How to prepare your station to pass the tightness test on the first try

An tightness test that fails is not just a technical problem: it is an event that triggers regulatory deadlines and suspends part of operations until it is resolved. The good news: 80% of failures are avoidable with proper preparation in the 2-4 weeks before the test.

This article is the practical preparation guide: what to do, in what order, and what common mistakes lead to failures that should have been avoided.

Why stations that "should" pass fail

In industry practice, failures concentrate on four avoidable causes:

1. Accumulated water in sumps, spill buckets, or the tank (not detected before the test).
2. Degraded seals in components other than the tank: sumps, spill buckets, flex connectors.
3. Incorrect tank configuration during the test (level out of range, presence of vapor).
4. Incomplete documentation of the test equipment or the site, which the inspector rejects.

What almost never causes a failure: the tank itself when it is relatively modern (post-2000, FRP, or lined steel). Tanks do fail, but rarely without having given prior warning signs.

The ideal preparation: a 4-week schedule

Week -4: Initial diagnosis

Goal: have a complete inventory of the current condition before starting any repairs.

Checklist:

[ ] Complete visual inspection of dispenser sumps and spill buckets.

[ ] Draining and cleaning of any accumulated water.

[ ] Verification of levels in each tank (they must allow testing between 20-95% during the test).

[ ] Review of the ATG alarm log for the past year.

[ ] List of tanks, lines, dispensers, materials (steel / FRP / mixed).

[ ] Documentation: date of last test (if any), previous reports.

Output: a list of prioritized findings: critical, important, cosmetic.

Week -3: Critical repairs

Goal: fix what you already know is failing.

Typical tasks:

Replacement of visibly damaged spill buckets.

Replacement of flex connectors with a cracked cover or material not compatible with E10.

Replacement of sump seals that look soft or swollen.

Sealing of penetrations in sumps (where cables or pipes enter).

Repair of ATG probes if they are giving erratic readings.

Draining of liquids into authorized disposal containers.

Do not attempt to repair anything that requires opening the tank or disassembling the measurement system. That is evaluated post-test if the problem points there.

Week -2: Internal verification

Goal: simulate the conditions of the official test without paying for it.

Tasks:

ATG probe inspection: verify that the level and water readings are consistent with a manual gauge stick + detection paste.

Temporary closure of dispensing during overnight hours: 4-6 hours of no activity so the system can "breathe." Document readings before and after: if there is unexplained drift, there is an active problem.

Breakaway test on the nozzles: verify that the breakaway connector operates at the correct force (typically 150 lb).

Verification of caps and seals on spill buckets: water-fill tests to confirm they do not leak to ground.

Output: confirmation that there is no active drift and the official test should not bring surprises.

Week -1: Operational coordination

Goal: ensure the conditions on test day are optimal.

Coordination items:

Brand distributor: schedule deliveries so the tanks are at the correct level (not full, not empty) at the time of the test.

Station operations: define the test window (ideally overnight for tanks under test) without affecting dispensing.

Personnel: notify the operating shift about the test, which dispensers will be out of service, and the protocol during the test.

Documentation ready: plans of the underground system, previous reports, current certifications.

Communication with the test technician: confirmation of date, equipment to be used (MESA 2D or other), estimated time.

Test day: execution

Station personnel present as observers.

Visual verification: certified equipment, current calibration of the tester.

Operations: dispensers connected to the tank under test taken out of service.

Documentation of the result in the local logbook.

The 7 checks that avoid most failures

1. Zero liquid in sumps

Why: the presence of water or product in sumps is a symptom of seal failure: the test technician will document it even if the tank passes.

Check: flashlight + inspection of each sump. Drain if there is water. Investigate the cause if there is product.

2. Dry spill buckets with a functional cap

Why: a full spill bucket or one with a broken cap lets water/product into the tank, contaminating the test before it starts.

Check: each spill bucket opened, drained, cap with functional seal verified.

3. Calibrated ATG probes

Why: old or disconnected probes give erroneous readings. If the test relies partly on ATG readings, a bad probe produces a false result.

Check: compare the ATG reading vs. a manual gauge stick in each tank. A difference of >0.5″ in height indicates a problem.

4. No unresolved active alarms

Why: pending active alarms can complicate test interpretation. If there is an unresolved "high water" alarm, the test may be affected.

Check: clear all active alarms with documented prior investigation: see ATG alarm protocol.

5. Correct level in each tank to be tested

Why: the MESA 2D method requires the tank to be between 20% and 95%. Tanks that are nearly empty or full compromise the thermal compensation.

