03. Case study

Not a pilot.
A working field.

Argus runs every day on producing heavy-oil wells in the Parrylands area of Trinidad, operated by Green Flame Resources. Progressing-cavity pumps on variable-frequency drives, satellite uplink, no control room. The dashboard on this site is the same software that runs that field. Screens marked SIM carry sample figures: drive and energy values are measured, production volumes come from the crew’s tank readings.

Monitoring has run continuously since June 2026, first on Pad 6 and then on Pad 4. In the interest of being straight with you: this field is where Argus was built, not an arm’s-length customer — we explain the relationship here.

Jun 2026Running since

5Wells live

2Pads

2 sTelemetry

0On-site servers

The starting point

Wells you can only
see by driving to them.

01Before ArgusA pump could trip at 2 a.m. and go unnoticed until the morning round.

02Before ArgusEvery reading meant a truck, fuel and a drive to the pad.

03Before ArgusSlow mechanical trouble showed up only once it had already broken something.

04Before ArgusNo number, no history, no way to compare one well against another.

What was deployed

An afternoon per pad.
No trenching, no server.

01Field cabinet per padA sealed enclosure reads each variable-frequency drive over Modbus and turns it into clean telemetry. Nothing was replaced; the existing drives stayed exactly as they were.

02Uplink over StarlinkThe pads sit behind satellite and carrier-grade NAT. Argus connects outbound through an encrypted mesh, so there is no static IP, no port forwarding and no on-site PC.

03Live in the cloudTelemetry from every well lands in one dashboard at a 2-second update, with history, comparison across wells and reports.

04Alarms to the phoneDrive faults, power loss, pump-off and torque anomalies push straight to the operator, typically inside a minute.

What it caught

The failures that
used to arrive late.

These are events from the running system, not scenarios.

A slow torque creep that fixed limits miss

On one well torque climbed from about 45 to 58 percent over roughly five hours. No single 15-minute step was large enough to trip a conventional threshold. Argus runs a second, slower reference window specifically for this shape and raised it as sanding, with the from and to values and the time.

Power loss told apart from a lost link

A satellite blip and a genuine outage look similar from a distance. Communication loss is classed separately, so a dropped uplink never masquerades as an equipment failure, and a real power loss sends a recovery message once it clears.

Wells that stop without anyone noticing

A stopped or offline well shows its real state, runtime and energy instead of quietly disappearing from the view.

Where it stands

Running, watched
and still growing.

The field runs in monitoring mode with remote control gated behind a site-wide safe lock, so the wells stay under local authority while the platform proves itself day after day. Downhole level and surface pressure instrumentation is going in next, and the same kit extends to further pads without adding infrastructure on site.

Want the same view of your own field? Email info@argusautomation.tech or request a demo below.

Put your field online.

First well live in days, not quarters.

Not a pilot.A working field.

Wells you can onlysee by driving to them.

An afternoon per pad.No trenching, no server.

The failures thatused to arrive late.