My first non-trivial hardware: PCIeDuino

Just a bit more than a year ago I’ve started to work on a hardware idea as a challenge – a mini-PCIe form factor Arduino clone, the PCIeDuino.  The inspiration was working with a bunch of embedded boards, especially the VIA VAB-600 Springboard, that had a mini-PCIe connector, but not that many accessories that can go in there. (Disclaimer, I work at VIA at the moment, though this wasn’t a work project). I thought it would be cool to put an embedded-grade microcontroller on these boards, mostly to expand the I/O capabilities.


The original PCIeDuino sketch in my lab book
The original PCIeDuino sketch in my lab book

Looking at the mini-PCIe specs, the connector has mandatory USB 2.0 lines so communication was solved. The area available, roughly 5.1 x 3.0 cm, is not too bad  -more than how big the Arduino Nano is in comparison. The biggest challenge probably was that there are only 3.3V power lines available, and most example designs use 5V power supply.

Maker Taiwan

Taiwan’s Open Hardware movement gets a boost

I sure felt different yesterday morning waiting for my turn to give my 30 minutes talk, but in retrospect I’m really glad to have been invited to the Boost Open Source Hardware Movement event, organized by the CTIMES magazine over here in Taiwan. It was the second time try, after Typhoon Soulik cancelled the original event.

CTIMES personnel introducing the Boost Open Source Hardware Movement event
Kicking off the event early Saturday morning

There were 8 speakers scheduled from different companies and background: RS Components, Via, Broadcom, Motoduino, TMI Holding…. and me from Taipei Hackerspace. First I was wondering how do I fit in there, and maybe my talk *was* out of place a bit. Most talks were in Chinese so I could grasp only basic stuff from them, although the slides helped – most people made slides following the “slides are my notes” style, which is not my style, but was welcome this time. It was also great to see seasoned speakers like Richard from Via, and Lucas from TMI giving fun (and informative!) talks.

I did feel I’m in the right company, though. Open Source Hardware is becoming more and more of my focus – or maybe I’m just realizing it now that what I do is called such. Part of the audience were industry professionals and part enthusiastic hobbyists & students, and I had some great chat in the breaks with both kind of people.

I heard about the National Science Fair here (what? I got to check that out!), and how much interesting work people do with interactive hardware and elementary/high school kids. There is waaaay more going on that I can imagine, and there’s a lot more potential to tap into.

Had a chance to gather some industry experience too. Naren from Broadcom, who was responsible for getting Raspberry Pi into production was telling the story how they were expecting only 10,000 orders altogether and got 350,000 on the first day (sold 1.8 million to date), so they had to scramble an entire new supply chain. Thought that came to me was that maybe that’s one of the biggest value of  Kickstarter/Indiegogo is to be able to get an order of magnitude estimate of the demand.

Also heard about Via’s experience in pursuing new design and materials with their APC platform (such as using paper for housing), and interacting with their community. Heard a lot from RS Components how they are building tools for their community, and building the community itself. The demo was fought with technical difficulties (first rule of presenting is not to assume a working/fast internet connection), but it was inspiring nonetheless, and gave me some big (and difficult) ideas for my project: have to see if there’s any good community for scientific/laboratory electronics and hardware building, and if there isn’t then build one.

View of the room during the break time
Break time between the talks, everyone’s scrambling for chat and cakes

One of my favorite part in these events though to meet friends’ friends. This time was no exception, even if there was too little time to talk, I left with quite a few people in mind who I will need to contact soon, because they are doing something awesome and connected to my area of interest, either in hardware or in startups. This is one reason I’m trying to be really generous with my time and making introductions between people who I recon would hit it off well: I was the recepient of that so many times that I got to give back too.

My talk

I was scheduled in the middle of the afternoon just before tea break, and I was one of two people who used English for their talk (out of the 8 speakers). I was more nervous

I’ve uploaded the slides used for the talk in advance as well, so now it’s all out in the open.

My favorite part writing this talk was probably the attempt to summarize the philosophy and values of the maker movement, in a way that would inspire others. Some bits:

  • Don’t accept crappy – everything can be changed and improved upon.
  • Aim for collaborative creation. Celebrate the weird. Don’t mock.
  • Do and then share the results for everyone to learn from it.
  • For things to happen, you have to show up. Don’t wait for someone else to start, build up and inner motivation
  • Everyone’s values are different, a ‘space is often a different canvas for everyone.

So far the feedback I got about the talk is that I should have mentioned the projects people were working on in the ‘space, and upcoming events. That would have made a better ending for sure, and I had an extra 5 minutes or so to do that. Definitely going to emphasize the practical aspect next time.

Another thing I noticed listening to (a bit of) the video is that I need to use much less “ehm” and “ahm”… I certainly don’t remember using any, and consciously trying to avoid it in my talks in the recent years, I guess I need to listen more carefully (and prepare better).

Any more suggestions? What else wasn’t good? Bring it on, I want to become better at this.

More about the Taipei Hackerspace is on the mailing list, which is open for everyone to sign up, ask questions, show their projects, and hear more about what’s up.


Smartphone headphone to laptop adapter

My involvement in the Taipei Hackerspace so far had two very beneficial effect on my thinking: first I have much more ideas what new things to create, then I have much higher probability of seeing those projects through to completion. This post is a write-up of a recent project, illustrating both of these points.

