Think you’ve seen every possible type of Arduino based hand held video game? [Kevin] managed to coax something new out of the theme with a very clever credit card sized console that uses some very interesting construction techniques.
The inspiration for this project began when [Kevin] dropped an SMD resistor into a drill hole on a PCB. This resistor fell right through the hole, giving him the idea creating a PCB with milled cutouts made to fit SMD components. With a little experimentation, [Kevin] found he could fit a TQFP32 ATMega328p – the same microcontroller in the Arduino – in a custom square cutout. The rest of the components including a CR2016 battery and OLED display use the same trick.
The rest of the design involved taking Adafruit and Sparkfun breakout boards, and modifying the individual circuits until something broke. Then, off to Eagle to create a PCB.
[Kevin]’s experiment in extremely unusual PCB design worked, resulting in a credit-card sized “Game Boy” that’s only 1.6 millimeters thick. The controls are capacitive touch sensors and he already has an easter egg hidden in the code; enter the Konami code and the Hackaday logo pops up to the tune of [Rick Astley]’s magnum opus.
Now [Kevin] is in a bit of a bind. He’d like to take this prototype and turn it into a crowd sourced campaign. In our opinion, this “Game Boy in a wallet” would probably do well on a site like Tindie, but any sort of large scale manufacturing is going to be a rather large pain. If you have any wishes, advice, of complaints for [Kevin] he’s got a few links at the bottom of his project page.
It’s not a GameBoy. The title is misleading.
Seconded. Call it handheld game or whatever, and don’t poison article titles and project names for those that will, some day, create the actual thing it is named for.
I would like to open discussion on if this is possible! What is the smallest microprocessor you could fully emulate a gameboy on? Now I am thinking about finding a screen with the correct resolution…
Someone did a sega master system emulator on an STM32 discovery board. That might be a good place to start the research.
I designed the Arduboy against the same physical layout of the buttons and screen scaled down to the size of a credit card. I’ve gone so far as to call it a Gameboy “clone” and some people still take issue with that. Just think of it like it is a clone but something went wrong along the way…
It’s certainly not a clone. A clone is identical to the original.
The word you are looking for is “inspired”, the device was inspired by a gameboy’s form.
“To produce and sell the Arduboy requires a Cad soft Eagle license that costs $820. This would be my initial goal of a crowd sourced campaign.”
KiCAD would be free ;-)
Normally i would argue that re-making the board in another cad program would not be worth the time compared to getting the license, but this board is so simple there should be no problem.
In some ways I wish I had started this project in KiCAD but getting started teaking myself… Eagle was the easy choice because board files from sparkfun and adafruit were easy to find and OSH Park handles them without needing to output gerbers. I’ve had offers to cross develop this in KiCAD for the release.
The real reason to do crowd source is to meet quantity discounts and minimums.
This is precisely *why* open source programs and open data formats are so important. If you don’t “get it” at the beginning of a project, then you are either forced to continue with your closed system (and force that decision on other people related to the project), or incur a huge cost (money or time) to extricate yourself.
I will not contribute any money if I know it’s going to be wasted on an unnecessary software licence. Others won’t feel the same way, so good luck anyway.
I’m using Kicad for all my work. Why ? Because at school I learned to route with Kicad. I also learned how to use Eagle but it still easier to use Kicad (even if it’s not as powerful). There is just one problem : libraries are so empty ! I need to create every components I used. Last days, I needed to make a 2x5mm LED. Something as usual as a 2x5mm LED should be on default libraries.
I’ve been learning Altium at school. and have a student license for it now. ironically it is cheaper for me to pay tuition fees than the Altium license costs…
As for libraries, you should become proficient in making them yourself. I used some sparkfun libaries along time ago and the footprint was wrong, ruined the entire board. much easier to spend a few minutes making a new footprint than to waste time and money on non-functional circuit boards.
By the way, there is already some components mean to be reverse mounted. Check this for example : http://www.kingbrightusa.com/industry-thinnest-ultra-super-thin-smd-display.asp
greg, the problem isn’t that he doesn’t know how to make component libraries, it’s that kicad users have to waste time reinventing the wheel with every project. the situation is like programming in a language without standard libraries.
It’s a tried and tested method of making circuitboards thinner through milling out space for the packaging of SMD components, the plethora of Nintendo Game & Watches dating back to the beginning of the 80s and the Tomytronic 3D handhelds in the early 80s use this technique for housing their main IC on the circuitboard. I’m sure there are many more other early handhelds which use this method.
