SNES chip failures

 

It’s all so tiresome.

I do truly enjoy repairing game consoles. The SNES was the first console I owned myself so it has a special place in my heart. I love repairing them, cleaning them, and getting them ready to be used again. Sometimes I think there’s nothing more rewarding than that. But other than the common broken power jack issue, chip failures are the main problems with Super Nintendos. Every so often I’ll get one that needs a recap, but chip failures are by far the most common problem besides broken power jacks. But no matter how rewarding it may to fix these things, it eventually becomes depressing endlessly removing one dead chip after another.

Failed SNES CPUs.

Bad PPUs aren’t as common as bad CPUs, but they’re by no means rare, either.

There are 4 revisions of PPU2. S-PPU2, S-PPU2 A, and S-PPU2 B are the three most likely to fail, at least from my small sample size. S-PPU2 C and CPU-B seem to be hardier than the earlier revisions.

The main chip failure on 1-CHIP and SNES Mini consoles is the S-APU.

I often hear people say bad CPUs on Super Nintendos are a rare issue. But it sure doesn’t seem that way to me. Maybe a quarter of SHVC consoles I get have a bad PPU or CPU. Of course, I’m getting consoles that have already been identified as broken, but still. People say it’s not an epidemic on the same level as, say, bad capacitors on PC Engine Duos or Game Gears, but I don’t think it’s unreasonable to assume that all the old revision PPUs and CPUs in these Super Nintendos will go bad in the not-too-distant future.

No more ALPS joysticks on recent JDM-055 Playstation 4 controllers?

This JDM-055 motherboard came from a broken PS4 controller I bought on eBay.

A few weeks ago I bought a box of broken PS4 controller parts on eBay. There were a few motherboards and some miscellaneous housing bits in the box. It was a good deal. But what surprised me the most was this board inside that had black joysticks with orange potentiometers.

The right stick on the board was cracked, but that wasn’t the interesting part. What was more interesting is that there were no signs whatsoever that anybody had worked on the board before. The controller had been opened up and there was some dust and dirt on the board, but no signs that anyone had touched anything at all with a soldering iron.

No signs that these were soldered in by hand.

These orange joysticks are readily available all over Aliexpress, eBay, Amazon, etc… so it’s not uncommon to find them inside boards that have been worked on before. From the “F” logo on the sticks I believe they’re made by Polyshine/Favor Union, which is the same company that makes the flexible conductive film for the buttons. You can even see the “F” logo on the conductive film in the photo at the top of the page, albeit upside-down.

This board didn’t look like it had been worked on before by anyone. All the solder joints for the joysticks looked original. Nobody had removed the original ALPS sticks, installed these black and orange ones, and then subsequently broken the right stick. This board had simply never had ALPS sticks in the first place. These orange ones were factory original.

I was so curious about this that I actually asked the seller if he knew anything about it. He confirmed for me that he hadn’t replaced any of the sticks himself and had purchased the lot from a wholesale seller. He said the board was like this when he received it.

I was mystified. I was dying of curiosity, but had no answers. The seller had said he believed the wholesale vendor he bought it from was selling Walmart customer returns. I wondered if maybe, like Gamestop does with some products, Walmart had refurbished the controller and used aftermarket parts. Still, it didn’t look like it had been hand-soldered, so it seemed unlikely to me.

Just today I received another ZCT2U controller with a JDM-055 board in it that I bought on eBay. Interestingly, it too has orange and black Polyshine/Favor Union joysticks in it. It also has no signs at all of any rework.

And the serial number, for those who want to try to find one in the wild.

I think at this point I can be reasonably confident that I’ve got my answer. These controllers were not refurbished. Neither Walmart nor any other retailer had anything to do with it. Sony has simply stopped using ALPS joysticks in their controllers and switched to using Polyshine/Favor Union joysticks. Or at least, that’s the case for some of the newest ZCT2U controllers. Maybe it’s just the ones with 1-982-707-21 as the part number?

I’ve worked on plenty of JDM-055 boards that have the usual red, green, and white ALPS joysticks in them, so it’s not the case that all JDM-055 boards are like this. But it looks like maybe the most recent revision uses these new joysticks. They’re almost identical to the similar-looking ones you can find on Aliexpress, but the center button (i.e. L3 or R3) makes the same kind of clicking noise the ALPS ones make. This is in contrast to all the orange and black ones I’ve bought on Aliexpress in the past, most of which have made a noticeably louder clicking sound than the ALPS ones.

