Worklog Wii U Mini Worklog

[Lazr]

So I installed Fusion, and now i'm stuck at logging in! Progress nonetheless!

But maybe it was a good thing that I held off on the shell.
I was looking at my WUP-40 board the other day, and I noticed that the back USB traces are just about right where the module ends:

What does it mean? Well the trim line I used originally used was pretty aggressive, since it went right by the DRAM passives. However, the WiFi module sticks out just enough that the trim like can be right below the USB traces. One (or two?) less things to rewire, and it makes all board revisions compatible! (not that they wouldn't have been compatible before, since I wasn't going to rewire the module anyway).

Speaking of the module..Remember that pin I needed to wire 3v3 to? It has a testpad, but more importantly it's within the trim line!

I didn't notice that this TP went to that pin until I was updating the SMC pinout on WiiUBrew. I much rather use the TP than the via under the module plug (and short 3v3 to GND with the module shield..)
Only unfortunate part is that my NAND relocation on the original Wii U Mini trim covers up the testpad. Oh well!

Anyway, here's what the current trim lines look like now:
A5X boards:

B1X boards (aka just WUP-50):

The plan is figure out how TF to sign into Fusion. Nold told me in the discord that there was a custom .desktop in the issues, but I couldn't find it. Maybe I need to look harder?

Next is to get the PCBs made for the PSU++ and assemble them. It's basically the same BOM as the OG PSU+, with some feedback resistors moved around/changed. This might be awhile, since money is very tight for me atm.
(The source for that is here).

Lastly, I'll trim one of my many Wii U's. It won't be a 50, those should probably be saved for portables. I'm thinking of using my WUP-40 (It's the best board rev behind the 50 imo, despite it being basically identical to the WUP-30). This will happen whenever I can assemble the PSU++ boards.

Last little tidbit, I've ordered some of the drh-flash-flex PCBs. About $30 in total. Not bad! I only ordered 10, and I haven't tested them yet, but the flex source is here. Hopefully I didn't fuck up anything. It is a pretty simple flex, but you never know until you actually test it..

 

[Lazr]

I haven't tested them yet

I've just tested them today/yesterday (times are weird). Post for that is here.

Also:
I've installed Windows on a hard drive, installed fusion, and then I did the same thing I did with OnShape: make a hollow cube and then never touch it again. This project was intended to make me learn 3D CAD before doing a full Wii U portable. I have no excuses really, I am just not wanting to do it lol.

 

[Lazr]

4 (ish) months later and I'm back in the zone!

For starters, I gave up entirely on the eMMC SDIO. Wouldn't be the first time its completely died on me, even doing everything "correctly". Because of that, I ordered an mSATA SSD from eBay. Yes yes, used flash storage bad, but it's not like the "new" options looked very good either. I'm unsure if any 'big' SSD manufacturer even make mSATA SSDs anymore. Moving along, it's a 128GB Samsung SSD, perfectly usable for what I need it for.

I'm not even sure if my current 3v3 reg can handle this + all the other devices on the 3v3 rail, we'll see.

The biggest downside to using a SATA MLC in general is that Wii/GCN VC (Injections or not) do not work with it. It's not a huge issue if you also have an external USB drive or use SDUSB, but it's still something to be aware of.

I haven't actually installed it on the Wii U yet, as all of my time today has been spent trying to get the RTC to turn on the console. My first idea was relocating something we (Yveltal lmao) worked very hard to get rid of: the ABLIC. I didn't actually do that first though, as while I was theorizing as to why the RTC worked on an untrimmed console vs an untrimmed console, I came up with the idea that maybe shorting all of the voltage rails on the daugherboard was the root cause, and I was right.

(From the BB discord in #modding-discussion)

I would also like to thank Yveltal for doing all the hard work for me and tracing out the voltage rails on the internal layers of the WUP-01, and being kind enough to send a screenshot of it (because I can't easily open .pdn files on Linux). Really saved me the time of tracing it all out myself. You rock, man!

(From the BB discord in #modding-discussion; sharing the full Wii U mobo screenshot here for posterity)

Few hours later when I finally got home, I got right to work and removed the daughterboard from the main board.
Everything was going well!...or so I thought.
While I was wiring everything, I had assumed that the entirety of the RTC was ran off the standby 3v3 rail. I was wrong. There is a single pin I missed when looking it over that is ran off the "standard" 3v3 rail (probably to detect when the console is on).

