Test Chamber for Validating Sensors

A project log for Breathalyzer for Blood Glucose

Portable, simple and inexpensive test to provide Type I diabetics an indirect measure of blood glucose via breath analysis.

tom-meehanTom Meehan 09/04/2019 at 01:060 Comments

Semiconductor gas sensors are relatively inexpensive but they come with one caveat - you have to calibrate them with known concentrations of the gas(es) you are using the sensor to detect.  It is one thing to detect if a gas is present but if determining its actual concentration is the objective (which it certainly is, in this instance) then we need to test each sensor against different known concentrations to determine their sensitivity to to our target gas (acetone) as well as the relationship between sensor values and gas concentration.

For this chamber, I have a number of design considerations:

Presently, I am finishing a chamber design and build that uses an inexpensive glass aquarium (I did remove the upper plastic rim since I believe it is made from ABS - very absorbent to acetone) and mounting sensors and fan inside, along with a watch glass for the evaporation of acetone.  In addition, a Calorique heating element external to the tank and insulation surrounding the whole structure.

The sensors that I am currently including are: BME280 (for temperature, humidity and barometric pressure), MQ 3 gas sensor, MQ138 gas sensor and TGS822 gas sensor.

The chamber has a silicone gasket (fabricated) to seal it. Wiring is routed into the tank via holes drilled in the glass and sealed with silicone.

Below are a few pictures of my progress so far:

Keep tuned, I'll have additional updates tomorrow!