Contents
Case Study

Murdoch had budget approval for an optical tensiometer. They chose a $7,485 smartphone system instead, then made contact angle calculate on rough, water-absorbing soil.

Last Updated
July 30, 2026
Industry
Agricultural and Soil Science Research
Location
Perth, Western Australia
Request Quote Book 15 Minutes with Abhimanyu
Validated against KRÜSS DSA100E Murdoch-CSIRO Innovation Research Fellowship Cited in peer-reviewed journals

A Murdoch-CSIRO research group at the Centre for Sustainable Farming Systems needed water contact angle on treated and untreated soil. Standard sessile drop workflows failed on it: uneven baselines, cluttered surface texture, and droplets that sank before capture. A 51 minute session with Droplet Lab's engineering team converted repeated calculation errors into working measurements, using the Tilted Angle app, Extended detection, and tight image cropping.

Abhimanyu Photo
Written by
Abhimanyu Bhandankar
Holds an MBA from Schulich School of Business and a BE in IT. He joined Droplet Lab in July 2019 and now leads sales and marketing.
CEO at Droplet Lab
Read More
Gurdeep-Saini-Photo
Technical Review by
Gurdeep Singh Saini
Holds a BASc in Mechanical Engineering (Ryerson) and an MASc from York University. He focuses on the custom AI behind the instrument.
COO at Droplet Lab
Read More
Abhimanyu Photo
Written By

Abhimanyu Bhandankar

CEO at Droplet Lab

Holds an MBA from Schulich School of Business and a BE in IT. He joined Droplet Lab in July 2019 and now leads sales and marketing.

Gurdeep-Saini-Photo
Reviewed By

Gurdeep Singh Saini

COO at Droplet Lab

Holds a BASc in Mechanical Engineering (Ryerson) and an MASc from York University. He focuses on the custom AI behind the instrument.

Primary outcome: A blocked analysis workflow, unblocked

soil images that would not calculate

Before calculation error
After 37° / 35° returned

On a representative soil image, the app returned a calculation error under Smart, Extended, and Manual point placement. The droplet occupied a small fraction of the frame, so contact points landed a few pixels off the true edge. After cropping the same image tightly around the droplet and using the Tilted Angle app to bend the baseline to the soil surface, the analysis returned left and right contact angles of 37° and 35°. The fix was image handling and mode selection, not hardware.

$7,485 vs an approved benchtop

Delivered configuration: base instrument, sliding angle module, precision dropper, dedicated Pixel phone, cuvette holder.

4 on one portable unit

Contact angle, surface tension, surface energy, and sliding angle, with lifetime licences included.

51 min one session

Length of the expert troubleshooting call that resolved the soil analysis blockage.

3 settings, not hardware

Tilted baseline, Extended detection, and tight cropping. The combination that made soil images calculable.

Executive Summary

Who

A research group at Murdoch University's Centre for Sustainable Farming Systems in Perth, Western Australia, led by Dr Tona Sanchez-Palacios, Murdoch-CSIRO Innovation Research Fellow. The group works on fluid fertilisers and on a biodegradable soil-applied treatment intended to change how water behaves at the soil surface.

Problem

The group had no contact angle capability at all. Soil water repellency was the quantity that would decide whether a treatment worked, and it was not being measured. Once the instrument arrived, a second problem appeared: real soil is a hostile substrate for standard sessile drop analysis. Uneven surfaces gave no straight baseline, surface texture confused edge detection, and droplets penetrated fast enough to be missed at the phone's capture rate.

Solution

A $7,485 Dropometer configuration covering all four measurements, plus a documented soil analysis procedure: Tilted Angle app for bent baselines, Extended detection mode for cluttered backgrounds, tight cropping before analysis, and an explicit rule for choosing between the Young-Laplace and Polynomial fitting algorithms.

Time to Value

First contact 15 July 2025. Demo 22 July. Vendor onboarding and purchase order 25 August. Instrument in Perth 17 October after customs. Onboarding call 23 October. Soil analysis procedure resolved 25 February 2026. The gap between delivery and working soil measurements was the real bottleneck, and it was closed in a single session.

Results

Images that returned calculation errors under every point-placement mode now return left and right contact angles. The team has decision rules it can hand to other users rather than instrument-specific intuition. The measurement campaign on treated versus untreated soil is underway; quantitative repeatability figures will follow the first full batch.