(From the BB discord in #modding-discussion)

After fixing that, it worked! I didn't even need to wire 3v3 to the pin since it worked then too, but I did so anyway "just in case".
I also learned that rewiring TP177 to the RTC isn't required, but it would still be good to know WTF it's for.

This single mistake cost me a few hours, but it also led me to relocate the ABLIC:

I did this for a few reasons, mainly:

• I unintentionally lied on one of my previous posts about the console working fine when the RTC turned on the PFET with the ABLIC gone. It does not. The 1v15 rail relies on the ABLIC for some reason, and I didn't notice this when initially testing. That's on me.
  • I tried hooking up the lines to a cheap 24MHz logic analyzer, and I found it weird the 1v15 line (not sure if its EN or PGOOD) goes high in between the 3v3/12v PGOOD and 1v PGOOD, but I have no idea what im saying if im being honest.
  • 
  • It might work fine if pulled up to 3v3 through a resistor or something? I never tested it though.
• So that I could have the standby 3v3 rail. I'm not sure if the PSU-PlusPlus will even work for this project anymore, but maybe if a pullup resistor is put on pin 3 of the 1v15 controller it'll work.
• Troubeshooting. The thought process was "Maybe the RTC needs something from the ABLIC to do it? Maybe the ABLIC needs to also be connected to the Gate?"

Heres a video of it finally working. I was super excited when I saw that the voltages were on without me doing anything but press the Power button.

The blue LED is blinking purely because the DRAM didn't have any power, but boot1 (the source of the blinking) is very good news.

The plan from now is simple:

• Figure out how to power the DRAM. Maybe I'll take the 1v5 regulator from the dead WUP-50 I stole the ABLIC from for this board. I got rid of the other regulator because it was quite literally falling apart.
• Rewire 5v. I have 24AWG silicone wire now, instead of the other crap I used before. This should clean it up really well.
• Wire the SATA SSD and set up the MLC on it. This might take the most time (because SATA and trace soldering two differentials), and I have to just hope that the dinky regulator can power it.
• Continue frankencasing. I cut the bottom piece shorter already, but i need to bond it to the faceplate as well. (I use epoxy and bondo for this yeah?)
• Rewire the buttons and LEDs from the SMC to the front panel connector, instead of having it hardwired directly (this also gives a bonus of finally using resistors for LED_Y and LED_B; LED_R already has one on the SMC daughterboard).
• Rewire the SD slot and front USB ports to a cut piece from another Wii U (Thanks Yveltal for sending me that WUP-40).
• Maybe redo the 5GHz relocation with the DRH flash relocated. EDIT (2/11/25 12:15): It's no longer a maybe, The glue stopped holding.

The light is very much visable at the end of the tunnel, but it is still some time away.

 

[Lazr]

So much has happened in the past two days, and not all of it good..

First thing I did after my last post was redo the 5GHz relocation. That went down smoothly, even though I couldn't test it until MUCH later.
I did relocate the DRH flash using one of my flexes, but that's easy stuff.

I don't enjoy soldering to the B2B footprint, but it's more than doable. Small glimmer of fun here was wiring up the "TP157" pin on the module to the SMC, and watching the Gamepad turn the console on.

I also rewired the 5v regulator, and it's MUCH cleaner compared to before.

I went to sleep afterwards. Went right back to it when waking up though!
I cut, sanded, and relocated the 1v5 regulator from that dead WUP-50 board in about an hour. This stuff gets real easy as you learn and do.

Cleaned up the pads on the mobo before relocation. Probably could have gotten away with not, but I wanted it to sit flat. I also relocated C55 over to the footprint of C41. I'm pretty sure it's not entirely needed, but it won't hurt anything either.

Completed all the wiring for this daughterboard. I didn't feel like bodging on a resistor to the EN pin on the controller, so I just wired it up to the RTC. It's like this on a stock console, so theres no harm in doing so.
I should also probably have another GND wire, but that is a thought I am having as of writing.

Full pic of the top before we get into to some real nice stuff:

Now for fun stuff. SATA!
Wired up two pairs of twisted magwire, 24AWG to some of the 3v3 pins, and 24AWG (not pictured) to GND at the back. Maybe could have gone tighter, but its fine for being hand-twisted (my current setup for twisting magwire is a dremel, which does work I just don't like using it)

Glued it down to the board. I'm not a huge fan of it sitting ontop of the core VRMs, but theres literally no where else to put it.

