Wii U Stuck on Logo

[ChefVortivask]

Hello everyone,

First time poster here. I decided to make this post since I am having issues with my Wii U. Currently, my Wii U will not boot up to the main menu. It stays stuck on the Wii U logo indefinitely.

There are a few things to note about my console. The console is the black 32 gb model. I believe it was bought during the 2012 holiday season (release) as we were expecting to play Pikmin 3 before it got delayed. The system has never been modded in any fashion. It was not the most heavily played console, and I can't even give a good estimate of when it was last used (at least more than 3 years).

I have already tried doing the UDPIH, however it does not seem to be working. When using the second recovery menu, I get a white screen with some blue lines (see attached image). I believe this is similar to what @fadafwet described in his post. I was able to get the logs from the system, which I have attached to this post. Any idea what the issue may be and how to fix it? Any help would be much appreciated.

 

[ChefVortivask]

That’s annoying, but at least we’re making progress. If I added the pull up, how exactly would that be done? You said before that I would put the resistor between CMD and 3V3, but I never thought those two were directly connected. I guess I need to take a look at a spec sheet or diagram (any suggestions or links?).

As for the oscilloscope traces, I have a couple ideas about how I can have those done. There’s an electronics shop at my university, but since this is a personal project, I probably need to be specific with what I want done so it doesn’t take up too much of their time. If they can’t do it, I may know a couple people who can help me out.

 

[SDIO]

You said before that I would put the resistor between CMD and 3V3, but I never thought those two were directly connected

They won't be directly connected, the will be connected through the resistor. One end of the resistor on the CMD line and the other on 3V3. Since the resistor has a relatively high value, the low side transistor in the outputs of the Card and the Host Controller can still pull it down relatively easily. But when they are not activly driving it the signal will be pulled high through resistor.

The spec says the pullup resistor can have a value between 100Kohm and 4.7Kohm. The smaller the value the strongr it would pull. You can start with a high value and see if it changes something and if it doesn't help add more resistors in parallel to reduce the combined resistance (or just use a smaller resistor)

For the oscilloscope it probably has 4 probes, so we would wan't to see at least CLK and CMD and then 2 of the DAT lines. You can try first DAT0 and DAT1 and then the other two to see if they any of them look strange.

If they have a protocoll analyzer for SDIO that could be helpful in seeing at which point we are, but that is optional, as we are more interested in the electrical side. For the CLK we want to see that it is even and sqary. For all the signals we want to see that they reach the high level, somewhere around 3V3 and go down to almost 0V. Also we want to see that the rise and fall times are not too long. Since they probably have experience I would assume they would notice if a signal looks unhealthy.
Also not it will start out with a clock at ~400kHz which should then be rised to 52Mhz, we want to see if it looks good at both speeds.

 

[ChefVortivask]

Thank you for that information. I'm currently working on getting help with that, but in the meantime, do you have a link to that spec sheet that I could take a look at? I'm trying to wrap my head around what is going on, which may be difficult since I haven't been formally trained in electronics. I don't even understand how exactly CMD and 3V3 are connected through a resistor since I thought 3V3 to the NAND-AID powers the SD card which can then send out its signals (which is probably wrong). I guess the main thing I'm confused about is where exactly the resistor would go. Obviously you've stated it goes between 3V3 and CMD, but where exactly on the board would it go?

 

[SDIO]

where exactly on the board would it go?

for example between then CMD pad and the 3V3 pad on the NAND-AID. But any other point where it can conect to the signal would also do.

3V3 is not only power supply but 3V3 also represents the logic high level

 

[ChefVortivask]

If it really does just have to be somewhere between CMD and 3V3, then the pads on the NAND-AID would probably be the easiest. I thought you’d say something like that, but I didn’t think it would be that simple.

Post automatically merged: Jan 11, 2024

I want to give an update. I contacted the electronics shop I mentioned and this morning I went and met with guy in charge of the shop (I actually sent him links to some posts in this thread so he has a better idea of what I need to do). They have an oscilloscope I can use, but no protocol analyzer for SDIO. As previously stated, that shouldn't be a big deal.