Highlights

0 → 37°/35°

Calculation error to returned angles on the same soil image

~50 cm

Soil column height the group needs to reach, driving current stand feedback

$7,485

Full four-measurement configuration, delivered to Perth

Quote teaser

"Switching to the tilted one, and being able to crop the photo, actually solves all the problems."

Murdoch University, at a glance

Research focus

Fluid fertilisers for agronomic purposes, and a biodegradable soil-applied treatment designed to change water behaviour at the soil surface

Products / applications

Biodegradable smart material sprayed onto soil to influence water behavior (supporting goals such as improved water-use efficiency, reduced irrigation needs, and yield resilience)

Measurement stage

R&D method development and formulation screening: quantifying wettability on treated versus untreated soil

Users (roles/shifts)

Research fellow and PhD researcher, with a wider team being trained on the workflow

Materials and surfaces tested

Fine sandy soils and coarse aggregated soils in tubes (~5 to 10 cm), soil columns (~50 cm), sprayed material deposited on paper, and leaf samples

Key constraints

Non-uniform surfaces with no straight baseline; rapid droplet penetration on repellent soils; sample heights up to 50 cm; portability and power for eventual field work

Real soil breaks the standard sessile drop workflow

Three separate problems, only one of which was about the instrument.

There was no measurement before this

Unlike a manufacturing QC case, there was no incumbent method to improve on. The group was not measuring contact angle at all. The soil project is roughly a year old, and the team identified early that they needed to quantify soil hydrophobicity to make any claim about a treatment working. The instrument was bought specifically to close that gap, before it became a bottleneck rather than after.

Why soil is a hard substrate

  • A non-uniform soil surface gives the software no straight baseline to lock onto
  • Surface texture around the droplet reads as edge candidates, so automatic detection picks the wrong contour
  • Droplets on repellent soil are asymmetric, so left and right contact angles genuinely differ, sometimes by more than 5°
  • On less repellent soil the droplet penetrates faster than the phone's 10 to 15 fps capture can reliably catch
  • Viewing angle matters: a dip in the surface on one side makes the apparent baseline sit lower than on the other side
  • Coarse aggregated soils cannot be flattened without destroying the property being measured

What was actually at stake

Water contact angle is the signal that decides whether a soil treatment has done anything. If that number is unreliable, formulation comparisons stall, and the case for scaling from lab to pot trials to field trials cannot be made. The group had a March measurement deadline and a team waiting to be trained on a procedure that did not yet exist.

What success would require

  • Angles that calculate reliably on rough, porous, absorbing surfaces
  • A procedure simple enough to teach to several team members quickly
  • Rules a user can follow without understanding the underlying fitting mathematics
  • A path toward taller samples and eventually field conditions

The fix was image handling, not hardware

What Was Deployed

A $7,485 Dropometer configuration: Flagship base model ($5,000), sliding angle module ($695), precision automatic dropper for controlled drop volume and flow rate ($1,495), a dedicated Pixel phone ($295), and a cuvette holder for liquids. A fabric sample holder for leaf samples was added after the demo. Lifetime software licences for all four measurement types are included, covering contact angle, surface tension, surface energy, and sliding angle.

The Origin

A demo booked specifically to stress-test the software

Dr Sanchez-Palacios found Droplet Lab through the website on 15 July 2025 and configured a quote himself using the online builder, listing the use case as fluid fertilisers for agronomic purposes. He then asked for a demonstration with a specific purpose: to determine whether the software could handle the sample types and protocols the lab intended to establish.

He was not shopping without a budget. He already had provisional approval to purchase an optical tensiometer. In his own words, the Dropometer looked good, but he wanted to check the software before committing.

The demo ran on 22 July. It changed the configuration rather than the decision: a fabric sample holder for leaf samples was added to the quote the same day. What followed was seven weeks of university vendor onboarding; public and product liability insurance, a modern slavery declaration, and banking proof on letterhead before the purchase order was issued on 25 August. The instrument reached Perth on 17 October after a customs hold.

Decision Rationale

Why not the instrument already approved

Three options, one of which already had budget sign-off.

Option 1 ·
$10,000+

Buy the approved benchtop optical tensiometer

Budget approval was already in hand. Established platform, established protocols, and a specification sheet that answers procurement questions without argument.

Approved and available, but permanently lab-bound. No route to 50 cm columns, pot trials, or field measurement.
Option 2
$0

Continue without contact angle data

Keep assessing soil treatments on downstream proxies and defer wettability measurement until later in the project.