Also made myself a little diagram, because I couldn't find anything that so easily just showed each pin and what its for.

Wired it up and it worked first try! I did have weird issues with setting up the MLC on it, but i'm just gonna chalk it up to the current 3v3 regulator being unable to reliably provide for it. It eventually finished after a few hours (MLC setups should only take a few minutes fwiw). I ordered a PTH08080, which should be able to provide more than enough for the SSD + everything else on the mobo. Mobo takes like, 300mA of power draw, though i'm not sure if that includes the LDOs for 1v25 and 1v1. Anyway, those rails are also low power, so it's still perfect.

Shitty pic of progress I sent to a friend.

And there we go! More than enough space for all the games I care for.

At this point, I was hyped. Seeing this come together the past few days have been a treat. This is where the downfall starts, however.
I noticed that I couldn't power the console off from the power button, but I could from the Gamepad...weird. I found it especially weird that I could power on and navigate minute_minute with the same button.

I'm not sure why I thought this was a good idea, but I tried to tie all module pins on the SMC (besides TP157) high. I thought "oh maybe its waiting for all the modules to be ready".
That was wrong, and in the process, I completely killed the ABLIC (dead short to GND), 5GHz module (3v3 has a very low resistance to GND. The Gamepad can't connect either), and some of the RTC (1v5 would never be enabled. The 15v PFET was always being "turned on"). I'm also thinking some of the SMC too, due to how low the resistance between the standby and "main" 3v3 is.

I tried to replace the ABLIC, and while it technically worked, that was not the only thing I killed on the daugherboard, so when trying to power it up, I killed that one too. At this point, I was super irritated and just went to bed.
When I woke up, next to nothing happened until way late into the evening. The whole day was just thinking as to WTF the deal was with the power button.

This is when I started to look deeper into the damage I had caused.
First thing I did was remove the dead ABLIC daugherboard. I also killed the MOSFET, which is kinda an important part to a switching regulator. I didn't bother trying to replace anything on that board, and I just cut another daugherboard off the same WUP-40 I cut the front USB ports and SD card slot off of. It's not much bigger than the other daughterboard, though I had to get super close to some of the components here. All of the wiring is the same as before. Tested it every time I changed something to make sure I didn't just kill another one. This was also the time when I learned the RTC wasn't doing so good.

ATP I decided it was probably better to test this on another working system.
My second LOLWUT I tried pressing the power button on, and it shut off just fine.
I had the idea that maybe TP177, the one thing to the RTC I didn't wire this time because I didn't think it was needed (and also the only difference between the trims besides all of the SMC voltage rails being shorted) was the cause, but it just didn't make sense.
To prove that it was/wasn't the cause, I desoldered the TP177 wire on the LOLWUT. Imagine my surprise when pressing the power button stopped working there.
I had thought that it wasn't needed since the Wii U booted just fine without it. Now im really curious as to what all its for (besides "safe shutdown" things).
(Sidenote: Love how its easier to test on a trimmed console now compared to an untrimmed)

Now I knew what the problem was with the button, but I had to find a replacement RTC.
Luckily, I have a board I was needing to put into the "Scrap Wii/U Box" and had a perfectly usable RTC I put on to try and get it working again (keyword "try").
Going back to the Wii U Mini trim, I desoldered everything off the relocated SMC daugherboard in order to remove it from the board. I did a shitty job securing it anyway, so it would have to come up sooner or later anyway.
Hot air'd the old RTC off, put the new one on, rewired everything, and it started working properly again!
Probably going to have to replace the SMC as well though.. Not as big of a deal as I do have spares.

Plans for later are to rewire the 5GHz module and wait for my PTH08080 to arrive. If it doesn't fix the unstable booting issues, then i'll be replacing the SMC.

Theres probably some things to mention I already forgot about, but I'd rather post what I do remember than not at all.

 

[Lazr]

it will be a working Wii U running on custom regs

lol that didn't last for very long.

The RTC im using here was is from the WUP-40 board, so this RTC should be functioning completely fine.

This RTC was not fine, and I destroyed a perfectly usable WUP-50 because I didn't realize this sooner.

Really during these early days, I didn't know what I was doing or how I wanted it to turn out. Really glad I took that long hiatus from this to just go and learn more.

Anyway, for today's update:
I got the shell shape pretty much done. Just needs a bit (understatement) more processing before it'll be presentable.