I want to clarify a few things as I move closer to getting the oscilloscope traces. Firstly, will I be powering on the Wii U for these tests? I assume so since we're looking into voltage, but will I be using de_Fuse for these tests or will I remove the SD card from the front of the console that contains minute, thus starting the Wii U in its natural state? I realized this morning that I still have the version of minute that doesn't initialize the mlc, so I would think that would work against us.

My other question is regarding the number of probes and channels. It was stated that I'll be wanting to use four probes, one for CLK, CMD, and two of the data lines. I'm still looking into how to use oscilloscopes, but would we have four probes on one channel or two probes on two channels? The probes are also hooked, so I'm not exactly sure how I would go about keeping these probes attached to four separate points at once.

I'll be going in again on Tuesday so I may have more questions up until then.

 

[SDIO]

Sorry just saw now that you have added some thing.

Each probe is on it's own channel. There are probes which can grab onto things. So if you have short wires soldered to the pads and the clk trace you could hook on that. Yes you would power it up. You can try with and without defuse to see if that changes anything. If you only have two channels then that's probably also fine for now, as we just wan't to see if the waveform looks good. Even one channel would be good enough for that.

Also @Lazr1026 now has SATA adapters, so that would also be an option

EDIT: also make sure that the GND from the probes is only connected to the GND of the console. Connecting it to anything else might damage the console or the scope

 

[Lostbhoy]

I hope you don't see this as trolling but you've been at this for nearly a year and now the talk is about oscilloscopes! How much have you spent on equipment??? Could you not have bought another wii u by now??

Honestly, I admire your dedication and wish you luck!

 

[ChefVortivask]

@Lostbhoy You're not the first person that's been somewhat astonished that I've been tackling this for so long, so no worries. I understand that from the outside looking in, it seems a bit ridiculous that I've stuck with it and not just bought a replacement, but I feel like I'm very close to fixing it. I honestly haven't spent that much money on equipment or materials since I had most of the equipment before starting this. The only things I bought for this project were some micro SD cards, some PCBs, micro SD card slots, a Raspberry Pi Pico, and some soldering iron tips. Since I can use the Pico and the soldering iron tips after this project, I don't really see them as a loss. Thankfully I didn't have to purchase an oscilloscope. If I did, I probably would have just bought a replacement motherboard or something since it would be cheaper than an oscilloscope.

 

[SDIO]

I bought a faster logic analyzer to roubleshoot my wii u. That also wasn't cheap

 

[Lostbhoy]

@Lostbhoy You're not the first person that's been somewhat astonished that I've been tackling this for so long, so no worries. I understand that from the outside looking in, it seems a bit ridiculous that I've stuck with it and not just bought a replacement, but I feel like I'm very close to fixing it. I honestly haven't spent that much money on equipment or materials since I had most of the equipment before starting this. The only things I bought for this project were some micro SD cards, some PCBs, micro SD card slots, a Raspberry Pi Pico, and some soldering iron tips. Since I can use the Pico and the soldering iron tips after this project, I don't really see them as a loss. Thankfully I didn't have to purchase an oscilloscope. If I did, I probably would have just bought a replacement motherboard or something since it would be cheaper than an oscilloscope.

That's not as bad as I thought then and yeah the tools will come in handy elsewhere but the journey and the knowledge gained will be highly beneficial to you and it's the satisfaction at the end... I do hope you get success man, you deserve it for coming this far! I woulda chucked it long ago!

 

[ChefVortivask]

I do hope you get success man, you deserve it for coming this far!

Thanks man. You're definitely right about the learning and experience I've been gaining. Also, it's not like I can use the Wii U in its current state, so in my head I figure it's worth fixing it. If I somehow mess something up and irreparably damage it, then it's not really a huge loss since it's not like it was working months ago to begin with.

 

[ChefVortivask]

There are probes which can grab onto things. So if you have short wires soldered to the pads and the clk trace you could hook on that.

I would think that with the hooked probes I could just clip onto the wires. I don't know if there are any other probes available, so it might be best if I just solder some wires to the pads. I would just need wires for CLK, CMD, and the DAT lines right? I should need anything for 3V3 or GND since there are multiple places I can hook onto for GND.

