Bluz Coin: New Form Factor for Wearables


#1

Clearly when bluz started, making a device that is pin compatible with the Photon/Core made a lot of sense. It allowed us to work with all existing shields and accessories and made the transition to bluz easy for existing projects.

However, there are many applications that would benefit from simpler and smaller form factors. The first one we wanted to try out was Wearables, so create a small circle form factor similar to the Adafruit Flora. Here is what we are thinking:

This is 1.2" in diameter, so just a tad smaller than a half dollar. It is a fully functional bluz board, it has the RGD LED, two buttons, LED for D7, everything. The only limitation is there are only 6 GPIO pins (including I2C and UART pins), but other than that it is a fully functional bluz board.

On the back would be a battery clip for a CR2032 battery, so this would be fully standalone. We can add some new holes to make it sewn down easily as well.

Anyway, what do people think? Any interest in this new type of form factor? Any thoughts or feedback? If we were to try it out, we would probably prototype a small-ish number of devices and gauge the level of interest, so let me know!


Cypress BLE sensor beacon with Bluz GWS
#2

#3

Why the GPIO limitation ? Real-estate/placing/routing ?


#4

Yeah, pretty much. To get the coin cell on the back, we had to limit it. I could probably squeeze on a few more, or we could just make it bigger too. I was going for “small as possible” while still using the same 0603 components as on bluz. I could probably go even smaller if we went with 0402 and a smaller RGB LED, but then there would probably be even less GPIO.


#5

Which way does the coin cell enter the holder ?

If it is from 6 o’clock (looking at the image you posted), then I see that battery removal/insertion would interfere with the GPIO pads, but if it inserted from 12 o’clock instead, could you make use of more of the circumference ?

If it is already from 12 o’clock - could you sneak the cell holder far enough north to free up space on the periphery for more GPIO ?


#6

Yea, the battery is inserted from 6 o’clock. I don’t see where that would interfere though?

We could probably move the battery around a bit to get more IO on there, this was just meant to be a high level concept starting point.

Do you have some application in mind that would need more? Or is it just a matter of trying to fit as many as possible?


#7

I figured it would interfere if you soldered wires to any pads that might obstruct battery access (since the cell slides in flush to the board.) If you use a different holder that was not flush to the board, that might assist too. Even if the wires exited the front, they can/would project out the back (and shorting to the battery on insertion might not be a good idea.)

I’ve got no app in mind; just that no-one ever seems to have enough GPIO, and as soon as you have less than a vanilla DK, you have to explain something, and not having to explain anything is always better.


#8

Connectorize it! Tired of all the little boards that have to be hand wired. Something like the tiny and cheap JST connectors.


#9

I think it’s a great little chip. Is there also the possibility of having shields to go with it. Such as light and motion monitor, or weather sensors.
An application I am thinking about is monitoring my house plants. Thus it would be great to have a chip set in each plant pot that would monitor the light, temp, humidity and possible soil moisture. But then again, just having a development shield would work great too.

Also, and I may have missed this before, but could it be used as a location beacon?


Cypress BLE sensor beacon with Bluz GWS
#10

I love the idea of this little guy, both for wearables like a button to trigger something, or a temperature-humidity/moisture/open-close status device that could readily be deployed. Naturally, I would hope the price point would be significantly less than DK & battery shield together.


#11

Let’s see if this conversation can be revived.
I could use 5 BLZ Coins today, although I would prefer some changes.

  • The I/Os should be on standard spacing, 100 or 50 mils. This will allow soldering headers for expansion shields.
  • I would make it square; why waste the PCB space. Even for a wearable application one needs to connect hardware and wiring is the only way… messy, more work and not easily serviceable. In other words, the Coin does not have a place to add sensors, etc., so the end product is bigger. I guess one could stick LEDs in the holes and make light patterns. Very limited use. Wait, if it’s square you can’t call it the Coin anymore. The Square!
  • Definitely need more pins. I would say all the pins on the DK.
  • Is the chip at the bottom the SPI memory? There are smaller packages, Then you can bring out more pins or add sensors.
  • Another way to provide pins is with alternate FW configurations. This would make it incompatible with existing code, but it would be easy to fix.
  • Is the antenna connector at top right? I would make a solder pad and use the space for pins.
  • It should have an optional temp+humidity sensor (price). That will be the most common function for the Coin.
  • I don’t know how well this works and how much processing it needs, but I have heard of devices using accelerometers as buttons. Tapping the Coin with your finger can indicate button press sequences. There are MEMS sensors that also include a temp sensor, all over I2C. The bus can be reused for other sensors and you minimize the need for pins. (sorry if I am preaching to the choir).
  • The battery should be optional (price) to a pair of solder pads. The first thing I would do is design a solar shield, kinda-like this BLE beacon (sorry, I think this beacon is really cool).
  • Price… less than $10 for the bare bones version, with a quantity structure. C’mon, you know this will sell more Gateway Shields.
  • Did I mention more pins?

Based on the picture, the artwork design seems done already. It looks like you repackaged the DK design, so the parts already exist as well. If this is never going to happen, how about publishing the design files?
I design hardware for a living and have access to SMT soldering equipment. I can help you build some, debug them and make BLE sensors great again!


#12

That is quite the list of changes!

Our hardware designs are always open, you can see our schematics and board layout for the bluz DK here: https://github.com/bluzDK/bluzDK/tree/master/v2/HWDesign

Ultimately, that may be better for you to work with than the Coin files as it was never put into production or tested, whereas the DK design files were actually built. You are absolutely free to make whatever type of form factor you would like, you just have to give me one :smile:

We have a new provisioning system in place so it is easy for you to build a bluz DK compatible board, then provision it for the Particle cloud. You can buy the modules that we use direct from our store: https://bluz.io/store

The BOM is also online in the above referenced GitHub repo, and you can buy all those parts on DigiKey or somewhere similar.

If you pursue this, I would love to see what you make! Ultimately, we could help market and sell it as well if there is a real interest.


#13

OK. Not quite the level of enthusiasm I was hoping for. The challenge is certainly tempting.
I am wondering why this design doesn’t exist already. It seems like the logical next step in evolution.
The support in this forum and the web is unbelievable and the Particle web services are priceless (literally and not).

Would you share the Coin files? It would be easier to start from that design instead of the DK.
I have some general questions, if it’s no bother:
How many Coins would the gateway shield be able to talk to at one time?
In the interest of making the Coin smaller, I read here that the SPI Flash could be removed to rely only on internal memory. Would any functionality be restricted or lost?
Would you be willing to modify the FW to use only the internal memory?


#14

We definitely appreciate your enthusiasm for our concept design, but ultimately we didn’t follow through because we didn’t see a business opportunity. At the end of the day, we are a very small team and need to pick and choose the projects and pieces that can help move us forward. While I would love to see other form factors for bluz, it just wasn’t something that we needed to do to help move bluz forward. And, it is something we would love to see community members do!

I would be happy to share the Coin files. If you can DM me your email address, or ping us at hello@bluz.io, I can get them to you. I don’t want to generically publish them since the designs could have issues since we never built it.

The hardware gateway can talk to up to 3 devices at the moment, but we are working to expand that up to the max of 8. I hope to have that done soon.

You could do it without the SPI Flash, but you would lose functionality, mainly OTA updates. For a development board, that is critical as otherwise you would have to program once and be done with it. I think it would be better to find it in a smaller package, which is available. That would make things much easier and wouldn’t require any firmware changes.