ARMACHAT - Doomsday wireless QWERTY communicator

Experience freedom of communication without borders, operators, networks, fees !!!
Create your own network, your own rules !!!

Similar projects worth following
Starting from
bobricius has 969 orders / 65 reviews
Ships from Slovakia
For HACKADAY PRIZE 2020 Project moved and enhanced with 2 new models.


All UPDATES moved and continues on above page.

VOTE / LIKE on this new page, THANKS ;)

On tindie store are only first boards with slow display refresh, not recommended for production.

Finally I make small distance test, no direct view, between houses in village.
(Max power, Spread Factor 12)
-700m unit on garden desk
-1500m unit in window, about 4m above ground

Experience freedom of communication without 
fees !!!

Create your own network, your own rules !!!

First prototypes manufactured in Aisler :

FUSION model - combine best features from both prototypes Manufactured in PCBWAY:

Hardware features:

  • SAMD21G18 for desktop version (pocket version use SAMD21E18, more easy to solder in TQFP32)
  • QWERTY keyboard
    - 40 keys (4x10) desktop version
    - 30 keys (5x6) pocket version
  • 18650 li-on battery
  • 1.8 TFT LCD display 128x160 ST7735
  • RFM95 long range radio module
    - 3 freq. by your country
    - IRQ / wake up
  • Beeper
  • Battery charging
  • Software RTC with alarms

Working software features:

  • Send message ... compose and send message to specified recipient or broadcast 
  • Receive message ... message is received on background and stored in memory.
  • Message editor ... write message about +/- 80 characters long ...
    80 is now best readable on full display ....
  • Message memory ... all send/received messages are stored in memory
  • Display memory ... display memory content
  • RTC .... display time
  • Battery ... display battery voltage
  • Config ... user radio configuration spread factor / power
  • time set
  • check message delivery confirmation
  • contact list display and check online contacts
  • select recipient or broadcast
  • sort from memory by contact
  • BEST working feature - resend undelivered messages
    if delivery fail, message is stored with "undelivered" attribute. Every station at power on broadcast "Ready receive status", then if any near station receive this broadcast and have undelivered message automatically try send this messages,
    if then receive confirmation mark messages in memory as DELIVERED :)

TODO software features:

  • add message encryption
  • think about message forward (mesh)
  • add some USB link features like work as usb modem, send receive/message from PC terminal
  • GPS position parse

STEM Learning:

What you can learn on this board?

  • modern architecture based on M0 arduino zero compatible, with lot memory and native USB
  • QWERTY keyboard matrix reading
  • complex input/output HID processing read keyboard data,
  • create and display graphics, save data to flash, send results over radio or usb
  • radio communication with world best, most advanced long range radio technology
  • work with I2C sensors or process GPS data
  • learn lot about network architecture, packets, communication, addresing, delivery confirmation
  • RTC clock and alarms
  • power management, sleep, wake by alarm, wake on receive

( all this things I learning now on this project)

Production and marketing plans:

Hardware is very simple and easy to solder by hand (in pocket version with SAMD21E18), low active component count make it very attractive as learn to solder kit. Unfortunately SAMD21 come without bootloader, this is little problem because SAMD need special tools to burn bootloaders, I have no experience how much is complicated burn bootloader in factory on bare unsoldered chip. I suggesting sell this as semi populated board > PCBA all active components except buttons and radio module (Unsoldered radio module give small advantage because there is 3 world frequencies) Why TH tactile switches? because is very important to precise align with front cover plate and this can be little hard with hand.

SWOT analysis


  • no competition with almost finished and working communicator
  • compact design
  • qwerty keyboard, much user friendly than encoders or arrow keys
  • officially supported Arduino powerful microcontroller with lot free memory and computation power
  • using only 15% memory now, unlimited future enhancements
  • extremely simple just one easy solder chip and buttons, cheap and low component count
  • low power consumption
  • color graphics display


  • no community, no users (maybe because I have not started selling )
  • currently no mesh networking or advanced message routing
  • my programing knowledge is limited (I quickly learning)


Read more »

View all 4 project logs

  • 1
    • there is nothing special, I build 6 prototypes, with hand soldering, without issues

    I am not saying best, they are just my steps :

    • Solder CPU, and all active/passive components
    • Burn bootloader> Connecting Atmel SAM-ICE> Chose Arduino zero native USB> Burn bootloader
    • Solder Display
    • Try some demo/test software
    • Solder radio RFM95 module
    • Solder buttons

  • 2
    Basic software:
    There is also nothing special, All parameters can be configured on gui menu.
    • Download code, libraries, Burn.
      #include <Keypad.h>
      #include <RTCZero.h>
      #include <Adafruit_GFX.h>    // Core graphics library
      #include <Adafruit_ST7735.h> // Hardware-specific library for ST7735
      #include <SPI.h>              // include libraries
      #include <LoRa.h> //
      #include <FlashAsEEPROM.h>

View all instructions

Enjoy this project?



