SteamTrust · Flagship industrial solution

Find the steam you’re paying to lose.

Wireless tags on every steam trap, a ranked repair list every morning, and emissions evidence your auditor can check.

Estimate your losses Talk to us

The SteamTrust repair list for Plant 2, ranking steam traps by yearly cost, with trap 114 blowing steam at the top and fuel saved after repairs.

Illustrative interface

Trap 114 · Line 2 just now

Blowing steam, costing about $3,100 a year. Fix first.

A stuck-open trap is invisible.

Most plants check their traps once a year with a clipboard. The expensive failures don’t make a sound.

Stuck open, nobody notices

It passes live steam to the drain 24 hours a day. The only symptom is a slightly higher fuel bill across the whole plant.

15–30% fail without a programme

The US Department of Energy benchmark for plants with no formal trap maintenance for three to five years.

Found once a year, at best

With an annual walk-round, the average failure runs for about six months before anyone sees it.

Now the loss is a number someone has to report.

Across Southeast Asia, emissions disclosure, carbon pricing and export buyers are asking for measured figures. An estimate is no longer enough for an auditor.

Lead with the fuel bill. Close with carbon.

Emissions disclosure in Malaysia and Thailand
Carbon pricing in Vietnam and Indonesia
EU carbon border levy
Fuel prices staying high

A SteamTrust evidence record for trap 114 showing temperatures before and after repair, steam, fuel cost and CO₂e saved, ready for audit.

What are your traps costing you?

Move the sliders. Every figure is computed from published benchmarks and our business-case assumptions, never generated.

Failed traps leaking now: 45
Steam lost a year: 6,480 t
Fuel money lost a year: $129,600
Recoverable once repaired: $71,280
CO₂ avoided a year: 672 t

Estimate only. Modelled on 144 t of steam a year per failed trap and 55% of the loss recovered once traps are repaired. Your site survey replaces these assumptions.

Get a site survey

One system. Three people who care.

Plant engineer

A ranked repair list every morning, instead of a yearly walk-round with a clipboard.

Sustainability manager

A measured, auditable Scope 1 reduction for the published report.

Finance director

Savings measured continuously after each repair, not claimed once and forgotten.

Sense. Decide. Prove.

Off-the-shelf radio hardware, our software, and a hard line on where AI is allowed.

01
Steam trap
New sensor or an existing signal
02
Radio tag
LoRaWAN, battery powered
03
Gateway
On the roof, 4G back-haul
04
Platform
Classifier and evidence ledger
05
Dashboard & report
Repair list, PDF and Excel

Every trap is classified asHealthyBlowing steamBlockedCycling too fastIndeterminate

Sense

Tags read each trap’s temperature, or an existing 4–20 mA, Modbus or switch signal, and report every 15 minutes.

Decide

A trained model classifies each trap. Below 85% confidence it is marked indeterminate, and no saving is claimed.

Prove

Readings and repairs go into a hash-chained evidence ledger. Reports use a versioned method; a language model never writes a number.

Four ways to connect.

We use whatever your plant already has, and fit new sensors only where it has nothing.

Read a serial cable

Where traps already report over Modbus, one tag reads a dozen of them from the register.

Copy a 4–20 mA signal

A rail-mounted isolator copies the signal. The original wire carries on to your control system untouched.

Read a switch

Dry contacts on traps that already have them are read directly.

Clamp on a sensor

Most traps have no instrumentation, so we clamp a temperature probe on the pipe. No cutting, no rewiring.

We never cut into your control loop. Our system watches and advises; it never switches anything and is not part of any safety system. If our device fails, your plant carries on.

Evidence an auditor can follow.

The report your sustainability manager is buying. Conservative by design: it states its uncertainty and never claims a saving before a repair is confirmed.

Steam system emissions reduction · Verification pack

Sample · illustrative figures

1 · Summary of claim

Continuous monitoring of 248 steam traps identified 43 failed traps. 31 were repaired and confirmed back in normal operation. Verified reduction: 486 tCO₂e.

2 · Boundary and basis

Scope 1, stationary combustion · GHG Protocol Corporate Standard · ISO 14064-1 · IPMVP Option A

3 · Calculation chain

Step	Method	Result
Trap condition	Classifier on 15-minute surface temperature; confidence ≥ 0.85 to book a loss	43 failed of 248
Steam loss	Orifice flow at rated orifice size and measured header pressure	1,847 t of steam
Fuel equivalent	Measured boiler efficiency, feedwater temperature and blowdown	5,301 GJ
Emissions	IPCC 2006 emission factor for natural gas	486 tCO₂e

4 · Uncertainty

±14% at 95% confidence. The reported figure is the lower bound, not the midpoint.

5 · Evidence for the verifier

Raw readings, classifier version, repair confirmations and emission-factor sources, kept 7 years in a hash-chained ledger.

Scope 1, 2 and 3, simply.

What you burn

Your boiler’s fuel. A leaking trap makes it burn more, so fixing traps cuts Scope 1 directly.

What you buy

Steam bought from an industrial-park utility still leaks through your traps. Same fix, different line on the report.

Everyone else

Your export buyers’ Scope 3 includes your Scope 1. That is why their questions now reach your plant.

Two ways to buy.

Both start with a site survey and a quote. We don’t publish prices.

Subscription

Buy the tags once, then subscribe to the dashboard, alerts and reports. Our cloud, updated continuously.

Ask for a quote

On-premise licence

Own the software on your own servers, with yearly maintenance. For data-residency and no-cloud IT policies.