Pi in a Gameboy Advance Build - WIP

backstander

@obsidianspider
Wow! Now that's progress!
I love the animated GIF haha!

obsidianspider

It's Friday night, so that means it's time to cut some stuff up in the garage.

I was trying to fit the Pi 3 inside the Gameboy Advance game that I bought from eBay (Is Madagascar any good? I still have the PCB in good shape.) and I determined that I would need to make the Pi narrower, and also a bit shorter. I decided that I would only really at most need two USB ports (Maybe one for USB audio (I haven't determined if I can live with the on board audio from the Pi) and maybe one for a USB port for plugging in a keyboard or something.) so I could trim the end of the Pi down a bit more.

Yes, it still works. I think. It powers up, no funky red LED flashing and I can SSH to it.

The Pi now fits inside the Gameboy Advance case with the end covered by a trimmed and gutted Gameboy Advance game.

Its looking pretty good in the case with the control PCBs

Next up I'm going to try connecting some things to the Pi 3.

obsidianspider

I got the TFT soldered to the Pi and things seem to work.

They seem to fit well in the case, but I don't have anything but the Pi and screen in the case in this mockup.

obsidianspider

I don't care if it's almost 11 PM, I'm really encouraged, so I decided to try hooking up a USB port to the remnants of the board, and it works!

moosepr

@obsidianspider awesome stuff! It's really coming together now!

obsidianspider

@moosepr It's definitely not a project for the weak willed. I'm still not sure where the heck I'm going to possibly put a USB port, but today I'm going to work on connecting leads to the USB pads and then will probably work on the audio circuit. I'm not sure about how I'm going to do the battery thing yet. I've seen posts in various places about a low voltage smart shutdown and things like that. Apparently if you over-discharge a LiPo you'll kill it.

First though I'm in a parade this afternoon so no time to fuss around in the garage where I tend to lose track of time. ;)

moosepr

@obsidianspider yeah lithium batteries don't take too kindly to being really flat. I think the adafruit boost doodad has protection built in. If your not using that then there are other options. Some cells have built in protection, or you can add an extra protection board

obsidianspider

Today I removed the USB port from the Pi and soldered up the appropriately colored wires. I applied some hot glue over the wires for strain relief. I sliced off the excess since I needed the board to fit inside the Gameboy Advance game.

When I tried to fit the Pi inside the gutted cartridge I found that I still needed more room as the wires and glue made the Pi too thick. I removed the top of the "front" part of the game and I also shaved the area for the screw hole flat on the inside.

Then with that going well I soldered up the test USB port to the end of the wires, and that worked, so then I tried desoldering the plug from my USB sound card. That didn't go so well. When I was putting the board in my PCB vice I heard a crack, and sure enough, the capacitor for the headphone plug ground cracked off. I thought that maybe it wouldn't be a huge deal, so I soldered across the pads to make the connection to ground. I figured that headphones can't draw that much, so eliminating the capacitor may be ok. Well, it wasn't. After I soldered up the sound card to the Pi I was getting a blinking red light and the Pi would not boot.

Just in case it was my USB leads and not the sound card I tried connecting up a USB port and keyboard to the Pi along with a controller, and sure enough, everything booted just fine. That said, when I plug in a USB drive, the Pi won't detect it, and I'm not sure why. The drive does work in my Genesis hub Pi 3, so I don't think it's a Raspbian issue. I am thinking that there may not be enough power getting to the USB port, so I'm going to try an alternate power source like straight from GPIO and see what happens.

I put the GBA game on the end of the Pi and it runs just fine, so even smooshed in the cartridge, no bad contacts.

0_1489272897773_pi3-usb-leads-connected-to-controller.jpg

obsidianspider

Well, I'm thinking maybe it isn't a power issue. I tried going to GPIO for power and then I started getting lightning bolts on the screen, so instead I used a breakout board to solder up the USB power directly to the 5V in that I'm using to power the Pi and my USB drive still isn't being recognized. The controller and keyboard both work just fine, but those also aren't using USB 2.0 speeds. (Per lsusb -t they communicate at 1.5 Mbit)

I also tried plugging in a USB 2.0 hub. That's not recognizing, a USB Wifi adapter isn't recognizing, and no USB drives are recognizing. The USB drive has a steady blink once per second or so.

I'm not convinced I didn't mess something up with the USB ports when cutting things down, or maybe connecting up that damaged sound card fried something.

obsidianspider

I tried going with the "nuclear option" to get the drive to read by adding dwc_otg.speed=1 to /boot/cmdline.txt and after booting the drive is recognizing, but then the controller wasn't. Thinking maybe I fried the controller with all of my power testing I tried plugging the controller into my laptop and it worked just fine, so lowering the USB speeds seemed to break the controller functionality.

Removing the "nuclear option" and rebooting caused the drive to no longer be recognized, but my controller started working again.

I read somewhere that data lines for USB can be touchy and I saw somewhere that they should be twisted so you don't have issues. I'm not sure if something like that will help or not.

obsidianspider

It looks like twisting the data wires worked. Some searching around online showed that in USB cables the data lines are usually a twisted pair in a USB 2.0 cable, so I tried it.

Plugging in two USB drives allows both of them to show up with no issues.

$ lsusb -t
/:  Bus 01.Port 1: Dev 1, Class=root_hub, Driver=dwc_otg/1p, 480M
    |__ Port 1: Dev 2, If 0, Class=Hub, Driver=hub/5p, 480M
        |__ Port 1: Dev 3, If 0, Class=Vendor Specific Class, Driver=smsc95xx, 480M
        |__ Port 2: Dev 10, If 0, Class=Mass Storage, Driver=usb-storage, 480M
        |__ Port 4: Dev 11, If 0, Class=Mass Storage, Driver=usb-storage, 480M

$ lsusb
Bus 001 Device 011: ID 058f:6387 Alcor Micro Corp. Flash Drive
Bus 001 Device 010: ID 0781:5406 SanDisk Corp. Cruzer Micro U3
Bus 001 Device 003: ID 0424:ec00 Standard Microsystems Corp. SMSC9512/9514 Fast Ethernet Adapter
Bus 001 Device 002: ID 0424:9514 Standard Microsystems Corp. 
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub

 $ df -h
Filesystem      Size  Used Avail Use% Mounted on
/dev/root       7.3G  2.1G  4.9G  31% /
devtmpfs        364M     0  364M   0% /dev
tmpfs           368M     0  368M   0% /dev/shm
tmpfs           368M  5.2M  363M   2% /run
tmpfs           5.0M  4.0K  5.0M   1% /run/lock
tmpfs           368M     0  368M   0% /sys/fs/cgroup
/dev/mmcblk0p1   57M   21M   37M  36% /boot
/dev/sda1       2.0G  1.3M  2.0G   1% /media/usb0
/dev/sdb1       965M  514M  451M  54% /media/usb1

I also tried plugging and unplugging a number of times, and also trying to plug and unplug my USB controller and a drive in and out and things seem to be ok.

$ lsusb -t
/:  Bus 01.Port 1: Dev 1, Class=root_hub, Driver=dwc_otg/1p, 480M
    |__ Port 1: Dev 2, If 0, Class=Hub, Driver=hub/5p, 480M
        |__ Port 1: Dev 3, If 0, Class=Vendor Specific Class, Driver=smsc95xx, 480M
        |__ Port 2: Dev 16, If 0, Class=Mass Storage, Driver=usb-storage, 480M
        |__ Port 4: Dev 15, If 0, Class=Human Interface Device, Driver=usbhid, 1.5M

$ lsusb
Bus 001 Device 015: ID 0583:2060 Padix Co., Ltd (Rockfire) 
Bus 001 Device 016: ID 0781:5406 SanDisk Corp. Cruzer Micro U3
Bus 001 Device 003: ID 0424:ec00 Standard Microsystems Corp. SMSC9512/9514 Fast Ethernet Adapter
Bus 001 Device 002: ID 0424:9514 Standard Microsystems Corp. 
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub

$ df -h
Filesystem      Size  Used Avail Use% Mounted on
/dev/root       7.3G  2.1G  4.9G  31% /
devtmpfs        364M     0  364M   0% /dev
tmpfs           368M     0  368M   0% /dev/shm
tmpfs           368M  5.2M  363M   2% /run
tmpfs           5.0M  4.0K  5.0M   1% /run/lock
tmpfs           368M     0  368M   0% /sys/fs/cgroup
/dev/mmcblk0p1   57M   21M   37M  36% /boot
/dev/sda1       2.0G  1.3M  2.0G   1% /media/usb0

I'm really glad that it was an issue of the data wires needing to be twisted and not something I fried on the Pi.

While I was trying various things and moving power around I removed the micro USB port for power. I had the same problem as the HDMI port with the surface mount contacts and the through hole pads being too hard to heat up all at the same time. I ended up messing up some more traces down there. I really need to learn a better way to do that, but everything works.

That's enough for today.

Tomorrow I'll do some messing around with the onboard audio to see how crummy it is, or to see if I need to shop for another USB sound card and not crack a capacitor off.

moosepr

@obsidianspider I was going to say try twisting the data lines but you beat me to the punch! I think is so that any interference hits both wires together and kind cancels itself out. Or maybe USB is just magic, and the twisting helps make the magic flow

obsidianspider

@moosepr From what I read online it's to reduce/eliminate crosstalk. Kind of like in an Ethernet cable.

obsidianspider

I broke the spare, old USB sound card I originally intended to use when putting it in my PCB vise last weekend, so this week I ordered a replacement from Amazon. It arrived today, and wow, this thing is small.

It works well, and sounds better than the old one that it's replacing, so that's good. I'm not sure if the old one caused this, but I noticed an annoying message when launching games:

lvl0: VolumeControl::init() - Failed to find mixer elements!

Some googling around leads me to believe that it's a common thing and "that's just the way it is" unless you want to rebuild EmulationStation from source, which I don't.

I saw that the RetroPie wiki had a step that was supposed to address the issue, but it didn't, and frankly, step 5 didn't make sense to me, so I also added the method for enabling USB audio that I found over at sudomod as I think it is clearer and makes more sense to a novice, like me, than the other method that is in the wiki. I'll let the powers that be decide if the "alternate method" should be the preferred one. I still feel like I don't have the authority to make a decision like that.

Now that I know the sound card works, I tried taking the shell off, and it's made really well. Some prying with xacto knives didn't have it budge, so I'm going to see if I can get it apart without breaking it.

obsidianspider

Some repeated scoring with an xacto knife along the seam of the plastic got me inside, and this is a very dense device.

The headphone and microphone jacks are surface mount, which I stink at removing, and the usb plug is surface mount with through hole supports. I mangled the HDMI port and Micro USB on the Pi 3 when I removed those, but on this thing, I can't do that. Not to mention that this is very small, and I'm sure it will load up with heat quickly. I'm not going to attempt to remove anything tonight as it's getting late and I don't want to mess it up. If anyone has any tips for desoldering that USB port, or for removing the headphone and mic jack without heat soaking the whole board and messing something up, please chime in.

I made sure to test the board with the shell off before trying to remove things, and a few rounds of Zed Blade for this week's MAME ROW had me confident that it's working as it should.

moosepr

@obsidianspider yeah those headphone Jack's look a bit tricky. Can you even fit a soldering iron between them?

I would be tempted to get the cutting disk on the dremmel and slice the Jack's down the middle. That should let you then unsolder the outer pins, then remove the plastic, then the inner pins maybe?

obsidianspider

@moosepr said in Pi in a Gameboy Advance Build - WIP:

@obsidianspider yeah those headphone Jack's look a bit tricky. Can you even fit a soldering iron between them?

Maybe, but not easily, and I'd end up melting plastic all over the tip.

I would be tempted to get the cutting disk on the dremmel and slice the Jack's down the middle. That should let you then unsolder the outer pins, then remove the plastic, then the inner pins maybe?

That sounds like a good strategy since I'm not trying to preserve the jacks. I think the plan will be to try removing the mic jack since I don't intend to ever use that functionality, so if I mess it up it's acceptable. Then I'll plug it into the Pi to test. Then I'll remove the headphone jack, and plug it into the Pi and make sure that it is detected. After that I'll try attaching leads to the headphone jack and connect it up to one of the new headphone plugs I got from China (maybe it's good they came as a 10-pack) and test again. Then if that's working I'll go after the USB plug, and test again.

obsidianspider

I made some good progress tonight. @moosepr 's suggestion about using a dremel on things before trying to desolder them worked well. Just for a sense of scale, that sound card PCB is about 2cm square.

I managed to connect up the volume wheel and Adafruit amp, and everything works.

For anyone who is also going to try to figure this out with their own build, running speaker-test -c2 at the terminal will cycle white noise back and forth between left and right so you can make sure you have the proper channels identified.

When I plug in the headphones, the sound to the speaker shuts off, so that's good, but I am noticing some electrical noise coming from the speaker that is tied to activity on the Pi. (When the green light is flashing on the Pi indicating CPU activity, the buzz changes.) I don't get the buzz from the headphones when plugged in, so I don't think it's coming from the sound card. The buzzing is also present in the speaker even if the audio is only going to the headphones. That leads me to believe that the Adafruit amp isn't filtering anything, or at least if it is, it's minimal, so I'm going to have to see if I can find any information on how to filter the noise and I also want to see how to incorporate the shutdown circuit into triggering when when I plug in the headphones.

moosepr

@obsidianspider yay for destruction! It's lots easier when you can unsolder one pin at a time.

Love the scotch tape setup!

obsidianspider

@moosepr the tape was to make it easier to see what was going on in a photo and also to try to isolate audio from power wires due to that buzzing. Some reading is showing that I likely need some ferrite beads and that I should be using some resistors to bridge the stereo to mono, but it's 3:27 AM, so I should probably stop researching and go to sleep.