Sadly, the controller I received today doesn’t seem to power on. The seller said it was connected to a faulty charger which caused the battery to burn up. That seems to be a common issue, but I’ve seen it most often on JDM-011 and 020 boards. I’ve had plenty of those with burnt up charge ports and melted plastic from the BD9200 chip on the board failing and overheating, presumably due to the controller getting plugged into the wrong kind of charger. But usually those at least power on. They overheat almost immediately, but they do at least come on briefly, whereas this one isn’t even detected by my computer at all. Maybe if I get some time one day I might mess around with it and see if I can figure out how to fix it.

Regardless of whether I fix this thing or not, I think it’s incredible that from the looks of it Sony has dumped ALPS. If I’m not mistaken, Sony controllers have had ALPS joysticks in them since the original DualShock controllers for the Playstation, which means they’ve been together for 25+ years. If Sony and ALPS have parted ways, then this is like some kind of huge celebrity breakup. It’s tragic, in a way. But exciting, too, particularly for modders and tinkerers, since Favor Union joysticks are usually easier to find and not faked anywhere near as much as ALPS sticks are.

A few more tips about black screen SNES consoles

I’ve been getting some SNES Mini and 1-chip consoles to repair lately. The most common issues are:

  • Black screen on all games
  • Black screen on some games. Other games, such as Mario Kart, freeze at the publisher splash screen

The latter issue is exactly the same problem as what happens when you simply remove the sound module from an SHVC console. It’s seems to be caused when the CPU can’t communicate with the sound hardware for any reason. In some cases, it’s due to some broken traces on the board somewhere. But as I’ve mentioned before, it’s often due to a failed APU. On GPM-01/02 and RGB-01/02 revisions it can also be due to a failed S-SMP.

    With a bad APU or sound module, the console freezes here on Mario Kart.

With a bad APU or sound module, the console freezes here on Mario Kart.

Or here on many Koei games, like my test game, Super Mahjong Taikai

Or here on many Koei games, like my test game, Super Mahjong Taikai

The first problem, however, is something that I can’t normally fix. Unless there’s some obvious problem, such as rust or corrosion in the cartridge slot or some broken traces, then I just give up on those boards and set them aside.

Something I discovered quite by accident, however, is that a bad APU can actually cause the first problem as well as the second.

Recently I grabbed one of those black screen SNES mini boards to steal its APU for a 1-chip board I was repairing that had a known bad APU (it would freeze at the “KOEI” screen). Once I installed the APU from the black screen SNES mini into the 1-chip board, the 1-chip board also began displaying only a black screen, even on games that were previously loading to the publisher screen and then freezing. At first I thought I must have accidentally left a solder bridge connecting some legs of the chip or screwed up in some other way with bad soldering, but after checking and rechecking, that wasn’t the issue. I then removed the APU and the board started loading games and then freezing again at the publisher screen, just as it had been doing originally.

To test my theory that the APU itself was causing the black screen issue, I went over to the SNES mini board from which I had salvaged the APU. I powered it on without any APU installed at all and, sure enough, it would load games up to the publisher’s logo and then freeze.

To be certain, I took one more known good 1-chip board that was only missing an APU and installed the APU from the Mini board into it. It started showing just a black screen when I installed it. Once removed, it again loaded games and froze at the publisher logo.

Finally, I went over to my stack and found two 1-chip boards that I had labelled “black screen”. I removed the APUs from both of them. One of them started loading games and freezing at the publisher screen, showing that it had the same problem as the Mini board (i.e. just a bad APU). The other board, unfortunately, still just showed a black screen. It must have had some other issue.

But the major point here is that when troubleshooting boards that show only a black screen, it’s possible the APU is actually causing the issue. I used to think the only issue caused by a bad APU was the loading and then freezing at the publisher splash screen issue, but this experience taught me that I was wrong.

Maggots? In my SNES?

It's more likely than you think.

It’s more likely than you think.

No, don’t worry, this isn’t my personal console. I got this one on eBay. Mine, of course, is immaculate, as you would expect. Incidentally, buy only from me or else this is the kind of thing that’ll show up at your doorstep after shopping online.

But wouldn’t you know it? This disgusting pile of filth and disease worked fine after I spent several hours washing it. I had to replace the capacitors, but otherwise it was fine. It’s aggravating that horrific-looking systems like this survive despite such neglect and abuse and yet so many well-cared for pristine-looking systems stored in safe and clean conditions by responsible, civilized people just die for no good reason.

