This is just a case and demo software for a small, USB-powered Geiger counter that has a display and speaker to alert if counts per minute exceed a threshold. The intent is just to let it run and sound an alarm if the radiation level gets too high. Cost is around $100 for the parts from Amazon, Adafruit, Mouser, etc. It also reports accumulated counts every minute on the USB port in plain text, one minute's reading per line, if anyone wants to log or analyze.

The Geiger counter is a widely available and fairly cheap "Portable Handle Geiger Counter 0.01 μSv/h DIY Kit Nuclear Radiation Detector With Geiger-Müller GM Tube Gamma Beta Ray Suitable for Arduino". I included a mounting pocket for an Adafruit 5V ItsyBitsy (https://www.adafruit.com/product/3677), a small speaker (Mouser.com part number 490-CMS151135078L100), an Adafruit 128x128 OLED display (https://www.adafruit.com/product/1431) and a couple of capacitive touch switches (TTP223 Capacitive Touch Switch Button Self-Lock Module for Arduino) in order to build a complete system that graphically and numerically tallies counts and alarms if need be.

The display frame mounts to the main box using 2x M3x5 screws. Four more M3x5 screws secure the Handle Geiger Counter into the box. The holes to mount the cover that came with the Geiger counter were offset slightly on the cover so other covers may or may not carry that same offset. Hopefully they are all identical but no guarantees.

The ItsyBitsy records the Geiger counter events and keeps a running minute by minute count. If levels cross a threshold it sounds an alarm using the pancake speaker and lights up the OLED display (wip). A computer connected to the micro-USB port on the ItsyBitsy can log and timestamp activity for longer history but this is just a simple running display of counts per minute vs minutes of time. Touching a capacitive touch switch lights up the OLED for a few minutes as well as switches between graph and numeric displays (wip). I blank the display and only wake it up when a capacitive switch is touched or if in alarm since it's OLED.

The capacitive switches, ItsyBitsy, and display are all held in place with thin double-stick foam tape. I also cut a piece of card to put between the ItsyBitsy and the Geiger board to keep the wires insulated and away from any high voltage spots on the back of the Geiger board. The card can also prevent screws or nails in the keyhole slots from being able to touch the Geiger board.

The main box is 119 mm x 70 mm x 27 mm tall. The display housing is 77 mm x 52 mm x 9 mm tall. The USB cable I’m using is a CableCreation Right Angle Micro USB 2.0 Braided Cable, 90 Degree Vertical Right USB 2.0 A Male to Micro USB Male to reach through the slot to the ItsyBitsy and then just hang straight down.

Photos showing wiring and the display, touch inputs, and speaker are now up. The Geiger counter connects to the ItsyBitsy using USB, G, and pin 7 on the Itsy board to 5V, GND, and VIN on the counter board (P3). Those connections aren't shown in the new photos I just posted. Due to a change in naming, and because the ItsyBitsy has hardware MOSI, and SCK, I had to rename the references in the Adafruit display demo (https://learn.adafruit.com/adafruit-1-5-color-oled-breakout-board/wiring-and-test) to MOSI, and SCK (from MOSI_PIN and SCLK_PIN) since that is how the pins are described in the ItsyBitsy header file, and comment out the defines in the code since they are already defined. DC_PIN is 8, CS_PIN is 6, and RST_PIN is 4. That's all the graphics demo exercises. I haven't tried reading from the SD card but the wires are there. Also, CS_PIN (6) connects to OC (OLEDCS) on the display board.

The text on the display (A) are the numbers for the last completed count in both cpm and uSv/hr using 151.0 as the conversion factor for the M4011 tube. I’ve also seen 153.8 listed but it’s an approximation anyway. (B) is the current count growing in. The current count bar gets updated as counts accrue during the 60 second integration period. (C) is the display of past counts. Two hours (120 individual cpm measurements) worth of counts are shown on the display and scroll off the left as new counts are added. (D) is a countdown bar for the current one minute count integration period. When it is full height, the minute is just starting. It is shortened proportional to the elapsed time and the integration period ends and the display is updated when it zeros out and a new integration period starts.

The software may have bugs but looks ok so far. It now also computes and draws a least squares straight line fit through the data. It also displays the average count and slope of the counts numerically.

The firmware has an alarm level now and will sound if the average count goes above a threshold. There's also a current count value in the lower right corner of the display, a red horizontal bar showing the alarm level, and the best fit line of the count data is now black for better visibility. The alarm level is still a #define. More to go but 0.6 exercises everything except the SD card reader/writer so all of the hardware can be tested. Higher functions are still missing. The alarm is a bit softer than I thought it would be. Harder to hear outside a room.