Festerstoybox wrote 07/14/2020 at 10:44 point

I'd love to but one from you, or at least have a list of parts I need to build my own. 

  Are you sure? yes | no

ko4dfj wrote 07/02/2020 at 23:55 point

Need one!

  Are you sure? yes | no

Devin St. Clair wrote 06/27/2020 at 02:45 point

Can we purchase a kit somewhere?

  Are you sure? yes | no

bobricius wrote 06/27/2020 at 20:58 point

in future bare PCB boards

  Are you sure? yes | no

Hsingai Tigris Altaica wrote 05/14/2020 at 20:35 point

Have you thought of joining forces with Meshtastic?

  Are you sure? yes | no

José Miguel wrote 05/01/2020 at 15:30 point

I love the idea (communication with nothing else than your own hardware rocks!), and it has caught my attention as I've just bought a pair of LoRa transceivers!

Will the protocol be documented/available to the public? It would be nice if the protocol used were clearly documented, so more people can benefit from it and join the network with their equipment :D. That would also make your device more popular (although it is already receiving some love)

  Are you sure? yes | no

asutrem wrote 04/30/2020 at 02:30 point

Do you have a BOM list for the fusion version?

  Are you sure? yes | no

x893 wrote 04/26/2020 at 09:30 point

Cool project !

Why not use any smartphone to use big screen and touch ?

Connect device via BLE or USB OTG ?

Solar for changind.

Next after doomsday.

  Are you sure? yes | no

bobricius wrote 04/26/2020 at 09:43 point

How long your smartphone can operate? is sensitive device, did you see smartphone without broken display? (I not), absolute not possible home made construction. You have good project Disaster radio. This is radio education board ;)

  Are you sure? yes | no

jeroen wrote 04/27/2020 at 19:54 point

I did exactly this using the rocket scream Lora board. Put much effort in porting Lora drivers to  FreeRTOS for low power (standalone) operation and build a working basic Lora 'Chatroom'. Here a vid of my phone and laptop using serial/OTG VT100 terminal communicating over LoRa:

But I really like this project, no phone dependency is great. Wonder how the keypad feels! Let me know when these board are 'available' would love to play around with these.

(my project is stalled because I got a payed job)

  Are you sure? yes | no

Jonathan wrote 04/26/2020 at 07:57 point

This would be great with a BB10 style keyboard, look at this for example:

  Are you sure? yes | no

bobricius wrote 04/26/2020 at 08:03 point

Yes, I know, but I want standard components not replacement parts.

  Are you sure? yes | no

Jonathan wrote 04/26/2020 at 09:02 point

That makes sense

  Are you sure? yes | no

Dave's Dev Lab wrote 04/22/2020 at 23:20 point

pretty cool design! i was looking at doing something similar, but haven't had the time. thoughts about how to add a barcode scanner?

  Are you sure? yes | no

Jdcrunchman wrote 04/16/2020 at 23:42 point

Do you actually have a unit up and working?   Who is the main coder?

  Are you sure? yes | no

bobricius wrote 04/17/2020 at 23:06 point

I am main coder :)

  Are you sure? yes | no

Jdcrunchman wrote 04/16/2020 at 23:41 point

Do you have a Slack account setup for this project?    Is there any code up on GutHub yet?   Who is working on the code,  and what coding environment are you working under?  Arduino,  MicroPython?

  Are you sure? yes | no

bobricius wrote 04/17/2020 at 23:08 point

Gode is not published now, work in progres. Working in ARDUINO IDE. In theory it can support Micropython.

  Are you sure? yes | no

Jdcrunchman wrote 04/16/2020 at 23:34 point

It should be EMP hardened....   With my 4 years USAF experience operating on Cold War era radar equipment,   EMP hardening is something I can contribute to.   But more importantly,  my experience with Micropython on the ESP32 and SX1276 LoRa transceiver is going to be a big issue with battery drain.....   Developing a power management unit is a MUST.

I would be most interested in looking at the code,  but more interested in field testing it under urban conditions,  in places where multi-path can be a problem,  bouncing off stucco buildings, where there is a lot of those in Vegas.

Has anyone submitted field test results,  perhaps "Answer back" numbers.....

  Are you sure? yes | no

bobricius wrote 04/17/2020 at 23:13 point

With on display, receiving mode and all work on full is power consumption 50mA. I have only one quick test, I have no resources for more testing. I am working on this less than month.

  Are you sure? yes | no

bald engineer wrote 04/14/2020 at 18:51 point

Very interesting design. Will you be sharing the schematics and code for it? Is your plan to sell these on your Tindie store?

  Are you sure? yes | no

bobricius wrote 04/14/2020 at 22:23 point

Design files are on github, some software +/- week. I have plane sell but without radio module, buttons and battery holder

  Are you sure? yes | no

RomanS wrote 04/14/2020 at 09:21 point

It's better to use EBYTE E32-433T30D / E32-866T30D LORA 1w transceivers The range tested with my version of LORA communicator, is up to 8km!

