great, but I'd like to try and use a different charging IC next time for higher voltage inputs so I can use my switch charger (saving me having to carry two chargers around in my backpack when I'm on long business trips or whatever).

Finally, the GC+ from Pizza Lord, LM49450 (Gman) and some poorly placed FFC connectors which are now finally on the top of the PCB instead of the bottom (what the hell was I thinking last time?) They are not in the most convenient locations (i.e I will have to fold over the ribbon cable for the bottom connector for it to work) but I know that they will fit in these positions from the last revision at least.

Finally, the entire PCB - top and bottom so far (you might have noticed I made it more "square-looking" compared to last time) - there was a lot of wasted space on the previous board that I could have used, given that there was plenty of space in the case for the PCB to be wider/longer. I added some nice silkscreen I guess, some personal aesthetic if you will . Finally, one huge issue I noticed last time was that it was very difficult to get the external components like the USB-C port, volume wheel and on/off switch to fit last time, mainly because the holes were only located in the top half of the base of the portable. This time, in order to machine it, I have made sure that the two halves of the case will fit nicely over those parts instead of having to solder them all whilst the PCB is in the portable. Might be obvious to most, but I overlooked it initially for my contest build (also this is required in order to be able to CNC those parts!)

That's all for now. Hope this wasn't too long, but I'm still excited and morale is high! Can't wait to get started on it again in the flesh soon.

 

[Dmcke5]

Updates, updates, updates! I am currently working hard on finishing the second revision of the PCB!
Today was mostly spent using tRestrict, setting the grid to a very small value and having fun neatening up all of the messy polygons. Thanks to YveltalGriffin for suggesting this idea! Although it does not really impact the functionality of the board, it gives it a nice personal touch and makes the board look a lot neater than the auto-generated ground/power fills EAGLE generates for you.

But, actually, speaking in truth, nothing much has changed from the previous revision in terms of what components are included on the board. I have included an SD reader IC that will allow me to seamlessly read the onboard Micro SD card. The SD card will be accessible and removable easily by pressing on the SD card (i.e a spring loaded card) so that I can easily update the Micro SD whenever I feel like that be necessary. Sometimes, in my portables, I have put a smaller 128GB card and wish I was able to easily update it to a larger card to hold more games. Well, this will now be achievable thanks to the IC! The SSOP package is designed in such a way that the peripheral components can be placed close to the IC pins in a large capacitor/resistor "bank" so to speak. It is a little cramped since I am using 0603 packages, but this will make it easier to assemble of course. It also came in a larger QFN package as well which might have saved me some space, but the SSOP package works nicely for me I guess.

I also ended up replacing that horrific potentiometer from Digikey (yeah, they don't have many good volume wheels to be honest) that was in the previous revision with something with a much bigger radius but thinner profile. This hopefully will make it easier to turn. I am aiming for the feeling of a GameBoy color feeling volume wheel that is really smooth and pleasurable to turn... uhh... on... and off

Here are some more pictures of the PCB so far. I want to wait a good month before ordering this time to triple check the schematics again and make sure there is no mistakes like last time. As you can see, the routing/planes are a lot neater than the previous revision, and everything has more breathing room compared to before which is a great improvement. The TPS regulator modules are so great, hopefully they are not too difficult to solder into place since they are not QFN but rather a uSIP package.

Here is the top half of the board. This houses the USB-C (still using the STUBS4500) and Gmans power profile code. I plan on porting this over to Jeff's code once I get an opportunity, though. As you can see, there are a lot of vias, with the VBUS signal being connected entirely with a large power plane and several stitching vias. This made it a lot easier to fit the components into a small area of board. The USB data switching and card reader dankness is on the left hand side there (look at all those caps/resistors...) Luckily, they are as close as possible to the chips input/output pins. The GL827L has an onboard 3v3 stepdown regulator and is powered entirely from 5v, so I didn't need to change anything with my switching circuitry/power the card on a separate 3v3 LDO regulator or something. This allowed me to easily incorporate it into the design, without changing much at all (the previous rev used a generic USB-A port which was huge and added a lot more height). I will be able to shave off a couple mm of case thickness now, too!)

View attachment 13965

Here is the RVL-PMS, running the BQ25895M controlled over i2c in combination with the newer (to me, at least) fucking awesome TPS regulators. More stitching vias were used here to connect everything. I made sure to pull 3v3 and 1v8 signals for the U-AMP/GC+ from the bottom of the board (i.e routing them downwards) so as to not cut into the VSYS plane. It is important so that everything can breathe well on the board and there be no bottlenecks for the current to flow to where its needed. I think it looks really good for a noob like me! Without Yveltal, this wouldn't be looking as nice as it is now - a big thanks to him for always teaching me more about PCB layout and giving me little hints along the way (still have a long way to go, though... ). For now, the BQ25895 is great, but I'd like to try and use a different charging IC next time for higher voltage inputs so I can use my switch charger (saving me having to carry two chargers around in my backpack when I'm on long business trips or whatever).