DSCF2842

DSCF2843

DSCF2851

DSCF2848

DSCF2847

DSCF2846

DSCF2845

Unusual original Nintendo 3DS motherboard revision: CTR-CPU-40

DSCF9148

I received a 3DS console today that had been cracked in half at the hinge. As I opened it up to remove the motherboard I immediately noticed that the tiny little IR board that normally connects to a plug next to the P7 connector was missing. The IR module on this motherboard was integrated and not removable. Also strange was the fact that the little piece of tape used at the factory to secure the touch screen connector was black, rather than the typical white. I’ve seen a few boards with black tape rather than white, but in the past I always assumed this was because they had already been worked on before I received them and the last technician had used black tape. Now I know I was probably wrong. This was an entirely new motherboard revision.

The back of the board

The back of the board

Integrated IR

Integrated IR

There are some minor yet consistent differences I’ve noticed from one console to the next, such as the color of the mainboards, color of screws, slight variations in the labeling on the WiFi board, and the fact that most special edition consoles and consoles bundled with a game have the charge ports soldered much more securely to the board than the original black/red/blue models. Normally, however, the motherboard itself is basically the same. This is different from those types of minor variations because it’s a completely different motherboard revision from what I’ve seen before. Though it’s not something I normally pay attention to, I don’t think I recall ever seeing anything other than CTR-CPU-01 and the occasional CTR-CPU-20 boards.

Big blobs of solder secure the charge port in place, unlike the weak connection in earlier models

Big blobs of solder secure the charge port in place, unlike the weak connection in earlier models

The label on the WiFi daughterboard differs slightly from most models, but the board itself is the same DWM-W082 as all 3DS consoles.

The label on the WiFi daughterboard differs slightly from most models, but the board itself is the same DWM-W082 as all 3DS consoles.

In any case, this board does look more or less the same as any other model besides the fact that it has integrated IR and some silver screws in a few locations that normally have black screws. The color of the screws does differ from model to model, but I’ve never seen silver screws in these locations before. It’s fairly common for the external screws for the housing to be silver on special edition or bundled consoles, but I’ve never seen anything other than black for the screws beneath the SD card slot. Additionally, some special edition and bundled consoles have only two, rather than three, screws securing the L button in place and only three, rather than four, screws for the R button. This console follows that pattern but it also replaces the normal black screws for the shoulder buttons with silver ones.

Silver, rather than black, screws beneath the SD card slot.

Silver, rather than black, screws beneath the SD card slot.

Silver, rather than black, screws for the shoulder buttons as well

Silver, rather than black, screws for the shoulder buttons as well

Unfortunately, the previous owner used super glue in a misguided attempt to repair the cracked housing and even glued the battery cover in place, leaving no possibility of salvaging it. But the serial number sticker inside was still legible:

DSCF9143

For now I’m keeping this board. I put it inside one of my own consoles and sold the motherboard it replaced. I’ll probably keep it around for a while to tinker with and maybe I’ll eventually sell it.

Quick and easy fix for a Super Nintendo suffering from “black screen of death”

DSCF7636 - Copy

One of the most common problems people seem to have with Super Nintendo consoles is they find that the console powers on but only displays a black screen with no audio, even with known good games. I don’t have a tremendous amount of experience repairing Super Nintendos, but it is something I do occasionally for fun and because I find that I usually learn something in the process. I generally stay clear of consoles with this type of problem though, since it could be caused by just about anything. Determining the exact cause can be next to impossible. Normally these boards look perfectly fine visually, so figuring out where there’s a broken trace can take forever. They’re usually not worth repairing. Sometimes, however, you get lucky and there’s something very obviously wrong that you can see visually. It still may take some careful inspection with a magnifying glass, but if there are any signs of corrosion or other damage, it’s always worth trying to fix it, even if it doesn’t look that bad visually or strike you as something that’s likely to be the cause of your problem.

You can see here I've put down a bit of solder along a trace running near U15. This trace had a lot of corrosion on it and appeared broken visually. It is so tiny that, rather than fix it with a wire, I just lay down a bit of solder along the broken length of it to repair it. In the end I determined that it was not, in fact, the cause of the problem

You can see here I’ve put down a bit of solder along a trace running near U15. This trace had been broken due to corrosion, but it was so tiny that, rather than fix it with a wire, I just lay down a bit of solder along the broken length of it to repair the problem.