But the thinness of this device is most impressive.
I was planning to use the exact same technic for my electronic business card. But there is one problem with this kind of component placement : you can’t route under them. And routing under components can sometimes be a lifesaving operation.
Haku! I love your comment! I thought I remembered seeing this kind of cutout placement for the MCU but was unable to find images online. I liked to take electronics apart as a kid so perhaps this is where I first saw one. Please let me know if you can find an example and share it!
A big part of this project was to release this concept, this theory, into the community. I want to release a library with all of the components I used and a few more… I will also publish the details of how to create your own parts accounting for tolerances and board house rules.
I couldn’t find any suitable pictures online, so I dug into my own handhelds collection:
Tomytronic 3D Thundering Turbo:
http://www.haku.co.uk/pics/Circuitboard-Tomytronic3D.jpg
(I have the full collection: http://www.haku.co.uk/pics/Tomytronic3D.jpg )
Nintendo Game and Watch Donkey Kong:
http://www.haku.co.uk/pics/Circuitboard-NintendoGameAndWatch.jpg
YES! Thank you so much Haku!
So pretty :D
Please release this library as a kicad library first? You have now experienced the onerous effects of closed source. If you lack the experience in Kicad, just collaborate with someone who is more experienced, it seems like there would be plenty of volunteers here.
Me too, I would not pay towards buying you a license.
Btw: would a USB connection for charging and firmware flashing be possible?
Model 100 LCD assembly: half of the driver chips were mounted upside down (probably for routing), but for some reasons other than thickness they cut holes for them.
http://www.nycresistor.com/2013/01/06/trs80-model100-upgrade/
3rd picture down
They don’t want to route the part on the side facing the LCD, so this round about way let them flip the chip.
the eureka moment made me …what?
This makes Game Boy Micro look like Game Boy Macro!
Incredibly cool!
Best. Rickroll. Ever.
I hope it works, I’d buy one :)
Good Luck XD
Very neat, (but the basic idea, as Haku points out is not new).. perhaps the next step would be to make it flexible.. … also not new… but arguably, even cooler…
( for example… http://www2.imec.be/be_en/press/imec-news/archive-2009/imec-reports-ultra-thin-chip-embedding-for-wearable-electronics.html)
… I’m not knocking Kevin’s work in any way though… its a superb bit of design, well thought out and produced.
A flexible “game boy” would be unplayable my friend
Ok, I was planning on publishing this later but you are on the right trail anyways:
It is totally possible to reproduce this circuit into a single-layer dual-access flexible circuit. I have had a quote from a manufacturer. It starts to become financially practical at around 500 units.
There are thin-film li-on batteries of the correct dimension, and thin film speakers as well. As of this moment however, I can’t find a thinner graphic display. PLEASE contact me if you know of something.
Without having a thinner screen, the minimum thickness will still be ~1.6mm. It would be possible to eschew the standard circuit board for a substrate that is semi-flexible such as plastic. But again the screen is the least flexible portion.
I’m pretty sure flexible displays are something of a cutting-edge tech right now and not easy to buy on e-bay ;)
I’m sure there are flexible OLED available, that was one of their big things IIRC.
Good job!
For protection, he could add a second, much thinner PCB (like 0.8mm) and solder through plated holes.
Limpkin did a nice front plate and solder on backplate: http://www.limpkin.fr/index.php?post/2012/09/15/My-new-business-card
Limpkins was a big inspiration, showing me the way was possible. I’ve thought about maybe using plastic or even paper cutouts. If it wasn’t for the screen and the battery things could be reduced to about 0.9mm
Some people should be thrown into a vulcano…
..for making other people look inferior.
>To produce and sell the Arduboy requires a Cad soft Eagle license that costs $820. This would be my initial goal of a crowd sourced campaign.
So in other words Kevin refuses to learn Kicad for &800 cash money..
Somebody else’s $800 cash money, though.
He may just be unaware of it. I’ve heard of it, but didn’t know it was open source.
When the KiCAD site gets back up, I’ll give it a try.
Coming back to this, open source files that require an expensive program to use seems an ill-considered idea. I gather EAGLE doesn’t do more than two layers in the no-cost tier.
Eagle has the advantage over other cad software in that there is a free version that can open and do some editing on files made with licensed version, most other PCB cad software has no free version at all. (not counting KiCad ofcourse, which is all free)
DipTrace and DesignSpark both beg to differ.
Unless I’m mistaken its impossible to output Eagle files from Ki-Cad. I would like to support both platforms.