View attachment 13964

Finally, the GC+ from Pizza Lord, LM49450 (Gman) and some poorly placed FFC connectors which are now finally on the top of the PCB instead of the bottom (what the hell was I thinking last time?) They are not in the most convenient locations (i.e I will have to fold over the ribbon cable for the bottom connector for it to work) but I know that they will fit in these positions from the last revision at least.

View attachment 13962

Finally, the entire PCB - top and bottom so far (you might have noticed I made it more "square-looking" compared to last time - there was a lot of wasted space on the previous board that I could have used, given that there was plenty of space in the case for the PCB to be wider/longer. I added some nice silkscreen I guess, some personal aesthetic if you will . Finally, one huge issue I noticed last time was that it was very difficult to get the external components like the USB-C port, volume wheel and on/off switch to fit last time, mainly because the holes were only located in the top half of the base of the portable. This time, in order to machine it, I have made sure that the two halves of the case will fit nicely over those parts instead of having to solder them all whilst the PCB is in the portable. Might be obvious to most, but I overlooked it initially for my contest build (also this is required in order to be able to CNC those parts!)

That's all for now. Hope this wasn't too long, but I'm still excited and morale is high! Can't wait to get started on it again in the flesh soon.

View attachment 13963View attachment 13961

Looking good! Can't wait to see the updates to the case design when you get up to it!

 

[StonedEdge]

The new shoulder button prototype is complete (for the right trigger, at least). I changed the shoulder button to replicate the original SP style, which uses a hinge pin and a spring to keep the shoulder button from falling down. There are also little "flaps" that prevent the shoulder button from falling out as well - they are also designed to cover the gaps between the top part of the shoulder button and the case housing. I really wish that there was a flip/mirror function in Fusion 360 so that I didn't have to replicate this for the left shoulder button. If there is a way to replicate something like this without re-modelling, I'd love to hear how I can do that (although afaik its not possible). (Check below image, wow that was easy lol). The originals for the contest were rushed and literally were super glued to the dual tact switches, which is far from how I'd like them to be for the final model.

I'll need to run several test prints on this to tweak the feeling of it but so far I think it is a good starting point. The spring fits between the two holes of the lever and the pin holds it in place. One end of the spring goes into the little slot I made (temporary solution for now) and the other wraps around the shoulder button into the bottom groove.

I hate designing shoulder buttons because they are hard to get right the first time in terms of feel, but I'm happy with these for now at least as a prototype. Will report back after I have done some test prints to see how they are - its hard to know without properly testing. I plan on printing these on a resin printer since they are quite intricate, so FDM probably wont work so well to pick up the small details. (Contemplating buying either the Elegoo Saturn or the Elegoo Mars 2 Pro - leaning towards the Saturn since it has a bigger build area, though).

 

[Stitches]

The new shoulder button prototype is complete (for the right trigger, at least). I changed the shoulder button to replicate the original SP style, which uses a hinge pin and a spring to keep the shoulder button from falling down. There are also little "flaps" that prevent the shoulder button from falling out as well - they are also designed to cover the gaps between the top part of the shoulder button and the case housing. I really wish that there was a flip/mirror function in Fusion 360 so that I didn't have to replicate this for the left shoulder button. If there is a way to replicate something like this without re-modelling, I'd love to hear how I can do that (although afaik its not possible). (Check below image, wow that was easy lol). The originals for the contest were rushed and literally were super glued to the dual tact switches, which is far from how I'd like them to be for the final model.

I'll need to run several test prints on this to tweak the feeling of it but so far I think it is a good starting point. The spring fits between the two holes of the lever and the pin holds it in place. One end of the spring goes into the little slot I made (temporary solution for now) and the other wraps around the shoulder button into the bottom groove.

I hate designing shoulder buttons because they are hard to get right the first time in terms of feel, but I'm happy with these for now at least as a prototype. Will report back after I have done some test prints to see how they are - its hard to know without properly testing. I plan on printing these on a resin printer since they are quite intricate, so FDM probably wont work so well to pick up the small details. (Contemplating buying either the Elegoo Saturn or the Elegoo Mars 2 Pro - leaning towards the Saturn since it has a bigger build area, though).

View attachment 13988

View attachment 13989

Get the Saturn and we can be same-planet buddies!

Also you could print the entire case in resin for easy sand and polish

 

[StonedEdge]

Get the Saturn and we can be same-planet buddies!

Also you could print the entire case in resin for easy sand and polish

Actually, I have a big surprise for the case and how its going to be done. You'll have to wait probably 2 months though before its all done, but its going to be very nice!

 

[Aurelio]

Actually, I have a big surprise for the case and how its going to be done. You'll have to wait probably 2 months though before its all done, but its going to be very nice!