I wish I had taken a photo prior to the repair, but the opening photo up above shows the area where I found damage on this GPM-02 board. There are two audio RAM chips in the opening photo. U15, shown here to the left, is the one where there was a problem. Pin 12 of U15 had a bit of corrosion on it. I checked for continuity between it and the via it was going to and the connection was not totally broken, but it wasn’t exactly sound, either. If I fiddled with the multimeter probes I would get continuity, but it wasn’t consistent. Something told me to try using a wire to connect pin 12 directly to that via, just to see if it would help things. I put a bit of flux in the via, insert the tinned end of a small wire into it, crossed my fingers, and hoped that the solder would flow into the via and secure the tip of the wire in place. Fortunately it worked out as I had hoped and I was able to solder the other end of the wire to pin 12.

Without really thinking there would be any improvement, I went ahead and tested the console. To my surprise my test game worked fine. I then tested with about a dozen other games and they all worked. I was very surprised that this broken trace would cause a black screen for all games. Since mostly audio-related stuff goes on in this area of the board, I would have expected that symptoms of this type of damage might have been games playing but without any audio.

It's not pretty, but it works.

It’s not pretty, but it works.

In the end, I’m not sure if the damaged trace running near U15 actually had anything to do with the black screen problem. It was the first thing I noticed though when I opened the console, so I thought scraping away the corrosion and patching it was worth a shot. Fixing it alone didn’t solve the black screen problem though. Before reassembling I tested the console without the wire from pin 12 to the corresponding via and was able to reproduce the black screen problem, so a bad connection to/from pin 12 was clearly responsible for the issue. The other damaged trace may not have even been bad enough in the first place to cause a problem.

Though it looks pretty sloppy, when reassembling the console I simply placed a bit of electrical tape over the blobby length of solder on the patched trace. The console has been working fine for some weeks now, so it seems nothing is shorting.

Unresponsive 3DS face buttons?

This is the same console owned by the blighted mudcrab whose shoulder buttons I had the pleasure of cleaning previously.

This is the same console owned by the blighted mudcrab whose shoulder buttons I had the pleasure of cleaning previously.

While unresponsive shoulder buttons are far more common, sometimes the face buttons of a 3DS or other console become sticky or unresponsive. If the button works when pressed very hard but not when pressed lightly, it’s almost certainly due to dirt accumulation. If the button feels sticky or jammed, it’s most likely because there’s dirt built up around and beneath the plastic buttons, as in the above photo where thick rings of filth have built up around the circumference of each button. If the button doesn’t feel jammed but it’s not as clicky as it once was, it may be due to dirt under the conductive pads on the motherboard. In either case, it’s necessary to remove the motherboard from the housing (or at least remove the screws securing it in place and flip it over, if you’re in a hurry, like I was).

Here's the rubber pad between the plastic buttons and the contacts on the motherboard. Doesn't look too bad, does it?

Here’s the rubber pad between the plastic buttons and the contacts on the motherboard. Doesn’t look too bad, does it?

But peel it away to reveal a cache of shit and grease.

But peel it away to reveal a cache of shit and grease.

Remove the buttons one by one and clean the slots. Toothpicks help.

Remove the buttons one by one and clean the slots. Toothpicks help.

Clean the buttons, too. As you can see, they're usually even dirtier than the slots in which they sit.

Clean the buttons, too. As you can see, they’re usually even dirtier than the slots in which they sit.

Of course, while you’re at it you should also clean the D-pad and power buttons. It’s the same nauseating process as for the ABXY buttons, so I didn’t bother photographing them. When you have to disinfect your camera after each part of the process, you’ve got a strong incentive only to photograph the most essential steps.

Don’t close up your console yet. The most important part is to clean the contacts on the motherboard. If you’re a particularly depraved slimebucket—or if the console has suffered liquid damage—you should clean beneath the contacts as well.

As with most things, these contacts don't look too bad from afar.

As with most things, these contacts don’t look too bad from afar.

But look more closely and you'll be horrified.

But look more closely and you’ll be horrified. We need to clean this sludge off with isopropyl alcohol and a cotton swab.

For the worst cases you will need to clean beneath the contacts as well. Mainly this is only necessary when there was liquid damage, but in this particular case the owner was such a slovenly clodhopper that there was a horrifying amount of dirt underneath the contacts even in the absence of anything else that looked like liquid damage.

You can peel the contacts up with a toothpick, but I actually find it easier to use my fingernail. As long as you’re careful, you can just stick them back down on the board when you’re finished cleaning. The adhesive is very strong, so it can be reused. However, if you do accidentally destroy the contacts you can replace them either from another motherboard or by buying replacements on eBay or Aliexpress for about $2, so there’s no great risk involved here. Just don’t scratch the motherboard.