Check: coordinate with the distributor for a delivery 24-48h before the test that leaves the level ideal.

6. Organized site documentation

Why: if the technician arrives and there are no clear plans, sump locations, or line layout, test time is lost.

Check: a physical + digital folder with: tank layout, sump locations, previous certifications, manuals for relevant equipment.

7. A real operating window (not aspirational)

Why: if the promised "window" actually has continuous traffic, the test is contaminated with every customer arrival.

Check: a 4-8 hour window with minimal or zero dispensing. For high-volume stations, that means overnight hours of 1-5 AM or early Sunday morning.

Typical mistakes that show up on test day

"We thought the tank was going to be filled today"

The distributor's delivery arrived at the same time as the test technician. Result: the tank is being filled and cannot be tested. Cost: rescheduling the test (the technician charges extra) and paying for wasted time.

Prevention: written confirmation with the distributor 48h in advance.

"Nobody knows where the plans are"

The technician needs to know which tank feeds which dispenser. Without clear plans, they lose 1-2 hours reconstructing the layout on site.

Prevention: physical plans filed + scanned, accessible to shift personnel.

"The sump is full of water, we can't open it"

Recent rain + a broken spill bucket = a flooded sump. The technician cannot inspect the flex connector's seal. A partial test = a partial result.

Prevention: inspection and draining 24h beforehand, even if it was clean a week earlier (it rains unpredictably in Panama).

"The water alarm goes off and nobody knows why"

The ATG alarm is active; it was silenced months ago but was never investigated. During the test, it complicates interpretation.

Prevention: investigate and document ALL alarms in the preceding weeks: see alarm interpretation.

"The dispenser doesn't isolate from the tank"

Some dispensers have internal valves that must close during the test. If the valve does not work, the test is contaminated with product from the dispenser.

Prevention: functional testing of isolation valves in the prior week.

"We had never done a test here"

A station with > 5 years without a formal test. There is usually an accumulation of small problems that individually do not fail but together do.

Prevention: if you have gone many years without a test, hire a preliminary technical inspection (not the official test). Lower cost, identifies problems before the official test.

The most expensive mistake: doing the test "to see what happens"

There is a temptation to treat the test as a diagnostic ("let's run the test and see if there are problems"). It is a costly mistake for two reasons:

1. The official test produces a regulatory result: if it fails, it is on record and triggers deadlines.
2. It does not pinpoint the cause: the test says "there is a leak > 0.10 gph in this tank" but does not necessarily say where (sump, line, the tank itself).

The correct strategy is: diagnose first (preliminary technical inspection), repair what was found, and then run the official test. The combined cost of diagnosis + repairs + official test is lower than the cost of an official test that fails + remediation plan + retest.

The costs of not preparing

A test that fails triggers a chain: official retest, inspections to locate the leak, repair, lost revenue from the suspended tank and, if the leak is structural, a remediation plan with a report to MIAMBIENTE. The reactive path always ends up more expensive than upfront preparation.

Frequently asked questions

When should I start preparing for a scheduled test?
At least 4 weeks in advance. If the station has gone more than 5 years without a test, 6-8 weeks to allow for potential repairs.

Do I need to hire the same provider for preparation and the test?
Not necessarily, but it helps. The provider who will perform the test knows what inspectors look for and can prepare the site to minimize the chance of failure.

What happens if I discover a major problem during preparation?
Better to discover it this way than during the official test. Pause the test schedule, repair the problem, resume. The authority (Fire Dept./DINASEPI) prefers a documented maintenance plan over a failed test.

Does preparation affect normal operations?
Visual inspections and draining: minimal (<1h per component). Critical repairs: may require temporarily closing 1-2 dispensers. Coordinate with site traffic.

Is it worth doing an "internal test" before the official one?
Yes. An uncertified static test (with a portable pump or similar method) can detect major problems before the tightness test. It does not replace the official test but reduces surprises.

Conclusion

Failed tightness tests are almost always preventable. What makes the difference is not the test method but the condition of the system on the day the technician arrives. That condition is the result of decisions made weeks earlier.

If your station has a scheduled tightness test and you want technical support during preparation, contact Master Services. We perform preliminary technical inspections + assistance with corrections + coordination of the official test with a certified technician.

—

The procedures, frequencies, and regulatory references (SNE, ACODECO, MIAMBIENTE) described in this article correspond to industry practice. The exact preparation for a specific station depends on its current condition, history of previous tests, and site configuration. A personalized quote and plan are prepared after evaluation.