from this website:
"Allows select users to take part in Keurig consumer research panels. Invite only. If you’ve been asked by Keurig to use this app to take part in its consumer..."
I'd do it if they paid me to do it!
I ma taking note of the Particle silence on this one.... black ops I guess.
I see 
@Tomforti, so the P1 is basically a serial to TCP bridge acting as a data logger back to Keurig’s servers (or any data collection site). I would love to know what the target site is 
A little bit of sniffing can answer that question. My $$ is on aws kinesis…
I’m going to go ahead and suggest now we know what is Platform Teacup as listed in the platform-id.mk file, so that could probably be used to somewhat figure out what lines to tap, potentially.
Looking at the board, if you have a JTAG reader, all of the pins for that are broken out. As @ScruffR said, D- and D+ sure look like they come off underneath the USB area (P61 and 62). It would be an interesting experiment to just hook up a usb cable and some resistors, as per the manual. It’s possible that there is just some debugging info on the USB port still.
Unless wires are crossing over underneath the P1 module itself, it looks like TX/RX are around pins 40-50, which wouldn’t make them the standard serial port. Looking at the code itself, the only difference in the main firmware seems to be switching of the LED wires.
Considering it has the Reset and Setup buttons already broken out, and the usb, it would be quite easy to “repurpose” this 
I just want the eagle files for that board! I asked @zach for something along those lines during the last interactions he did. Bet one of the many smart people could replicate something very close pretty easily…just saying.
@LukeUSMC, I’m going to whip up a board based on what I see and add a proto area where the pin breakouts are. Or maybe I replicate the board, keep the connector and do a piggyback board?
Is there anyway the board came from Particle? I mean the board that the P1 is sitting does say spark.io
@peekay123 It would be awesome if it had a 20 pin (or whatever is appropriate) ribbon cable connector and that could interconnect this board to a proto board or a custom board pretty easily. The problem I have with most “shields” is expanding them is a pain. You have to solder on headers then use female->male jumpers or something else not as reliable…not to mention not everyone is capable of soldering headers in such tight spaces. A single ribbon cable to a breadboard, proto board or custom PCB seems to be the easiest and most reliable way of interconnecting to a pre-built “module” board. This will also help with packaging if someone is going for the smallest footprint possible as the boards could be stacked, side by side, perpendicular, whatever is appropriate for the case in use. In my selfish use case this leaves me only needing to design a PCB that is appropriate for the things unique to what I am building and gives me a known good Particle module design.
Instead of trying to replicate PCBs, you could probably just plug the RJ11 cable directly into a Photon. The red, black, yellow, and blue/green wires line up with the VIN, GND, TX, and RX on the Photon. I’d triple-check the voltage on the lines to be sure, but it looks like you might be able to plug it straight in (assuming 0.1" pin spacing on the RJ11 breakout side).
I don't know for sure, but I don't think so. The P1 is surface mounted to their PCB, and it's actually the P1 that has the (dated) Spark.io on it on the bottom. Since they kept a keep-out zone for the antenna, that's what you're seeing.
@wgbartley, I don’t want to replicate the PCB for getting the data off the Keurig unit! I want to duplicate it because it would be a great proto board.
I guess you should be able to connect it to any wifi. So I suggest you setup a wifi using e.g. a laptop and then use Burp or Wireshark to inspect the traffic.
This is how to do the same thing for an iPhone and Mac. But the principle should apply to the P1.
The only thing is if the software on the device is using SSL, then you have to make it accept the custom SSL certificate from Burp (which would involve a firmware update i guess).
Another approach is to set any computer connected to the same wifi in promiscuous mode and use Wireshark to listen to the traffic. Google: promiscuous mode wireshark.
This might be the easies way, but Burp gives you a better view on data as wireshark shows the individual TCP packages, where Burp understands the HTTP protocol, and can relate the TCP packages.
Could you snap a picture of the RJ11 connector that shows the wire colors inside? I’d like to hook up a serial monitor to the Keurig and see if the Keurig will communicate. Hidden features? Software update port? Interface to wireless remote?
thanks!
Red=3.3V, Black=ground, yellow=TX, green=RX
So I ran wireshark and captured CoPA and TCP packets to a AWS server (IP: 52.0.31.156) but not sure how to understand any other parts of the information it has.
What is the port being used on the TCP connection?
Thanks Tomforti. What I’m looking for is the pinout on the coffee maker side.
This a top view (release clip on top)