  Are you sure? yes | no

bobricius wrote 04/14/2020 at 12:26 point

interesting, thanks very much ! In next board I try add header with compatible pinout

  Are you sure? yes | no

RomanS wrote 04/14/2020 at 15:23 point

but the real challenge is to provide  LORA MESH and LORA WAN integration to cover really wide areas )

  Are you sure? yes | no

bobricius wrote 04/14/2020 at 15:31 point

yes, mesh is cool, I think it is just software problem. After I tune HW then take focus on mesh.

  Are you sure? yes | no

Kosma wrote 04/13/2020 at 22:37 point

how long this device work? days? montsk similar kindle?

meybe put solar panel ?

  Are you sure? yes | no

bobricius wrote 04/13/2020 at 22:45 point

untested, I measured consumption now it is about 50mA, full display backlit, radio onreceive, CPU at full power, no power saving. 26mA at low backlit. You can easy calculate work time to battery capacity. I start with power saving after I finish radio communication.

  Are you sure? yes | no

greenaum wrote 04/13/2020 at 18:35 point

Really great appearance and looks like fun! Is it only limited to 500M range though? Perhaps in a disaster you could scatter repeaters around, similar hardware but no screen perhaps, just some indicator LEDs and a couple of buttons.

Does it really need complicated hardware to flash a bootloader onto the pocket-version CPU? Can't one of the already-bootloaded chips program it? Programming MCUs is usually just a matter of asserting one or two pins, before or after putting the address and data onto their busses. Just toggling pins in the right order, essentially. I'd be amazed if there weren't existing Arduino projects to program bootloaders onto blank chips, although possibly only for the AVR versions.

I know you didn't ask, but... a mono LCD with a switch for the backlight might save lots of power. Or even the Sharp memory LCD that only uses power to alter the display, none at all to retain it. For slow-moving things like text it'd be well suited. Might not suit all your needs though.

Hm talking of gateways, maybe you could set up a gateway to IRC or whatever the kids are chatting over these days. If they still do text chat. None of those stupid smiling pile of crap icons! It would be fantastic if, however it's done, this could enable international chat, for free, and so simply.

I really like the way this looks, it's really neat and compact, no wasted space, all functional. Maybe you could make a micro-version to wear on your wrist, your wife can call you in when you're out in the shed inventing things!

As for the Chinese selling cheap clones of your design, I'd welcome it. I don't think this is something that's going to make your fortune. If you were serious about that you'd need to set up proper manufacturing as a real business. Me, I'd be happy having invented it, if there were people all round the world with copies they bought affordably, all able to chat with each other. Saves you the bother of being an underpaid solder-monkey. If you want to make money there's better ways than hobbies, and hobbies are more fun.

  Are you sure? yes | no

bobricius wrote 04/13/2020 at 22:18 point

I working on this project about mont ... software side few hours, before this, I have no idea about radio communication or about graphics display. 

Repeaters, gateways, everything is possible. 

I am not happy with this displays, I planed before Eadog102 display, in final version I removed it, is more expensive. Sharp memory display are ugly and hard readable,.

about SWOT ... I write it for PCBway contest. I have no money and capacity for mass production.

To bootloader, burn bootIader is not very hard ... maybe is possible with arduino, I not tested because I have Atmel-ICE but it is not easy compared to atmega328 with USBASP. But work with samd21 and another atmel arms is easier, no crystals, more memory, no usb chip, cheaper like atmega32u4 ...

I need making money, they keep my family alive. I created this project because I want learn all this things desibed in STEM section, nothing more, project name is more joke like device for save word = it is toy for my self learnig and I share some results ;)

  Are you sure? yes | no

greenaum wrote 04/19/2020 at 17:24 point

Ah OK, it's more for an imaginary doomsday. Fair enough, still can be fun even if you only have a few units.

  Are you sure? yes | no

Kosma wrote 04/15/2020 at 16:41 point

irc ? probably better way are NNTP usenet

  Are you sure? yes | no

Ahron Wayne wrote 04/13/2020 at 17:05 point

Well, it looks sweet :) but is it hardened against EMP blasts?

  Are you sure? yes | no

bobricius wrote 04/13/2020 at 17:51 point


  Are you sure? yes | no

greenaum wrote 04/13/2020 at 18:37 point

Keep it in a metal tin.

  Are you sure? yes | no

Jdcrunchman wrote 04/16/2020 at 23:39 point

You are going to need a lot more then a metal tin....  Some of this USAF equipment I worked in while in USAF is pretty darn sophisticated...  The antenna is going to be your real problem....

  Are you sure? yes | no

Mike Szczys wrote 04/13/2020 at 16:51 point

Cover plate looks great. Is this a circuit board using aluminum substrate or something different?

  Are you sure? yes | no

Mike Szczys wrote 04/13/2020 at 18:09 point

Oh that's cool! Yeah, this laser engraving process makes so much more sense. Neat material.

  Are you sure? yes | no

Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates