Smart Monitored Water Bottle

The Project

We set out to make a water bottle which tracks consumption for elderly people with neurodegenerative diseases such as Alzheimer's' or Dementia. First, we coded in Micro Python on Raspberry Pi Pico with the HX711 weight sensor and assembled hardware using techniques such as soldering. Then we coded the HX711 to print raw values, as masses were added or removed from the sensor. Next, we coded a graph in Python with the change in raw data as set masses of 100g to 600g, using an interval of 100g. We then found the gradient and the offset offset, we then calibrated HX711 to read out in grams in order for easily analysis by carers. After this, we designed the first prototype on FUSION 360 before 3D printing and assembly of parts. We have also developed a charging system. In the future, we will need to integrate Wi-Fi with the ESP32 PicoWireless and to allow it to connect to a wireless network. This will transfer data from Raspberry Pi Pico to a website or an app this will track the data for carers. Due to COVID complications, the project has not quite yet been finished and is still a work in progress.

Hardware
Health
Community

Team Comments

We chose to make this project because...

With an aging population, memory issues and the problems such as dehydration that accompany them are becoming an increasing issue today. We decided to design a smart water bottle to monitor hydration and alert patients to drink regularly, specifically to support those with neurodegenerative issues.

What we found difficult and how we worked it out

To charge the bottle, there’s the risk of electronics mixing with water. Using Faraday’s Law of Induction, we used electromagnetic inductive charging with receiver and transmitter coils transferring energy, inducing current and charging the bottle wirelessly through a waterproof electrical insulator

Next time, we would...

We would work using a different sensor as our current one is not very sensitive and has a low resolution. For example, using a water level sensor instead of a strain sensor load cell. We would also integrate on board Wi-Fi into the development process and use EM induced LEDs to show charging state.

About the team

  • United Kingdom

Team members

  • Megan
  • Elisabeth