Why test a continuity plan at all?
A continuity plan written on paper is a theory. Plants discover its weaknesses during real failures: the standby compressor that trips on high temperature after ten minutes, the tie-in valve nobody can find at night, the dryer bypass that has seized, the operator who has never switched the sequencer to local control. A controlled test finds these problems on a day you choose, with people and spares ready, instead of in the middle of a production run.
Testing should be proportional to consequence. A plant whose air feeds sterile pharma production or furnace safety controls needs a more rigorous programme than a fabrication workshop. Either way, the aim is the same: evidence that the plant can survive the failures it says it can survive.
Define the test before you touch a valve
Every drill should be written down and approved before it starts. The plan should state:
- The scenario: which credible failure is being simulated, for example loss of the largest compressor, a dryer trip, a feeder trip or a sequencer failure.
- The boundary: which equipment, headers and users are included, and which are excluded.
- The critical duty: the flow, minimum pressure at the point of use (in bar and psig) and air quality that must be maintained.
- Acceptance criteria: what counts as a pass, such as header pressure staying above a stated value and the standby machine reaching full load within a stated time.
- Abort criteria: the conditions at which the test stops immediately, for example pressure at a critical user falling below its limit or an alarm on any machine.
- Restoration criteria: how the system returns to normal and how you confirm it has.
- Roles: who leads, who operates, who watches production, who can call a stop.
Live, partial and simulated tests
| Test type | What happens | Good for | Limits |
|---|---|---|---|
| Tabletop review | Team walks through the scenario using drawings and procedures | Finding gaps in roles, contacts and documents | Proves nothing about equipment |
| Component test | One item is exercised, such as a standby start or a drain test | Regular, low-risk checks | Does not test interactions |
| Partial live test | A real failure is simulated during low demand with abort limits | Proving standby load acceptance and controls fallback | Needs careful planning and production agreement |
| Full drill | A complete scenario including rental air connection or generator start | Proving the whole plan and the people | Highest disruption; schedule during shutdowns |
Most plants build up: tabletop first, then component tests, then partial live tests in low-demand periods such as weekends or planned shutdowns.
What should a resilience test programme cover?
| Area | What is proved | Evidence to record |
|---|---|---|
| Standby compressor | Starts automatically, loads fully, holds pressure at summer conditions | Time to load, header pressure trend, discharge temperature |
| Critical-demand coverage | Remaining machines carry critical users after the largest machine is lost | Pressure at critical points of use, non-critical loads shed |
| Treatment redundancy | Standby dryer or bypass maintains acceptable air quality | Outlet dew point, filter differential pressure |
| Controls fallback | Machines run on local pressure settings if the sequencer or network fails | Pressure band achieved, any machine conflicts |
| Sensor failure | A failed pressure transducer does not cause wrong loading | Alarm raised, safe default behaviour |
| Cooling and ventilation | Machines stay within temperature with one fan or pump out | Room and discharge temperatures over time |
| Power | Critical machines restart in order after a supply interruption or generator changeover | Restart sequence, trips, time to pressure |
| Rental air connection | A portable compressor can be placed, connected and loaded | Access, valve operation, hose and coupling fit, time taken |
| People and communication | Operators know the steps; contacts are current, including night shift | Who was called, response, gaps found |
A standby compressor drill, step by step
- Agree the test window with production and choose a low-demand period.
- Record the baseline: running machines, header pressure, dew point, room temperature and demand.
- Confirm the standby machine is healthy: oil level, no active alarms, intake filter clean, isolation valves open.
- Brief everyone on abort limits and who can stop the test.
- Simulate the failure by a controlled stop of the lead compressor, following the manufacturer’s procedure.
- Watch the standby start, load and header pressure recovery. Note any alarms.
- Hold the condition long enough to see temperatures stabilise, especially on hot days.
- Restore normal running in the agreed order and confirm pressure, dew point and machine status.
- Write up the result, including anything that did not go to plan.
Testing rental and emergency air connections
If your plan relies on rental air, test it at least once. Questions to answer in the field:
- Can a diesel portable or trailer-mounted electric compressor reach the tie-in point, and is the ground firm enough?
- Does the hose coupling or flange match, and is there an isolation valve and a non-return valve?
- Is an aftercooler, moisture separator or dryer needed so that rental air meets your quality class?
- For electric rentals, is there a suitable power point, or will a generator be needed?
- Who on site authorises the connection and who operates the rental machine?