Also apologies for the lack of updates. I've been busy recently and haven't had a good chance to go probe the Wii U yet. I hope to have some time this upcoming week.

 

[SDIO]

Yes that would help. Also you should try to get GND close to the signals. You don't need 3V3

 

[ChefVortivask]

I was able to probe the Wii U with the oscilloscope today and have some interesting findings to share. I didn't test with de_Fuse (just the Wii U as is), but now I have a much better handle of the oscilloscope.

Starting with the CLK signal, it appears to me that the signal is fine and acting as it should. As stated previously:

For the CLK we want to see that it is even and sqary.

I've attached some pictures and you can see that the CLK signal which is the yellow wave corresponding to channel 1 is indeed a square wave. I noticed that upon powering on the Wii U it takes a couple of seconds for the CLK signal to respond (or start, not sure the proper terminology), but once it does it stays as a square wave with a frequency of 375 kHz. It never reached 3V3, only about 1.7 V.

With the CMD signal (blue wave, channel 2) the wave shoots up around 3.1 V upon turning on the Wii U and stays there. I never observed the voltage change, let alone go down to 0 V. The frequency also fluctuates a lot, but it's possible that may be due to noise. It's possible it was at 400 kHz initially and didn't notice, but it shoots up to well beyond 52 MHz to several hundreds of MHz. I even saw it climb to over a GHz at some times.

I also probed the D0 and D1 (channels 3 and 4, purple and green respectively) lines but didn't notice any change when powering on the Wii U. I take that to mean something is wrong as there probably should be some response from them.

I've attached several pictures to give a better illustration of what I did and what was measured. I have a theory that it's possible that the D0 and D1 lines weren't responding and CMD was acting weird because the micro SD card isn't formatted or set up as an MLC, but I could be wrong.

 

[V10lator]

I noticed that upon powering on the Wii U it takes a couple of seconds for the CLK signal to respond (or start, not sure the proper terminology)

I guess that's normal: It needs some time until the boot process reaches the point to initialise the SDIO / eMMC controller.

but once it does it stays as a square wave with a frequency of 375 kHz.

Shouldn't that be around 50 MHz? I'm not sure, so better wait for SDIO.

it shoots up to well beyond 52 MHz to several hundreds of MHz. I even saw it climb to over a GHz at some times.

I'm no expert on this but I don't think any signal should ever be faster than the clock (CLK). 52 MHz would be around 50 MHz, so that sounds like the speed it should have (but also the speed the CLK signal should have). Again, take what I say with a grain of salt and better wait for SDIO to reply through.

I also probed the D0 and D1 (channels 3 and 4, purple and green respectively) lines but didn't notice any change

Sounds like an issue, too. I guess at least D0 should show activity from the host controller trying to init, even when there's no card nor the eMMC connected. Again, I'm no expert on this through.

 

[SDIO]

It's normal that it starts out with 300 something kHz, it will only ramp up the clock once it determined the capabilities of the card.
The 1.7V seem low, but that might be normal before it knows if the card supports 3V3.
You should see activity on CMD. Maybe you have to set a trigger for that. The Wii U will only try twice, if it doesn't get a response it will give up.

For the hight frequency, that might just be noise (maybe the line is floating) or it could be a single transition on the signal, which happens much faster

 

[ChefVortivask]

Maybe you have to set a trigger for that.

By trigger, do you mean the trigger on the oscilloscope? It was briefly mentioned to me by the head of the electronics shop, but since I saw signal from CLK, we thought I was fine.

 

[SDIO]

yes trigger on the pscilloscope, it can trigger on rising or falling edges and start captruing at that point. After the Wii U gives up, you won't see any more activity on the other lines.

 

[ChefVortivask]

I did some looking into the trigger and from what I can tell, it's used to sync the oscilloscope to the signal? In any case I'll probably need someone at the shop to help me with that. Would I just be adjusting the trigger via trial and error until I see the CMD signal? Also I have a variety of resistors now so I should be able to test them once I find the signal.

 

[SDIO]

You would set the trigger to the rising or falling edge on the CMD signal. (see which doesn't trigger too early)