The group had already concluded this was untenable. Contact angle was the quantity the whole comparison rested on.
Chosen
$7,485

Smartphone-based four-measurement system

All four surface measurements on one portable unit, with lifetime licences, an automatic dropper for repeatable drop volume, and a sliding stage for roll-off angle.

Four measurement types and a field-capable form factor, at a fraction of a benchtop's capital cost.

The decision was made on software, not price. Dr Sanchez-Palacios had approval for the conventional instrument and asked for a demo specifically to test whether the software could handle his sample types before he committed to the cheaper option. That is the harder test, and it is the one worth reporting. The follow-on cost of the choice showed up later and honestly: soil is a difficult substrate, and the group spent four months between delivery and a working analysis procedure. Closing that gap took one 51 minute session with the engineering team, which is the part of the offer a specification sheet does not capture.

Next Step

Measuring on a surface that will not cooperate?

Soil, powders, porous membranes, and textured coatings all fail the same assumption that standard sessile drop analysis makes. Send us a sample or an image that will not calculate, and we will run it and tell you what is actually going wrong. That is how the Murdoch workflow was resolved: a single image, emailed mid-call, analysed on our end while they watched.

Request Quote Book 15 Minutes with Abhimanyu

Before vs After

Metric Before After
Baseline handling Contact Angle app only, straight baseline. Uneven soil never produced a clean baseline for the software to lock onto. Tilted Angle app. The baseline bends to follow the actual soil surface.
Image framing Full frame capture. The droplet occupied a small pixel area, so contact points landed a few pixels inside or outside the true edge. Tight crop around the droplet before analysis, using the capture-screen framing box. Removes pixel-scale placement error.
Detection mode Smart mode against cluttered soil texture. Repeated calculation errors. Extended mode for cluttered backgrounds. Smart also works once the image is cropped well.
Algorithm choice No clear rule for Young-Laplace versus Polynomial. Polynomial for asymmetric droplets and whenever the needle sits inside the droplet. Young-Laplace for symmetric droplets clear of the needle.
Result per image Calculation error under Smart, Extended, and Manual placement. Left and right contact angles returned (37° and 35° on the test image).
Transferability Instrument-specific intuition held by one user. Written decision rules that can be handed to additional team members.

Rollout timeline

Timeline (high level)

15 to 22 July 2025: Evaluation

  • Website enquiry and self-configured quote; use case recorded as fluid fertilisers for agronomic purposes
  • Demo requested explicitly to assess software capability against intended sample types
  • Demo delivered 22 July; fabric sample holder for leaf samples added to the configuration

22 July to 25 August 2025: Procurement

  • University vendor onboarding: insurance certificates, modern slavery declaration, banking proof
  • Vendor approved 19 August; purchase order PO-021532 issued 25 August

September to 17 October 2025: Delivery

  • Assembly and shipment from Toronto; customs clearance in Canada and Australia
  • Instrument arrived in Perth 17 October; assembly guide and Dr Alidad Amirfazli's demonstration video supplied ahead of onboarding

23 October 2025: Onboarding

  • Live onboarding session covering project setup, density entry, manual and automatic injection, Bluetooth pairing, sessile stage levelling, focus, lighting, and baseline placement
  • Needle selection guidance: thinnest needle the liquid allows for contact angle, surface energy, and sliding angle; largest needle for surface tension
  • Calibration path established: measure needle outside diameter directly rather than relying on supplier figures

25 February 2026: Soil analysis resolved

  • Expert session diagnosing the calculation failures on real soil images
  • Tilted Angle app, Extended detection, and tight cropping identified as the working combination
  • Young-Laplace versus Polynomial decision rule documented for the team

What has been demonstrated, and what has not

Test method

Sessile drop water contact angle on treated and untreated soil. Tilted Angle workflow with a bent baseline set slightly above the contact points. Polynomial fitting for asymmetric droplets, which reports left and right angles independently and tolerates the needle remaining inside the droplet. Young-Laplace reserved for symmetric droplets clear of the needle, where it also returns droplet volume.

Sample size and operators

Measurement campaign in progress. Multi-user training underway. Sample counts and operator counts will be reported after the first full batch.

Repeatability / reproducibility

Not yet reported. The group is running its first standardised batch.

Notes / assumptions

Soil roughness and porosity produce genuinely asymmetric droplets, so a difference between left and right angles is a property of the sample rather than an analysis artefact. Contact angles on rapidly penetrating soils are time-dependent and capture-rate limited at 10 to 15 fps; a higher-speed path exists if droplet dynamics become the object of study rather than a nuisance.

