Join GitHub today
GitHub is home to over 28 million developers working together to host and review code, manage projects, and build software together.Sign up
Water Quality Sensors
Water Quality Sensors
- List of parts
- Power consumption
- Responses in extreme events
- Digitial Communication
- Data Sheets/Commands
- Board Enclosure
- Coax Extensions
- Sensors Enclosure
- Water Proofing
- Coax Link
- SNODE Link
- Water proofing test
Open-storm board supports the following water quality sensors manufactured from atlas scientific
- Dissloved oxygen
- Oxidation reduction potential
Field Deployments in Ann Arbor, MI
|Mallets Creek||County Farm Park|
List of Parts
Additionally water quality sensors would need the calibration liquids from atlas scientific.
Water quality sensor has three components
- Probe (water quality sensor)
- EZO Isolation board
- Circuit board
Probe (i.e sensor) in immersed in water, where it measures a particular attribute. This reading is reported from the probe as voltage (a analog reading) . This analog reading is converted into a digital reading by the circuit board. Circuit board would differ for each probe. If coax cable (which transmits the analog reading from the probe to the board) is not properly connected, the readings reported might be significantly skewed. It is advised to use a isolation board. Isolation
5 water quality sensors are connected in parallel to the board, even though they are powered using 3 different pins on the board.
- SDA, SCL, GND are connected in a single circuit
- 5v-power is powred using 3 pins. Each pin on the board is restricted on the amount power that they can draw from the board.
Following Wiring convention is followed though out the sensor
- SDA - Geen
- SCL - White
- Power - Red
- Ground - Black
Detailed wiring instructions are linked here.(Link the wire pdf)
Preparing the parallel circuit
|Snip the wire ends||Soder the lead||Crimp the wire end|
|![IMG_3375.JPG](/:storage/56o4r7lii7bep14i.JPG =600x700)||![IMG_1187.JPG](/:storage/h0y6ehaulmyam7vi.JPG =600x700)||![IMG_9412.JPG](/:storage/4gzvcvia80jjdcxr.JPG =500x700)|
Make sure the flap is raised, this flap acts as a lock that holds the crimped wire in the pin.
|Crimped wire into the 5pin holder||5pin female holder fixed to circuit board||They are locked in place|
Sensors when connected to the open-storm board consume a average of 330 mA for ~7-10 sec. These measurements vary based on firmware used.
While these sensors are out in the feild, there could be instaces when the sensors report skewd values. The following readings are what you can expect when such events happen.
Response of the circuit when the coax cables are not connected
Response of the circuit when shorted
|Conductivity||Really high value|
Response when water interfears with the circuit
In water sensor is short
|Ph||2.63 Keeps decreasing|
|DO||Rusults seem consitant|
Response in when not submerged
These repsonse tend to vary drastically. But once DO sensors are calibrated, they tend to show a value of ~8 when the sensor is dry
$$I^2C$$ protocol to communicate with the sensors. Detailed implementation can be found in perfect cell/atlas_wq_sensors.c.
These sensors by default are in UART mode, they can be converted to $$I^2C$$ though i2c,command. For more details refer data sheet.
Futher information about the operation and capabilities/limitations of the sensors can be found in the data sheets (linked below)
Open-storm board is encolsed in a water proof encolosure to protect the electronics from elements (rain, snow ...)
Depending on the feild conditions, distance between the sensors and node is adjusted. Currently 25ft and 10 ft coax extesnions were used. To build a differnt node, only the coax and conduit lenghts in the above order have to be altered.
Water Proofing the coax link
Coax link (coax extension + wq sensors) needs to water proofed. Any water in the link skews the reading.
Two layers of water seal is used to prevent the water from damaging the coax connection.
- Self Adhesive Tape
On top of the self adhesive tape multiple layers(3 layers) of the liquid tape is added to reinforce the seal.
- Liquid Tape
Water proofing the conduit
Coax extenders are enclosed in a flexible conduit to avoid any external damage.
To protect the sensors from the gravel and other things in the stream they are stored in a pvc tube. This thick walled pvc tube has holes in the bottom, where sensors are held in place by rubber stopper. These holes allow the water to enter into the enclosure.
Enclosure of the coax cables is conneted to the sensors enclosure using a thereaded connector.
Seal the coax extender link
- Water Sealing Putty
- Liquid Tape Layer
Water sealing putty is appied to seal the pcv link and allowed to set. A layer of liquid tape is added on top this for prevent any seepage. Care must be taken near the coax ends.
Water Proofing Test
Once the both the coax and pvc links are sealed, their integrity is tested by taking a reading when the coax links are dry and then the entire node is immeresed in water and readings are checked periodically. If the readings are consistent, sensors are ready to be deployed.
- Coax Seal Testing
Sensors need to calibrated before they can be deployed in the feild. Each one of the sensors have a unique method. Refer to calbration section in the data sheets.
Once these sensors are tested and calibrated they are placed in the enclosure.
Deployment procedure might depend on the feild conditions. But in general they involve a pole mounted node and the wq sensors deployed such that they are immeresed in water for major part of the storm. But if they are placed to close to the bed, they might incur damage though contact from mud.
General Issues experienced
- Protocol lock reset from UART to $$I^2C$$ This issue has been addressed in the firmware update from atlas.
- Coax seal integrety In the event water enters the coax connection. Sensor measurement are skewd, link have to be dried and redeployed.
- Conductivity sensor check Conductivity sensor is just measuring the current in the water. A quick test to test its integrity would be to check for a short on the sensor end (copper colored part) and coax cable
- Sear force seems to damage ph and conductivity sensor. This issue is yet to be addressed.