What is a Percolation Test?

A percolation test, or Infiltration test is when an engineer, Geo, or technician measures the rate at which water (or other fluid) flows into the ground of a measured hole. The idea is to understand the soil's hydrological characteristics at the layer where the test is conducted. The results of this test are either a Percolation rate or an Infiltration Rate. I am not going to get into the specifics of the math or the difference between these rates, but I will show you how the sensor collects these rates. This article will introduce a solution to some problems that are encountered during the collection of Percolation data, and presents some novel use-cases that are valuable for deep borehole use.

Hardware

The Percolation Pod consists of a microcontroller, hardwired into a digital sensor in a custom housing. Overall, the cost of all these components is fairly reasonable, with the most expensive being the electrical components. The wires and housings are all PVC pipe, and custom 3d printed flanges and containers. The sensor was built so that it could be attached to the end of a standard measuring tape and lowered into a borehole or auger hole. The whole sensor is powered by a typical Cell Phone style charging bank with a Type C charging cable.

Software

The code for running the test is conducted onboard the microcontroller. It collects data analog through a whetstone bridge and is converted to digital with a double-channel ADC. The collection software that bridges the sensors to the output is written entirely in Python, and executed with some onboard Linux logic. After collection, the data is stored. CSV format on an external USB stick, which can then be ejected and post-processed in Excel or Google Sheets to create distance vs time graphs. To control the collection rate and other factors, you must remote access the microcontroller or have an onsite computer and the login details.

Testing

The testing is ongoing, There are some issues with the ambient temperature affecting the resistance of the wires involved with the setup, this means at changing temperatures, the data is affected proportionally. As a result, every graph printed by the sensor is scaled differently in the distance domain (time remains the same). From what limited thought I have given to solve this problem, I would hypothesize that the best hardware solution consists of two secondary calibrating sensors that will enable us to measure a scaling factor before every test. As such, it would become a two-step test; one to establish the scaling factor, and the second step to collect data. Other options include wiring in a thermistor to measure ambient temperatures, which may simplify operation, but my gut is telling me that is more trouble than it is worth. The last solution is to let the scaling factor simply be post processed out and have a collection procedure that determine the scaling factor based on the data. This is the simplest solution.

Data

The data that is collected from the sensor is pretty easy to understand, it is delivered in a distance vs time graph, with the slope being the flow rate. Take a look below at a simple graph showing some real-time data. Note the scale on the y axis and the nature of the data points, there is a lot of noise and pre test data collection in the uncleaned graph above. This noise and useless data was cleaned off and presented with an appropriate scaling factor in the bottom graph.

Conclusion

The main idea behind this sensor is to automate the data collection, as the current methods are time-consuming and inconsistent from person to person. Not only that, but every jurisdiction seems to have different published methods of Perc or Infiltration testing. It is very easy for poor data integrity when collecting by hand. This industry is pretty much relying on word of mouth; by creating a device that collects data, we can ensure consultants and technicians are collecting data properly. The secondary advantage is overcoming the limitations of conducting the test by hand, as it is limited by the depth and width of a test hole. The person conducting the test has to be pretty much at the elevation of the layer being tested to read the measuring tape. The Percolation Pod enables us to conduct tests on deep narrow holes and even during a drilling program at any depth by simply lowering the sensor into the hole and clicking a button. This has particular application in commercial construction applications with basements as the hydrological data is critical for the design of the drainage/foundation waterproofing, or stormwater management design.

Acknowledgements

This research and on-site testing couldn't have happened without the partnership with Phillips and Associates who provided the opportunity to test the sensor in real-world applications.