Question What exactly is the U10 chip?

[loopj]

Hey all, I'm wondering if anyone has discovered exactly what the U9/U10 chip is. I'm trying to figure out if it is an off-the-shelf, commercially available component, or a custom chip that Nintendo or one of it's partners made?

It looks to be a SSOP5 package (the same footprint as a transistor SOT23-5 package) but since it is labelled U9/U10 rather than Q9/Q10 I'm 99% sure this is not a transistor.

One of my U10's that I damaged during desoldering has the marking B37G.

I'm wondering for a couple of reasons: a) I'm curious about the role this mysterious component performs, and b) I'm wondering if drop-in replacement chips can be found on mouser/digikey for people who damage their U9 and U10 chips.

 

[YveltalGriffin]

The U10 is a supervisory IC that generates a delayed 3.3V signal. It sends the signal to the Wii's Hollywood, letting it know that it's good to go to boot up.

Edit 2024-05-13: My old explanation was flawed. For future travelers: U10 holds its output LOW to keep the Hollywood in reset, until the 3.3v rail rises above ~2.9V, at which point the output goes HI-Z (floating) and the Hollywood comes out of reset.

It's probably pretty similar (if not identical) to this voltage detector IC: https://fscdn.rohm.com/en/products/databook/datasheet/ic/power/voltage_detector/bd45xxg-e.pdf

Technically, the U10 could be replaced with a different voltage detector IC, a microcontroller, or even a physical switch. But the U9/U10 is the simplest and smallest solution that exists, so there's no real reason to replace it. Plus, you get two of them for free with every Wii!

 

[loopj]

Interesting, so it looks like a ROHM BD45332G (3.3v, 200ms delay, open drain) or ROHM BD46332G (3.3v, 200ms delay, CMOS) might be a drop-in replacement. I can't tell at a glance if the open drain or cmos model is the correct one here.

In my case, I damaged U9 and U10 by being too clumsy with my removal. I'm thinking spending a couple of dollars on some replacement chips is cheaper than me buying a new motherboard.

 

[jefflongo]

Interesting, so it looks like a ROHM BD45332G (3.3v, 200ms delay, open drain) or ROHM BD46332G (3.3v, 200ms delay, CMOS) might be a drop-in replacement. I can't tell at a glance if the open drain or cmos model is the correct one here.

In my case, I damaged U9 and U10 by being too clumsy with my removal. I'm thinking spending a couple of dollars on some replacement chips is cheaper than me buying a new motherboard.

I would wager probably CMOS because there's no pullups on the output. IIRC the function is to time a reset interrupt on the Hollywood after it has ample time (200ms) to power up. @Aurelio knows for sure I believe.

I would be interested to see if this would work as a drop in replacement. I think it actually could work.

 

[Aurelio]

On the Wii motherboard there's actually a 4.7k pullup on the u10 output, but when we trim the wii we get rid of it. This makes me think that even if there's a pullup, the IC they used is CMOS.

 

[loopj]

Thanks jeff, aurelio!

I'm going to buy a few BD46332Gs with my next digikey order and try one out (for science). Will report back.

 

[Stitches]

 

[superior1777]

@loopj any updates on the science experiment?

 

[loopj]

Look what arrived

Will try adding it to a fresh mobo.

 

[YveltalGriffin]

Look what arrived

Will try adding it to a fresh mobo.

I'm eager to see if you can get them working. I ordered a couple a few weeks ago and couldn't get them to work on untrimmed or trimmed mobos. A deeper read into the DS indicates that the ER line might need to be pulsed to start the POR timer, but it also says that you can tie ER to GND if it's unused (which is what U10 seemingly does)


The sheet also shows application circuits with ER grounded. But I couldn't get the ICs to output 3v3 at all, even just hooked up to my benchtop PSU. Probably just a dumb mistake on my end. Please keep us updated!

 

[loopj]

Just tried this out on an untrimmed 6 layer RVL-CPU-30.

Bad news, as YvenltalGriffin says it does not work as a straight swap out, the Wii does not boot.

I confirmed my wiring is ok by removing the chip and replacing with the stock U10, and the console boots.

 

[Stitches]

At least the science was done

 

[mokus]

There’s a decent chance that 3.3v threshold part is too high. I’m getting a parts order ready for something else and thinking I might throw in a 3.0V version. Then after a swap, I can test the original in isolation and see if I can figure out more specifics.

 

[mokus]

I ordered the parts but then I got impatient and pulled the original anyway. I can confirm my guess was correct - the original has a 3.0V threshold. Additionally, it is open-drain and has about 160ms delay.

edit: improved my timing measurement setup a bit. This particular unit has a very consistent 169.6 ms delay. I’ll know for sure in a few days but I’m very confident at this point that the BD45302 will be a drop-in replacement.

Final update - swapped the new parts in (replaced both U9 and U10 with BD45302G) and it works . Booted right up like nothing had changed at all.

 

[loopj]

Amazing, nice work mokus! I feel like I was vaguely part of history here

 

[rainwaffles]

Noob here, I can't find anywhere to post this question but does anyone have any advice on how to remove the U10 chip easier? I somehow managed to damage both of my chips when using my hot air station. I brought the temp up to 400+ but it wouldn't budge when I pulled on it with tweezers, and a bit of the black casing chipped off both of the chips. I swear it's a lot harder to remove than other circuit boards I've tried. Does using some cleaner or flux help?

I still want to remove the chips and check if they work anymore, but if they don't I guess I can try ordering the BD45302

 

[Gman]

Noob here, I can't find anywhere to post this question but does anyone have any advice on how to remove the U10 chip easier? I somehow managed to damage both of my chips when using my hot air station. I brought the temp up to 400+ but it wouldn't budge when I pulled on it with tweezers, and a bit of the black casing chipped off both of the chips. I swear it's a lot harder to remove than other circuit boards I've tried. Does using some cleaner or flux help?

I still want to remove the chips and check if they work anymore, but if they don't I guess I can try ordering the BD45302

Check this out! https://manual.bitbuilt.net/step/u10-relocation/6

 

[rainwaffles]

Thanks gman! I do have a hot air station, and I eventually realized I just set the air speed to be too low -_-

I have one more question- can anyone explain how I can test the u10 chip to make sure it still works? I have a power supply and multimeter. Are there any other tools I would need?

Thanks again, I'm new to hobby electronics so this is a big learning experience

 

[loopj]

For future travelers, I think I found the original U9/U10 IC - the Ablic S-80129CNMC-JKO.

The JKO matches the part markings on the U9/U10, the pinout matches, and the timings and voltage thresholds line up with what we expect (2.9V, ~200ms).

This is an obsolete part, but hopefully it helps people find a suitable replacement - the Rohm BD45292 matches the specs (rather than the BD45302).

 

[Maxmo]

For future travelers, I think I found the original U9/U10 IC - the Ablic S-80129CNMC-JKO.

The JKO matches the part markings on the U9/U10, the pinout matches, and the timings and voltage thresholds line up with what we expect (2.9V, ~200ms).

This is an obsolete part, but hopefully it helps people find a suitable replacement - the Rohm BD45292 matches the specs (rather than the BD45302).

I tested it with BD45292G-TR and it works. Thank you!