Most vapour recovery units do not fail suddenly. They drift, performance declines slowly, and the change is only noticed when a measurement crosses a threshold or a test falls outside an expected range. By that point the cause is usually several months old and harder to isolate. The cure is a small set of metrics recorded consistently, chosen so that each one responds to a different part of the system.
The point of measurement is not reporting for its own sake. It is to make change visible while the cause is still easy to find and cheap to fix.
The four metrics worth tracking routinely
First, recovered liquid volume, recorded against throughput in the same interval. This is the single most informative figure because it normalises for trading activity, and a decline against throughput means the recovery process is losing effectiveness rather than simply handling less fuel.
Second, inlet conditions: temperature, pressure and, where available, hydrocarbon concentration. These explain why recovery has changed, since a condenser that is working correctly on a warmer inlet will recover less than one receiving cooler vapour.
Third, run time or duty cycle of the recovery unit, including blower hours and condenser operation. A unit that is running longer to recover the same amount is telling you something is degrading, whether that is a fouled surface, a leaking valve or a control setpoint that has shifted.
Fourth, condensate and water handling data: volumes collected, water drawn off, trap drain frequency. Liquid accumulation in unexpected quantities is a reliable early signal of a problem that will later present as a mechanical failure.
| Metric | What it detects | Sensible interval |
|---|---|---|
| Recovered liquid per unit throughput | Loss of recovery effectiveness | Daily or per delivery |
| Inlet temperature, pressure, concentration | Change in the load condition | Continuous where instruments allow |
| Blower and condenser run hours | Degrading performance or control drift | Daily |
| Condensate and water volumes | Carryover, leaks, seasonal load change | Daily during cold and wet seasons |
| Maintenance actions and findings | Recurring causes, hidden patterns | Per event |
Why a baseline matters more than a threshold
A threshold tells you when something has already passed a limit. A baseline tells you when something is moving. For a system that degrades gradually, the baseline is the more useful instrument, because it captures the slope rather than the crossing point.
The practical way to build one is to record the same metrics at a consistent interval for the first months after commissioning, when the unit is known to be in good condition, and to keep the conditions of each recording comparable. A figure taken during a busy afternoon is not directly comparable with one taken at 6 a.m. unless that is understood and allowed for.
Making the records useful
Records that never leave a notebook do not produce insight. The value comes from looking at them periodically, ideally at the same time as other routine checks, and asking a simple question: has anything moved in a direction that does not return?
- Plot recovered liquid against throughput rather than tracking volume alone.
- Note any change in fuel grade, tank configuration or throughput pattern on the same chart.
- Keep maintenance actions next to the readings, so a change can be matched to the intervention that followed it.
- Record season and weather where relevant, because condensation behaviour changes with them.
- Review the trend monthly, not only when something looks wrong.
Common measurement mistakes
Three mistakes account for most of the confusion. Recording volume without normalising for throughput, which makes a quiet week look like a failing unit. Comparing readings taken under different conditions as though they were equivalent. And recording only failures, which leaves the record with no information about normal behaviour to compare against.
A fourth, less obvious mistake is measuring too much. A long list of parameters that nobody reviews is worse than four that are reviewed every month, because it creates the impression of diligence without producing a decision.
Turning readings into a decision
A record becomes useful when it prompts a decision. The simplest discipline is to review the trend line at the same point each month and ask three questions: has recovered liquid per unit throughput moved, has run time per unit throughput moved, and has the pattern of condensate production changed. If the answer to any of them is yes, the next step is to identify which part of the system could explain it rather than to arrange a service visit on general principle.
Comparing two consecutive months is rarely enough, because weather, trading patterns and product mix all move figures around. Comparing the current month against the same month a year earlier, alongside the running trend, filters most of that variation out and leaves the movement that matters.
| Review question | If the answer is yes |
|---|---|
| Recovered liquid per unit throughput has fallen | Check condenser condition and inlet temperature first |
| Run hours per unit throughput have risen | Check for restriction in the vapour path and for off-design operation |
| The condensate pattern has changed | Check line slope, drains and trap behaviour |
| All three are stable | Continue the routine and keep the record intact |
Deciding what to record before changing anything
It is tempting to add instrumentation first and decide on the record afterwards. The reverse order works better: agree which four or five figures will be reviewed monthly, confirm where each can be sourced, and only then consider whether anything needs to be measured that is not currently available. Most sites find the gap is in recording rather than in measurement.
Frequently asked questions
How long before a baseline is useful?
Two to three months of consistent recording is usually enough to see normal variation and to distinguish real movement from noise.
Do we need new instrumentation?
Often not. Most sites already have the temperature, pressure and run-hour information needed. What is usually missing is the routine of recording it in one place and reviewing it.
What is the strongest single indicator?
Recovered liquid volume per unit of throughput. When that falls while throughput is stable, the recovery process itself has changed.
Luoyang Wohong Petrochemical Equipment supplies vapour recovery equipment and supports operators with commissioning data and performance review. If you would like help setting up a practical measurement routine for your units, get in touch.