Outcomes to date

Measured Outcomes

37° / 35°

Analysis unblocked

A representative soil image that returned calculation errors under Smart, Extended, and Manual point placement returned left and right contact angles once cropped and analysed with a bent baseline. Same image, same instrument, different handling.

3

Root causes identified

Baseline geometry, pixel-scale contact point placement on an uncropped frame, and detection mode against cluttered texture. All three were addressable in software settings and image handling.

10 to 15 fps

Capture limit quantified

The phone's capture rate is now a known constraint rather than an unexplained source of missed droplets, with a defined higher-speed path if dynamic penetration becomes the study object.

Operational Outcomes

The blocking issue was procedure, not equipment. Cropping tightly around the droplet and bending the baseline to the surface converted failures into measurements without any hardware change.

The group now has transferable decision rules when to use Polynomial versus Young-Laplace, when to switch to Extended detection rather than one person's accumulated intuition. That is what makes a multi-user rollout possible.

Field deployment constraints are now specific and actionable: stand height for 50 cm columns, backlight battery life at working brightness, and phone-holder clearance. All three are in Droplet Lab's hardware feedback loop from this deployment.

Plan your own

What wetting data is worth in a formulation lab

Murdoch's internal budget figures are not disclosed, and their measurement campaign is still running. The calculator below models your lab, not theirs. Defaults are anchored on Murdoch's delivered configuration.

Formulation ROI Snapshot

Estimate saved iterations and lab cost.

Soil prep, treatment material, column setup, or field plot cost.
Lab prep + test + analysis.
Directional only. Depends on how early wetting data enters your screening loop.

Result

~0
Iterations saved / month
~0
Monthly savings
~0
Payback period
~0
Year-1 net benefit

Monthly savings = materials saved + technician time saved from reduced iterations.

Financial Context

What the choice cost and saved

Murdoch's internal budget figures are not disclosed. The comparison below is the configuration cost against the class of instrument that had already been approved.

Capital cost · $7,485 delivered

Full four-measurement configuration including instrument, sliding angle module, automatic dropper, dedicated phone, sample holders, and lifetime software licences. Benchtop optical tensiometers in the class Murdoch had approved typically start well above this.

Infrastructure · none required

No dedicated bench, no environmental enclosure, no service contract, and no specialist operator. The instrument travels between lab spaces and, once stand and power constraints are resolved, to pot and field trials.

Method breadth · four measurements

Contact angle, surface tension, surface energy, and sliding angle on a single unit with lifetime licences, against separate configurations or modules on conventional platforms.

Figures describe the delivered configuration, not a modelled return. Research procurement value depends on grant structure, existing facilities, and shared instrument access. Contact us for a configuration quote.

What the team said

"Switching to the tilted one, and then also just being able to crop the photo, actually solves all the problems. I think that has actually answered all the questions we have, because now we know how to set it up."
Samantha Viljoen - PhD researcher

An open feedback loop

Delivered

Four-measurement Dropometer configuration with automatic dropper and sliding stage, delivered to Perth
Onboarding session covering assembly, injection control, Bluetooth pairing, and baseline placement
Soil analysis procedure: tilted baseline, Extended detection, tight cropping, and fit selection rules
Automatic dropper assembly guide supplied, covering syringe barrel alignment with the internal limit sensors

In Pilot

Multi-user training across the wider research team ahead of the measurement campaign
USB-C dongle for direct mouse control of the phone, for finer contact point placement
Needle outside-diameter calibration by micrometer, enabling Young-Laplace work alongside Polynomial
Backlight operation at 1 to 2 percent brightness to extend battery life toward field use

Planned

Freestanding, vertically adjustable phone stand for soil columns up to 50 cm
Phone-holder redesign to avoid contact with the handset volume buttons during capture
Glossary entries for Young-Laplace and Polynomial fitting in the Surface Science Hub
Extension of the workflow from tubes to columns, and from lab to pot and field trials
Logo
Logo
Logo
Logo
Logo
Logo

Next Step

Measuring on a surface that will not cooperate?

Soil, powders, porous membranes, and textured coatings all fail the same assumption that standard sessile drop analysis makes. Send us a sample or an image that will not calculate, and we will run it and tell you what is actually going wrong. That is how the Murdoch workflow was resolved: a single image, emailed mid-call, analysed on our end while they watched.

Request Quote Book 15 Minutes with Abhimanyu