Sap Flow Meter Updates

Sap Flow Meter Updates

October 19, 2021 -

Author: Ryan Christensen

Currently we are in a major revision change after the deployments at Oak Creek which showed a lot of issues within the design that was used for deployments between July 19 and August 6. This included things such as general user debugging, battery monitoring, as well as some adjustments needed for the sensor to function properly within Douglas Firs.

The Version 6 revision focused these key points:

• Coulomb counter for monitoring battery life, as well as an automatic shut off in case the battery level goes too low
• RGB LED for visual user debugging, allowing ease of use and quick identification of errors.
• All board to wire connections now use keyed rectangular connectors to prevent reverse current, or mismatch.
• Improving the FlexProbe sensor to work better within Douglas Firs with longer probe lengths.
• Sparkfun I2C Differential Drivers also had a revision, making connections much cleaner with dedicated connections.

We are currently in the assembling and integration stage, hoping to get it online and ready to compare against the Version 4 which was used during the summer testing. As for HardwareX, the document has been complete and submitted, and the team is waiting to hear back from an editor, hopefully soon!

July 19, 2021 -

Author: Ryan Christensen

• The HardwareX is coming along nicely and we are pushing forward to get it out soon!
• There are currently two fully functioning probe systems, with a third on the way. The only limit right now is the battery count as we only have one battery in the field.
• A problem has been identified with some of the current devices used, specifically 8 of the 13 I2C differential drivers from SparkFun that we are using were broken, either the SDA or SCL pins shorted to ground, or the chip was overloaded and fried.
• We have planned a deployment at Corvallis' Oregon State Oak Creek, this week.

Additionally,

• The Zenodo and Github will be updated to reflect the current version that is being deployed in the field.
• Research and Development is being done into some of the Future Steps focusing cost reduction, power efficiency, and ease of use.

---

February 2, 2021 -

Author: Ryan Bohl

• The first draft of the Hardwarex paper has been completed!
• There was a build party on 1/26/2021 for assembling one more probe setup and documenting the process. Videos of the assembly can be viewed in the OPEnS drive in the Sapflow meter project folder. There is a folder here titled "build videos." It was difficult to get footage in focus. Therefore, several photos were taken. These can be found in: 2021 uploaded files --> upload --> Hardwarex documentation --> 1/26/2021 build
• The individual PCBs have been fully assembled, but there have been difficulties in connecting them together. The link to the 20 awg wires within the BOM should probably be changed. It is very difficult to feed these wires through the small through holes. This is the last major step in the hardware assembly. Testing this hardware should be easy, as most pieces from the past functioning setup can be used.

Additionally,

• I am leaving the project within a week. I have revised the wiki to more closely reflect the current state of the project.
• I will work to create another functioning unit to store in the lab cabinet where it can be accessed by Jackson.
• Other Materials (Non PCBs) will be stored in OPEnS shed
• I will commit the current revision of the arduino code into github where it can be linked to the hardwarex paper through Zenodo
• Further Research and Development on the project is put on hold.

I will still be able to be reached through the hardwarex revision process.

---

January 12, 2021 -

Author: Ryan Bohl

• New probes have been rebuilt. Removing watchdog timers and increasing precision have improved testing results
• Project is ending current RD phase. There is a current focus on the Hardwarex publication.

Deployed 1 probe into 1 tree:

• Modified sleep times to reduce current draw and increase battery life
• New flex probes were manufactured with increased thermal paste
• Removed LoRa functionality in previous deployment. We hope to test LoRa again with more RD time

---

May 12, 2020 - Field Testing Unsuccessful

Author: Marshal Horn

• Field testing uncovered many hardware issues.
• Hardware and firmware changed significantly since lab testing.

Deployed 9 probes into 9 cherry trees. Ran into several issues:

• Current consumption was much higher than previous tests, due to differential I2C drivers
• Thermal resistance of probes was greatly deteriorated due to poor contact inside the stainless needles
• Possible hardware issues related to RTC time
• LoRa range was only 500m with 1/4 wavelength whip antennas

---