|
| 1 | +--- |
| 2 | +layout: post |
| 3 | +title: "Water-damaged Game Boy Color: Fixing the Speaker Output" |
| 4 | +date: 2021-01-18 12:30:00 -0500 |
| 5 | +categories: consoles gameboy repair |
| 6 | +--- |
| 7 | + |
| 8 | +In my [previous post]({% post_url 2021-01-15-water-damaged-gbc %}), I cleaned up the button contacts on a seemingly water-damaged Game Boy Color. Everything seemed to be working except for the fact that there was no audio from the built-in speaker. Well, today I figured that out. |
| 9 | + |
| 10 | +## Does the audio work at all? |
| 11 | + |
| 12 | +I have had other Game Boys where the onboard speaker does not output sound, but headphones work, so that's what I tried first. I plugged some speakers into the headphone port on the bottom of the GBC, and the sound worked! That was a huge relief. I knew that the headphone port has a switch inside it that is normally `on`. When headphones are plugged in, it changes the switch to the `off` position. So there was most likely something wrong with that switch. |
| 13 | + |
| 14 | +## Cleaning the Switch |
| 15 | + |
| 16 | +I wasn't sure where exactly the switch was located inside the headphone port, so I did some digging and found [this excellent post on Sam's Asylum](http://samzasylum.blogspot.com/2015/05/gameboy-color-speaker-repair.html) that showed exactly where to clean. Luckily, the designers of this headphone jack left a little window in the case to allow me to clean the switch without desoldering it from the board. |
| 17 | + |
| 18 | + |
| 19 | + |
| 20 | +I inserted a headphone cable to open the switch and make the contacts accessible. Using a set of precision tweezers, I gently scraped off the oxidation from both sides of the switch. When I was done, I used some canned air to blow out any oxidation dust and put the GBC back together enough to test it. Aaaaand the speaker still didn't output any sound. |
| 21 | + |
| 22 | +## Ok, how does this circuit even work? |
| 23 | + |
| 24 | +At this point, I knew I had to start poking around with my multimeter. I found this extremely detailed [schematic for the GBC](https://console5.com/techwiki/images/e/e6/Nintendo_GBC_Schematic.png) from Console5 and dove in. |
| 25 | + |
| 26 | + |
| 27 | + |
| 28 | +I've cropped the schematic to focus on the headphone jack, which is marked as `P5` on the right side of the image above. The red arrow points to pin 5, which is is the side of the switch that connects to pin 6 (`SW`) of the sound amplifier `AMP-MGB` (`U3`) on the left side of the image. Pin 4 of `P5` is connected to ground. When there are no headphones connected, the switch is closed, connecting pins 4 and 5 of `P5` together, which connects the `SW` pin of `U3` to ground. |
| 29 | + |
| 30 | +You may have noticed that I'm ignoring `EM4` (part number [BLM11B102S](https://docs.rs-online.com/ae7f/0900766b80021680.pdf)), which is in-between pin 5 of `P5` and the `SW` pin of `U3`. This is just a ferrite bead, and is most likely there to filter out noise. It's good to have, but not absolutely necessary. |
| 31 | + |
| 32 | +## Probing |
| 33 | + |
| 34 | +I set my multimeter into continuity mode and found the `SW` test point on the Game Boy's PCB. I touched one probe of the multimeter to that, and the other probe to pin 5 of the headphone jack (the red arrow in the image below). Unfortunately, I did not get a tone, so there was a break somewhere in the trace (the yellow line in the image below). |
| 35 | + |
| 36 | + |
| 37 | + |
| 38 | +Fortunately, I quickly found the break. I probed both sides of EM4, which showed continuity, so I tested both sides of the circuit from EM4. However, there wasn't continuity on either side! I wondered if the oxidation got under the pads on the PCB and lifted them up, so I scraped off some of the coating to reveal the traces on either side of EM4. I tested the continuity from one side to `SW`, and there was continuity. Likewise with the other side and pin 5 of the headphone jack. I knew my suspicions were correct. |
| 39 | + |
| 40 | +## Fixing the Problem |
| 41 | + |
| 42 | +I tried to reflow the solder of `EM4`, but the component was so light that it lifted off the board as soon as I touched it with the iron. I tried to bridge the trace and the pad with solder, but I wasn't able to do it. In the end, I decided to bypass `EM4` entirely with a piece of very thin (bodge) wire connected between the `SW` test point and pin 5 of the headphone jack. I was careful to route it around the button pads, cartridge connector, and screw holes. |
| 43 | + |
| 44 | + |
| 45 | + |
| 46 | +And... it worked! The audio comes out of the onboard speaker with no issues, and switches off when headphones are plugged in. I normally don't like to bypass filters like that, but I'll address that if the audio quality is a problem in the future. In the meantime, my wife's coworker can show this off to her daughter! |
| 47 | + |
| 48 | +<video controls> |
| 49 | + <source src="/assets/video/water-damaged-gbc-repaired.mp4"> |
| 50 | + Your browser does not support the video tag. |
| 51 | +</video> |
0 commit comments