We keep our own fleet of rental compressors from 50 to 2,000 cfm at 2 to 30 bar (about 30 to 435 psig), diesel and electric, and can help plan the tie-in and a trial connection. Our team has supplied and commissioned temporary air packages on large projects, including multi-compressor installations at high pressure for a natural-gas development in Andhra Pradesh, so we know the practical details that decide whether a connection works.
Failed tests are valuable
A drill that fails has done its job: it found a weakness safely. Record what happened honestly, investigate the cause and fix it. Typical findings include:
- standby machine set to manual start, or with an outdated pressure setting;
- dryer bypass valve seized or unlabelled;
- local controller pressure bands that fight each other once the sequencer is off;
- a room that overheats within minutes when one exhaust fan stops;
- a rental coupling that does not fit the plant flange;
- contact lists that name people who have left.
After the fix, repeat the relevant part of the test. Do not record a pass until the system has actually passed.
How often should you test?
There is no single right interval; set it by consequence and by how often the system changes. A practical pattern is frequent component checks (standby starts, drain tests), periodic partial tests of critical scenarios, and a fuller drill during an annual shutdown or after major changes such as a new compressor, a new sequencer, a new dryer or a header modification. Always retest after a real incident has exposed a gap.
Safety boundaries during drills
Drills must stay within the manufacturer’s operating limits and the site’s own procedures. Isolation, line breaking and electrical work remain under the site’s permit-to-work and lockout procedures. Never defeat safety shutdowns, such as high discharge temperature protection, to make a test pass. If a critical user’s pressure or quality approaches its limit, stop the test and restore normal running.
How Reddy Compressors helps
We help plants turn continuity plans into tested, working arrangements. As an independent multi-brand specialist we work on compressors from ELGi, Atlas Copco, Ingersoll Rand, KAESER, Chicago Pneumatic and other makers, and we are not an authorised dealer of any brand. We have tie-ups with all major and reputed brands. From our office at Tarnaka, Secunderabad, our team of over 20 service engineers and technicians covers all of Telangana and Andhra Pradesh, and we supply equipment, rentals and spares across India.
- Pre-test inspection of standby compressors, dryers and controls.
- Engineers on site during drills to observe, record and advise.
- Trial connection of rental compressors and planning of tie-in points.
- Corrective repairs, settings checks and spares after a failed test.
- Preventive maintenance and AMC to keep standby equipment ready.
What to send us
- Your continuity plan or a short description of the scenarios you want to test.
- List of compressors, dryers and controllers with make, model and nameplate photos.
- Critical users with required pressure, flow and air quality.
- Photos of the proposed rental tie-in point and access route.
- Preferred test windows, such as weekends or shutdown dates.
Frequently asked questions
How do I test my standby air compressor?
Choose a low-demand period, record a baseline, then stop the lead compressor in a controlled way and watch the standby machine start, load and hold header pressure. Set abort limits for pressure at critical users beforehand. Keep the condition long enough for temperatures to stabilise, then restore normal running and record exactly what happened, including any alarms or delays.
What is a compressed air emergency drill?
It is a planned exercise in which a credible failure, such as loss of the largest compressor, a dryer trip or a power interruption, is simulated under control. The team follows the continuity procedure, records how equipment and people respond, and checks that critical users keep enough air. The drill has written acceptance, abort and restoration criteria agreed in advance.
How often should compressed air backup systems be tested?
Set the interval by consequence and by how often the system changes. Many plants do frequent standby start checks, periodic partial tests of critical scenarios, and a fuller drill during an annual shutdown. Always retest after installing a new compressor, dryer or sequencer, after modifying headers, and after any real incident that exposed a weakness.
Can we test without stopping production?
Often, yes, if the test is planned for a low-demand period with clear abort limits. Component checks such as standby starts and drain tests rarely affect production. Partial live tests need agreement with production and a quick way to restore normal running. Tests that carry real risk to product or safety should be scheduled during planned shutdowns.
What should we do if a resilience test fails?
Treat the failure as useful evidence. Stop the test, restore normal running and record exactly what happened. Investigate the cause, which is often a wrong setting, a seized valve, an overheating room or an untrained operator, then fix it and repeat the relevant part of the test. Record a pass only once the system has genuinely passed.
How do we prepare a plant to connect rental compressors in an emergency?
Install a tie-in point near the compressor house with an isolation valve and a non-return valve, and a coupling or flange that suits portable compressor hoses. Make sure a portable unit can reach it on firm ground. Know your required flow, pressure and air quality, and arrange a power point or generator for electric rentals. Then do a trial connection.
Related pages
Technical content reviewed by D. Srinivas Reddy, Managing Partner and service engineer, Reddy Compressors. Last updated 17 October 2026. How our engineering content is reviewed.