It's amazing the D-pad worked at all with this level of accumulation.

It’s amazing the D-pad worked at all with this level of accumulation.

When cleaning it's fine if you don't remove 100% of the dirt.  It's not necessary to be perfect.  In this case the contacts are clean enough to restore completely normal functionality even though there is still a small amount of stubborn dirt left that I couldn't remove.

When cleaning it’s fine if you don’t remove 100% of the dirt. It’s not necessary to be perfect. In this case the contacts are clean enough to restore completely normal functionality even though there is still a small amount of stubborn dirt left that I couldn’t remove.

And that’s it. If people weren’t such swine none of this would be necessary.

Unresponsive 3DS shoulder buttons?

Well, have maybe you're a swamp dwelling slob, like the previous owner of this console.

Well, have maybe you’re a swamp dwelling slimelord, like the previous owner of this console.

The above is about average for consoles I receive. I’ve seen much worse, but it never ceases to surprise me how much dirt can accumulate inside consoles that look to be in decent cosmetic condition externally. This is the first time I had the idea to actually photograph it though. I doubt most people realize how much material gets transferred from their hands to the console and becomes trapped inside when they fail to wash their hands before playing.

I always clean beneath the shoulder buttons before selling a used console. I used to only do this for liquid-damaged consoles, but I’ve noticed it’s often necessary even on consoles that look like they’ve been taken care of responsibly. So now I don’t even bother testing the shoulder buttons first to see if they need it. I just assume they do and clean them.

Unresponsive shoulder buttons are almost always due to dirt accumulation. It’s really amazing what slobs people are. Fortunately, cleaning the shoulder buttons is very quick and easy, so even destitute mud farmers like the people who let this happen to their consoles can fix it themselves.

The left and right sides are pretty similar, so I only took photographs of the right side to show the process.

This is what we will be contending with.

This is what we will be contending with on the right side.

The left is not much better.

The left is not much better.  


First we remove these four black screws. On some consoles there will be only three.

First we remove these four black screws. On some consoles there will be only three.

Now we life the plate off and set it aside.

Now we life the plate off and set it aside.  

Next we life off this black plastic piece.

Next we life off this black plastic bracket.  

The button is held in place by a peg and a spring. Carefully remove them.

The button is held in place by a peg and a spring. Carefully remove them.

This would be the moneyshot.

This would be the moneyshot.

The best way to clean is with isopropyl alcohol. Not only does it make removing even tough dirt easy, it also eliminates most odors, which may be important if there’s liquid damage. You can do a good job with just a combination of a toothbrush, toothpicks, and some cotton swabs. Obviously you’ll want to focus your cleaning on the recess in which the plastic button sits, but you should also clean the button itself. If you’ve noticed that your shoulder buttons are unresponsive and don’t make as much of an audible “clicking” noise as they once did when you press them, cleaning them can sometimes help with that

Now we clean and dry the left side...

Now we clean and dry the left side…

and the right side.

and the right side.

While you have the console open, it’s also a good idea to clean out the area beneath the SD card slot and the slot in which the stylus is stored.

Remove the four silver screws that secure the stylus holder and SD card slot cover in place.

Until you open it up to check, you would never suspect how much dirt can be trapped in these areas.

Remove the four silver screws that secure the stylus holder and SD card slot cover in place.

Remove the four silver screws that secure the stylus holder and SD card slot cover in place.

Disgusting.

Disgusting.

Much better.

Much better.

Reassembly is easy as long as you haven’t lost anything along the way.

Make sure these two nuts are in there. The four screws on the battery cover screw into these.

Make sure these two nuts are in there. The four screws on the battery cover screw into these.

First, place the button into the recess.

First, place the plastic part of the button into the recess.

Then insert the peg...

Then insert the peg…

...and the spring.

…and the spring.

Now we can insert the button itself.

Now we can insert the button itself.

Next we replace the black bracket on top of the button. The ribbon cable for the button goes through a slot in the bracket.

Next we replace the black bracket on top of the button. The ribbon cable for the button goes through a slot in the bracket.

Replace the serial number plate and screw it back in. Take note that no screw goes in the upper left corner.

Replace the serial number plate and screw it back in. Take note that no screw goes in the upper left corner.

Here's what the left side should like like in the end.

Here’s what the left side should like like in the end.

The finished, disinfected product.

The finished, disinfected product.

If I get around to it I might write about how to repair most cases of sticky/unrepsonsive face buttons next. If I get around to it.