In-progress pic. Just showing how its all going to be laid out. In theory I could have moved the SD slot back quite a bit, but I'm still quite happy with this result

Top and Bottom pics of the shell after it was all finished. ATM I just "welded" the parts of plastic to the shell with my iron. Will bondo it later, but this snow needs to get out before I can go anywhere. Snowed outside and inside today!

I will say that the size diff between this and a stock console is really fucking pathetic, but its not like the Wii Mini was that much smaller compared to the standard Wii either..

Speaking of the standard Wii, it's just slightly smaller than it (though not by much!):

And here's it compared to Nolds Wii Micro:

I'm thinking of painting it White, but I'm not opposed to color suggestions!

Oh I also redid the 5GHz module. Took longer than usual because I forgot to solder TP99 to the module, and when I tried to lift the module I snapped some of the wires in half.. Oh well, it works!

Ignore that there's no LEDs at the front. I don't want to deal with those just yet, since I still need to be able to remove the Wii U from the shell easily for processing.
Plan is to finish the shell, and then I can start the mass wiring of the faceplate. I also need to figure out the Sync button since when I cut the SD daughterboard it didn't even cross my mind..

 

[Lazr]

My Mini console is better.

Yeah, it's done. It's far from perfect, but perfection can never be reached, so I'm happy with how it turned out.

It's been over a month since the last post, but most of that was doing the shell work. There's not too much to share from that, its just bondo -> sand it flat -> bondo -> sand it flat. Do that over and over on spots you want to fill and then paint. I also used some acetone and some Wii shells to create some "ABS cement" to secure the front panel to the shell better.

As for what I'm actually used to: the noods.
I think the first thing I did after the previous post was cut the front IO from another dead board (that poor WUP-01 with its missing SMC and RTC..). When I cut the other IO board, I forgot about the sync button, and I didn't really know what to do about that besides just cut another board. Bonus was being able to easily plug the bluetooth module up at the front like a stock console, so WiiMotes and other controllers should have the same range as a stock console.

The original spot I had for the PTH08080 was bad due to the inductor being too tall (so the board wouldn't sit flush in the shell), but that was an easy fix.

Then I think I cut the front panel connector from yet another dead board. I'm pretty sure that this singular unit has parts from all retail boards now. The wiring was a little bit annoying, since I had to wire it to the SMC which was on the opposite side and I had to somewhat "guess" which wire was for what, but eventually I got it done.

After that spaghetti, the only things left were just USB (BT included) and the SD Card.
Wiring USB was actually super easy. I didn't twist anything (which is probably why its was easy), but it works and I didn't do anything crazy with how long the wires were, just trace to trace.
I had issues with BT though, which really put doing anything with this at a pause, since I had no idea what the problem was. The SoC pin was pulled up to 3v3, which should have been the only thing needed, but it just wouldn't work. The GPIO pins on the module weren't connected to the SMC, but that was exactly the problem, despite the previous relocation being just fine without those pins. My hypothesis is that on the front IO board, those pins still had some pullups/pulldowns, so the BT module just wouldn't init or something, but I have no actual proof of that being the cause. Anyway, after it started to work, I got to work rewiring the front USB ports and then wire the SD Card.

USB was still no issue, and same with the SD Card. The only "issues" I had were just being stupid and swapping DAT0 and DAT1, so the SD Card wouldn't init.

And here are pics of the board before closing it up:

Now for anyone asking how to build one:
1. Cut Wii U Board
2. Cut Wii U Shell
3. Bond Shell parts together to make New Shell
4. Put Wii U in New Shell

Obviously shitposting there, but this was a really fun learning experience and I do not plan on making any more.

Shout-outs to my Dad for answering my questions about Bondo and painting plastic and everyone who gave advice in the discord for frankencasing.
Huge thanks to YveltalGriffin. This would not have been possible without him and his amazing Wii U Trimming guide and WUP-01 scans.

 

[Nold]

great work! just kinda sad that from the front (esp with the controller pad on it) you can barely tell it's different.. guess it's left to someone else to make a slim version

 

[Shank]

So awesome. Have you considered a version 2 where you stack a disc drive on top of the heatsink like a GameCube?

 

[Lazr]

So awesome. Have you considered a version 2 where you stack a disc drive on top of the heatsink like a GameCube?

i had this idea when originally "planning" (my inspiration for the idea came from the Wii Mini though) but until my MLC SDIO curse is lifted, probably always going to use SATA for MLC. I also don't really have any ideas on how to implement it, or if its worth the effort.