Skip to content
White supply air duct running under a ceiling
  1. Insights
  2. BMS vs EMS in a GMP facility: which HVAC instruments need GMP calibration?

HVAC & cleanrooms

BMS vs EMS in a GMP facility: which HVAC instruments need GMP calibration?

Not every sensor on an air handler is a GMP instrument. A documented, risk-based classification tells you which ones are, how to calibrate them, and how to defend the decision in an inspection.

  • By Nexus Metrology
  • 5 min read

Two systems, two jobs

Most GMP facilities have two kinds of system watching the HVAC. The building management system (BMS) controls it: fans, dampers, valves, heating and cooling coils, all driven by set points and control loops. The environmental monitoring system (EMS), where a site has one, watches the results: room temperature, relative humidity, differential pressure and sometimes particle counts in the areas that matter to the product, with alarms and records that quality relies on.

Many sites keep the two apart on purpose. The EMS can then be validated and placed under quality change control, while engineering keeps the freedom to tune the BMS. Other sites use a single system for both jobs, or a BMS with a validated monitoring layer. Each approach can work. What matters is knowing which instruments produce data your quality decisions depend on, because those are the instruments that need GMP calibration.

Not every HVAC instrument is a GMP instrument

A large air-handling system can carry hundreds of sensors. Calibrating every one of them to full GMP standard spends effort where it adds little, and spreads attention thin where it is needed. Calibrating too few leaves gaps that an inspector will find.

The regulations set the outcome, not the list. 21 CFR 211.46 calls for equipment to control air pressure, micro-organisms, dust, humidity and temperature where appropriate, and 211.68 expects equipment used in manufacturing to be routinely calibrated, inspected or checked under a written program. EU GMP Annex 1 asks for pressure differences identified as critical to be continuously monitored and recorded, and for temperature and relative humidity to be controlled within ranges that suit the product, the process and the people. ICH Q9(R1), published by FDA as guidance, adds the principle that makes the rest workable: the level of effort, formality and documentation should be commensurate with the level of risk.

A risk-based classification

A simple and widely used approach sorts each instrument into one of three classes. The names vary from site to site; the logic is what matters.

ClassWhat it meansTypical HVAC examples
GMP criticalThe reading demonstrates a product or process requirement, is used for a quality decision, or raises a GMP alarm.EMS room differential pressure, temperature and humidity sensors in classified or controlled areas; pressure monitors that alarm a critical cascade
GMP non-criticalThe instrument controls or supports a GMP system, but another, calibrated instrument provides the GMP record.BMS sensors that drive dampers or coils where an independent EMS sensor monitors the same room
No GMP impactThe reading has no bearing on product quality or GMP decisions.Outside air temperature, plant-room conditions, energy metering

The question is never which system an instrument is wired to. A BMS sensor whose value is printed on a batch-related report, or used to release a room, is acting as a GMP-critical instrument whatever its label says.

The impact assessment behind each decision

Document the reasoning for each instrument, or for each group of identical instruments in the same service. A short set of questions covers most cases:

  1. What is the reading used for? Control only, monitoring, alarming, or a record that quality reviews?
  2. Is it the only evidence? If the room has an independent EMS sensor, the BMS sensor may not be the GMP record. If it is the only source, it probably is.
  3. What happens if it is wrong? Could a drifted reading hide a room that is out of its limits, or trigger a decision about product?
  4. Would anyone notice? A failure that shows up immediately in another measurement carries less risk than a quiet drift nobody can see.

The answers place the instrument in a class and set what its calibration looks like. Keep them with the instrument list so an inspector, or a new engineer, can follow the logic. Annex 11 expects an up-to-date inventory of computerised systems and their GMP functionality, and risk management applied throughout a system’s lifecycle; a documented classification of the instruments feeding the EMS fits naturally beside that.

How calibration differs by class

GMP-critical instruments belong in the written calibration program. They need defined tolerances derived from the process limits they protect, NIST-traceable reference standards, as-found and as-left data, a loop check through to the value the EMS records, alarm verification, QA-approved records and an impact assessment whenever one is found out of tolerance. Their intervals should be justified, ideally from their own history (see setting risk-based calibration intervals).

GMP non-critical instruments still need to work, because a control sensor that drifts will push the room toward its limits and generate alarms on the critical side. A documented engineering calibration or check at a defined interval is usually proportionate. Annex 15 includes calibration of instrumentation in installation qualification, so these instruments should be calibrated before qualification work that relies on them.

Instruments with no GMP impact follow good engineering practice and the manufacturer’s recommendations.

Watch points

  • Data that changes class without anyone deciding. A BMS trend added to a monthly quality report quietly turns a non-critical sensor into a GMP record. Route that change through change control and reclassify the instrument.
  • Two values for one room. When the BMS and EMS both show a room’s pressure and disagree, people will ask which one is right. Decide in advance which value is the GMP record, and say so in the procedure.
  • Scaling and range mismatches. The transmitter, the BMS and the EMS must agree on range and units. A loop check is the only reliable way to prove it (see calibrating the sensors behind your cleanroom pressure cascade).
  • Classification that never gets reviewed. Revisit it after a deviation, a room reclassification, an HVAC modification or a new monitoring system.

How Nexus Metrology helps

Nexus Metrology calibrates BMS and EMS instrumentation on site and helps teams build the instrument classification and impact assessments behind their calibration program, drawing on 15+ years of GMP experience. See HVAC Utility Calibration and Metrology Program Development.

Sources

Regulations and guidance are paraphrased above; the originals are the reference.

  1. 21 CFR 211.46: Ventilation, air filtration, air heating and cooling (eCFR)
  2. 21 CFR 211.68: Automatic, mechanical, and electronic equipment (eCFR)
  3. EudraLex Volume 4, Annex 1: Manufacture of Sterile Medicinal Products (2022), paragraphs 4.16, 5.2 and 9.6 (European Commission)
  4. EudraLex Volume 4, Annex 11: Computerised Systems (European Commission)
  5. EudraLex Volume 4, Annex 15: Qualification and Validation (European Commission)
  6. FDA Guidance for Industry: Q9(R1) Quality Risk Management (2023)

Next step

Talk to a metrology expert

Questions about your own instruments, intervals or records? Nexus Metrology brings 15+ years of GMP experience and on-site field calibration to your facility.

Keep reading

All Insights articles