Surprise? You've been talking about it everywhere lmao

 

[MRKane]

I'll have forgotten by then so look forward to experiencing surprise! Then being impressed, then that feeling shifting to jealousy that it's so beautiful lol

 

[StonedEdge]

Surprise? You've been talking about it everywhere lmao

Heh yeah I guess I have. I guess the true excitement will be seeing how transparent I can get it after a bit of wet sanding with high grit sandpaper

 

[StonedEdge]

The main PCB is pretty much finished. I am going to wait a couple of weeks to order it as I want to make sure there are absolutely zero ERC/DRC errors this time around. I already noticed the 5V boost regulator package is broken and doesn't connect to the GND plane, so I'll need to fix that. I added some standoffs for the Wii (the large 6mm holes in the bottom left and top right corners) so that the Wii can be screwed down. It's a little bit different, in that the Wii motherboard screws to the front of the case (most Wii portables screw the Wii to the base half, but there is no space in my bottom because of the batteries). I had to re-route a lot of my board and move stuff around to make this work, but it will be worth it for assembly if I want to make more than one in the future!

I think I'll be getting black silkscreen from JLC this time. It just looks so good.

A big thanks to Dmcke5 for coming up with these standoffs! Since there is no room for front screw posts that seal the case from the bottom, we came up with right angled posts that screw directly into the countersunk holes in the top case. The bottom case has vertical posts that will have tapped holes in them. It looks ugly now, but with anodized black countersunk screws, I don't think you'll notice them when it is all together.

 

[thedrew]

I'm diggin the 3ds sliders this time around!

 

[Wesk]

So how much thinner is it now that you've moved to sliders?

 

[StonedEdge]

So how much thinner is it now that you've moved to sliders?

I haven’t put sliders in it. The main reason for the drop in thickness is because I removed the USB port for a Micro SD card slot. The top thickness is... yes, something like 6mm which is ridiculous, but there’s not many alternatives when putting sticks in a clamshell. I’d say the 3DS sliders would allow it to get down to the 25-26mm mark, though.

 

[thedrew]

Black silk screen from JLC is super slick, also easier to see solder joints for them QFNs imo. Is the headphone jack properly grounded here? Just happened to notice no via and the ground plane seems to cut off

 

[StonedEdge]

Nope, they are connected to the internal plane with these two vias.

 

[StonedEdge]

A couple of big updates to the project, I think!

New (very different design) shoulder buttons are in, and they feel wonderful. They use a spring hinge design that is built into the triggers themselves, rather than a physical spring. The hinge arm has had a weak point inserted, just enough so that the triggers will flex nicely and return back when released. The triggers are screwed into the back housing with two M2 x 5mm screws to keep them tightly secured. The triggers are now a lot, lot bigger, so much so that they take up the entire back housing. This makes them a lot easier to push than the previous ones which I'm really pleased about. (Reminder: the last one's were superglued to tacts!)

Something I am really happy with is that my portable is now going to have real analog triggers (no dual tact method) thanks to the wizardry of a couple of neodymium magnets and THT hall sensors aka electromagnetism. The sensors work by detecting the field generated by the magnet, and convert this field into a variable voltage reading that can be understood by GC+ 2.0 depending on the strength of the field. It's hard to predict the exact voltage ranges though that it will output, and some software changes will be required (implementing some scaling LUTs for the analog L/R readings and hard coding them into the firmware) to try and exercise the full range of the analog scale (0-3.3V). Hopefully I can get this working after reading the analog values that GC+ 2.0 configurator gives me. Thank god for Pizza Magic.

I already posted a video of them working in the Discord chat, but for those who didn't see:



Hall sensor extends half way off the PCB to be parallel to the 6 x 2mm magnet. About 3.5mm of travel is achieved. Probably the best I could get with an 84 x 82mm case.

Final digital press is actuated by a lever that is separately printed and secured with a M2 screw.

Here they are inside of the case (along with the Wii and 2 x 18650s. I've managed to shave about 2-3mm of thickness of the case entirely and made sure that everything fits snug).

The routing is pretty much completed now with a few minor tweaks required (I think I said it was complete above but things keep changing, including placement of parts and things as I move stuff around). I added a spot for the BT module as well as added the full MX/RTC circuitry, replacing the larger CR2023 battery with a CR1025 (less RTC life, but a much smaller footprint so it fits all in a tiny section of the PCB). There was no room on the Wii anymore to have these, the tolerances are really, really tight in this build now. This thing is uncomfortably small. The "red X's" are roughly where the SMD digital buttons will sit. I haven't put them on the board yet, but they will be activated with a small lever - it all seems to work in the CAD simulations, but we will have to see in practice how it holds up over time!

Random digital SMD button that offers a "very clicky feel" for the digital press.

Snapshot of the front side of the PCB. Almost finished. Now with MX and Bluetooth (not pictured on the front).

Next steps are to get the digital buttons in place and shell out the rest of the front housing where excess material is not required. After that, perhaps a 3D printed prototype build and testing out the shoulder button functionality, probably with some protoboard or with a cheap little proto-PCB. It goes without saying that I owe a big, big thanks to Dmcke5 so far for helping me along the way with re-designing parts of my case, and giving me tips about getting it CNCd along the way! The man has a CAD addiction.

 

[Gman]

Very good idea on the shoulder buttons!