About two weeks ago, I have fixed a headphone for a friend: the wire near the jack was broken and needed replacement. It was a relatively straightforward project, but needed a bit of digging into audio wires and jack connections. I was really proud of the result (fixing things have this effect, I highly recommend doing more of that!), and kept thinking if I can come up with any other audio-related project where I can use the knowledge I learned, and this is what I have came up with:

I bring my laptop almost everywhere with me, and started to use my smartphone headset to listen to music since it sounds great and much lighter to bring with me than the large headphones. It also has a built in microphone, so if I can use that, then really don’t need any other equipment to make Skype/Google+ Hangout calls. Those are 4-conductor headphones for Left/Right/Ground/Microphone channels, but computers (PCs) can only use the 3-conductor Left/Right/Ground and Mic/Mic/Ground connectors. Let’s make an adapter so I can break out the audio lines and mic lines to the appropriate laptop connectors!

Hardware setup

The parts needed:

  • 1pc 4-conductor input socket
  • 2pc 3-conductor output jack
  • 1pc 1-signal (mono) audio wire (signal + ground lines), about 15cm per finished adapter
  • 1pc 2-signal (stereo) audio wire (2 signals + ground), same length

The Guanghua Computer Market and its neighbourhood has a lot of electronics stores. The appropriate output 3-conductor jacks were really easy to find, as were the audio cables. Those didn’t look as good as the Bose headphone that I repaired, but it’s good for a prototype. The hardest part was the 4-conductor input socket: they had some that should be mounted on a printed circuit board, and tried that one for the first prototype, but then they found me some better one that I can use with the cables. That was the most expensive part at about 20TWD ($0.70).

Wiring diagram showing the internals of the adaptor
Wiring diagram of the whole cable, M=microphone, G=ground, R=right, L=left

Above is the circuit schematic, probably a bit of a mess, but tried to keep it simple. Then aim of the whole setup is to get the G (ground) + M (mic) lines to one jack (the mono audio, thinner cable), and G + L (left) + R (right) to another (stereo audio, ticker cable).

One complication is that the order of M-G-R-L seems to be the “Apple Way”, that my HTC headphone adopted as well, while others have a more logical (and easier to solder) G-M-R-L series (eg. the Sony PSP headphones as I found searching for it). Thus this adapter would not work for every headphone. Maybe version 3 should have a switch to swap the Mic and Ground lines at the input socket?

Parts for the first version of the adapter
Version 1 of the adapter: 2 pieces of 3-ring jacks, 1 piece 1-line wire (that is signal + ground), 1 piece 2-line wire (2 signal + ground). The input jack is different for the final version, and heatshrink will not be needed.

After all the parts are collected, there’s some micro-surgery. Strip the audio wires carefully, and don’t have to leave much out, just enough to get to the jack electrodes, and such that the thick outside cover fits into the cramp that is there to hold things in place.

For the microphone wire, connect the two output channels (the tip and the 2nd ring), and the ground goes to the base one. For the audio output, the tip is the left channel, the 2nd ring is the right, and the base is ground. The audio wire I got was a bit thick to cramp, but it’s not too bad if the metal cuts into the plastic cover, as long as it doesn’t cut through it.

The output jack image
Soldering one of the output jack

The input socket was quite a bit trickier to solder, because the electrodes were so tiny. They are arranged in a 4 directions, going around clockwise as “tip – 2nd ring – 3rd ring – base”. Had to use some magnifying glass and one spare unsoldered jack to make sure I connect the right things to the right places, but it works – mostly. The hardest part is not to melt the individual wire insulation (they are not enameled wired as inside the manufactured headphones). Also, cramping the two wire together is tougher, had to cut off the sleeve of the socket so the wires fit within the plastic cover of the socket.

Soldering the input socket of the final version
Soldering the input socket of the final version, that’s pretty tight for the two wires, have to make sure not to burn off the plastic cover or them

This second version was done within about 15-20 minutes, though, since I had all most of the research done previously. Had to do some careful inspection that no shorts developed within the device because of the soldering, and then connect everything up.

The final design plugged into a laptop with headphones
The final design together with HTC Butterfly headphones, with the microphone visible on the left

The listening sound quality is pretty good, though I guess it could be improved with better soldering (which includes using a finer tip soldering iron), and more patient cabling work. The mic is pretty nice, my recordings sound much better than before, and it’s convenient to use too. Much less thing to carry around and definitely better than my laptop’s built in mic. This might even get me to make more calls (sorry Mum and Dad that I haven’t been phoning too much lately!)

Taking it further

The electronic markets here in Taiwan are full of gadgets, and I looked but couldn’t really see any commercial adapter that did the same thing. This made me think that maybe it could be interesting to make this into an actual product. First find some local factories that could make it, as I keep hearing my business-related friends manufacturing a lot of things. Then set up a Kickstarter/Indiegogo project with some reasonable (let’s call it hobby-level) financial goal, and see whether people would be interested.

This would need quite a bit of preparation, and what’s putting me off is the simplicity of the project (what’s too simple for Kickstarter?), and that a Chinese factory could rip it off faster then you can say Shenzhen. Still, it might worth it for the experience and contacts, will do some research, and in the meantime keep making stuff that interests me.

Update 8/6: as a commenter pointed out, these things exists already e.g. on Amazon, searching for “smartphone headset pc adapter”. I guess that simplifies the future a bit, have more time for better projects.