The cost goal is just an initial benchmark because it’s the only fixed cost I have. I need to calculate what the marginal cost of making kits and shipping them, and in the end this will probably account for a larger percentage of the final goal. If anyone has any experience doing this for crowd sourcing please contact me, I could use some help.
I’ll put this out there too: If I fail to reach my kickstarter goal I’m going to release what I have now without any changes so you’ll be able to do the KiCad work yourself regardless.
Yes, but eagle is better. That 800$ buys it forever, not just for this design. Altium is many times better than eagle still, but somewhat more expensive (around 5000$)
That $800 buys is it forever *for you*. The next guy who wants to work on it also has to pay $800.
This is my last comment on the subject: I have had several people reach out to me to develop the board free of charge in Ki-Cad and nobody has offered to develop with their existing Eagle license. End of the day, I’m still in the Eagle camp but this project has opened my eyes wider to the debate. I also use Photoshop not Gimp for the record.
I would like to see a picture of the back of the completed unit, but it’s nowhere to be seen ?!
It’s a 1 sided circuit, I’ll show the back and the schematics after the crowd source campaign (whether it meets its goal or not).
I see on your site that you want to make a 4-layer board, and yet, it’s already working as a 2-layer board. You are basically going to add a lot of manufacturing cost for no real return. And since you need to buy an Eagle license purely to get the extra layers it seems even less like a good idea. Sorry.
Andrew, I’m looking to cover my costs for doing additional work. It would take a much longer time to re-do everything in KiCad. I’m going to release the source files if the crowd source campaign is successful or not. Keep in mind if I would have never mentioned what I was developing with you wouldn’t even be upset over this.
Thank you for participating in the discussion here, despite some negative opinions (mine included). It is your project, and it is quite splendid. It’s true, if I didn’t know about it I wouldn’t care, but I do know, so it’s more a lament that you will take something cool and add cost and complexity. Most design engineers work hard to *reduce* the number of layers (and components) and overall cost.
Regarding Kicad, well, people often claim ‘no big deal, Eagle is free if you work within the restrictions’. I have always maintained that this is true, but it will bite you in the ass one day. Today is that day.
I still wish you good luck.
I love this design! The only aspect that I couldn’t find on his website was how he charges the battery.
Just tap the solder tabs on the battery real quick with a soldering iron!
To get a more flush appearance I have to adjust one of the tabs slightly with pliers.
I have had a bad problem buying these batteries DOA though that’s an issue needing to be sorted.
Very nice. I was planning to do something similar after seeing a few business card PCBs on here about a year ago. Annoyed that Kevein beat me to it, but impressed all the same. I had no idea that that had been done before in the 80s though!
I was planning to invert some LEDs in the same way as the uC but impressed he got a screen in there. I had no idea you could get batteries with tabs but was planning to do something to mount an even thinner CR2012.
Drop me an email, I would be happy to work with you once I get things sorted. :)
cool idea; I hope your funding comes thru.
I’d like to see one of these in real life and maybe use the concept later on.
This is a fantastic project! I would be very keen to know when crowd funding for this begins. Are you looking at using Kickstarter? If so I could most probably give you a whole heap of tips as to what to expect, I have run two successful campaigns (in fact my last campaign was a credit card sized LED game device!)
Brad, please do drop me an email!
http://i.imgur.com/JhKG5IS.jpg
Just to say, if anyone makes this, the screen should be quite a bit bigger. And check the response time on the inputs, to be sure they’re fast enough for game playing. Also an SD card slot would be nice, and maybe a few K more RAM? As it stands tho it’s still an excellent idea and I’d LOVE something this tiny I could keep with me, play whenever I like, and program myself.
Also nice would be either a rechargeable battery, or including a 2032 / 2016 charger that runs off USB, and is cheap! Maybe with a connector that also programs the card, even a little cradle.
Oh and, if you like, you could run any unused pins off to contacts that the user could drill thru and use later if he wanted.
Oh no! The ideas keep coming! The last one for now is, maybe a smart-card connector? So people who develop smart card apps could debug with it.
I’ve got some reservations about the form factor of this but the need to re-solder a new battery after every 9 hours of game play is an absolute deal-breaker for me. What I don’t understand is why he’s chosen the CR2016, look at the PowerSearch site and you’ll find heaps of rechargables like the PGEB0083559 which is 0.8mm thick, 65mAH and only $3.30 when purchased in bulk.
(Correction: powerstream, not powersearch).