Integrating RECD Monitoring with a Building Management System

In most large buildings, the diesel generator is already on the Building Management System (BMS). Operators can see whether it’s running, its load, fuel level and alarms from the same screen they use for chillers, pumps and lifts. When a Retrofit Emission Control Device is added to the exhaust, it often sits outside that picture: a local gauge or a small controller in the DG room that someone has to walk down to check.

That gap matters. The RECD’s most useful health indicators, such as differential pressure, exhaust temperatures and regeneration status, only help if someone sees them at the right time. On a busy site, “the right time” usually means “on the screen the operator already watches.”

This guide explains how to integrate RECD monitoring with a BMS: which data points are worth bringing across, the main integration methods, how to lay out graphics and alarms, how to commission the link, and what documents to ask for at handover.

First, Check What Your RECD Can Actually Provide

RECDs vary widely in their instrumentation. Before planning integration, find out exactly what your device offers. Ask the supplier for:

• A list of sensors fitted, such as differential pressure, inlet and outlet temperature, and back-pressure
• Whether there is a controller, and what it does (display, logging, regeneration control, heater control, alarms)
• Available outputs: volt-free (dry) contacts, 4–20 mA analogue signals, Modbus RTU (RS-485), Modbus TCP (Ethernet), or other protocols
• The register map or point list, if a communication protocol is offered (see RECD Modbus data: units, scaling and register mapping explained)
• The power supply requirements of the controller and sensors (see RECD controller and heater power supply requirements)

A simple passive RECD may have only a differential pressure gauge. A device with active regeneration usually has a controller with more data. Integration should be designed around what actually exists, or the specification should be upgraded at the ordering stage. See writing a technical RFQ for an RECD project.

Which Data Points Are Worth Integrating?

Not every value needs to reach the BMS. Start with the points that support a decision.

Point	Type	Why it’s useful in the BMS
Differential pressure across RECD (ΔP)	Analogue	Soot and ash loading; crack or bypass detection
Engine exhaust back-pressure	Analogue	Engine protection; compare with OEM limit
RECD inlet temperature	Analogue	Shows exhaust conditions reaching the device
RECD outlet temperature	Analogue	Thermal state; regeneration detection; overheating
Regeneration active	Digital	Explains ΔP and temperature changes
Regeneration count / last regeneration time	Analogue / counter	Duty-cycle and ageing trends
Heater status (if fitted)	Digital	Confirms active regeneration support
RECD common alarm	Digital	Simple, robust alarm for any fault
Individual alarms (high ΔP, high back-pressure, sensor fault, etc.)	Digital	Targeted operator response
Controller healthy / communication OK	Digital	Detects silent loss of monitoring

Points already on the BMS from the DG controller, such as running status, kW load, running hours and engine alarms, should be shown alongside the RECD data. RECD readings only make sense in the context of load and running state. See why RECD test results may differ from day-to-day site readings.

Four Ways to Get RECD Data Into a BMS

Option 1: Hardwired dry contacts

The RECD controller provides volt-free relay contacts for one or more alarms, such as “common fault” or “high ΔP”. These are wired to digital inputs on a nearby BMS controller.

• Pros: simple, robust, works with any BMS, easy to test, no protocol knowledge needed
• Cons: alarms only, with no values or trends; each extra signal needs another pair of wires and another input
• Good for: small sites, or as a backup alongside a data link

Option 2: Hardwired analogue signals (4–20 mA or 0–10 V)

Sensors or the controller output analogue signals for key values, such as ΔP or back-pressure, which BMS analogue inputs read and scale.

• Pros: values and trends without a communication protocol; independent of the RECD controller’s software
• Cons: each value needs its own cable and input; scaling must be set correctly on both ends; long cable runs near hot exhaust need care (see RECD cable routing near hot exhaust lines)
• Good for: a few critical values where the RECD has no communication port

Option 3: Modbus (RTU or TCP)

The RECD controller exposes its data in Modbus registers. The BMS, or a gateway, polls these registers over RS-485 (Modbus RTU) or Ethernet (Modbus TCP).

• Pros: many points over one connection; widely supported by BMS platforms and gateways
• Cons: needs an accurate register map, correct scaling and data-type settings; communication faults need detecting
• Good for: most medium and large sites

The Modbus protocol specifications are published by the Modbus Organization.

Option 4: BACnet via a gateway

Many commercial BMS platforms use BACnet natively. If the RECD controller speaks only Modbus, a protocol gateway can translate Modbus registers into BACnet objects. Information about the BACnet standard is available from the BACnet committee website (ASHRAE SSPC 135).

• Pros: RECD points appear as native objects in the BMS
• Cons: an extra device to configure, power and maintain; mapping errors are possible at two levels
• Good for: sites with a BACnet-based BMS and a Modbus-only RECD

Which option to choose

Site situation	Typical choice
One small DG set, basic BMS	Dry-contact common alarm, plus optional 4–20 mA ΔP
Several DG sets, Modbus-capable BMS	Modbus RTU or TCP from each RECD controller, plus hardwired common alarm as backup
BACnet-based BMS, Modbus RECD	Modbus-to-BACnet gateway, plus hardwired common alarm
Critical site (hospital, data centre)	Data link and hardwired critical alarms, on backed-up power

A combination is often best: a data link for values and trends, and a hardwired common alarm that still works if the data link fails.

Worked Example: Points Schedule for a Three-Set Site

The example below is illustrative, for a commercial building with three DG sets, each with an RECD controller offering Modbus RTU. A points schedule like this belongs in the integration specification.

Point name	Source	Type	Units	Range	Alarm?	Trend interval
DG1_RECD_DP	RECD1 Modbus	AI	mbar	0–200	High warning / high critical / low	1 min
DG1_RECD_BP	RECD1 Modbus	AI	mbar	0–200	High critical	1 min
DG1_RECD_T_IN	RECD1 Modbus	AI	°C	0–700	High warning	1 min
DG1_RECD_T_OUT	RECD1 Modbus	AI	°C	0–700	High critical	1 min
DG1_RECD_REGEN	RECD1 Modbus	DI	On/Off	—	No (event log)	On change
DG1_RECD_FAULT	RECD1 Modbus	DI	Normal/Fault	—	Yes	On change
DG1_RECD_COMMS	BMS internal	DI	OK/Fail	—	Yes	On change
DG1_RECD_COMMON_ALM_HW	Hardwired contact	DI	Normal/Alarm	—	Yes	On change
DG1_KW	DG controller	AI	kW	0–500	(DG alarms)	1 min
DG1_RUN	DG controller	DI	Run/Stop	—	No	On change

Repeat the same structure for DG2 and DG3. Consistent naming makes graphics, alarms and reports much easier to manage.

Alarm values are set in the alarm response matrix, not chosen ad hoc. See how to design an RECD alarm response matrix for operators.

Designing the BMS Graphics

A good RECD graphic helps an operator answer three questions in seconds: Is the DG running? Is the RECD healthy? Is any action needed?

Practical design tips:

• Place RECD data on the DG set’s page, not on a separate screen. Show the exhaust path visually: engine → RECD → stack.
• Show load next to ΔP and back-pressure. These values only mean something at a known load.
• Use clear status colours consistently with the rest of the BMS: normal, warning and alarm.
• Indicate regeneration clearly, for example with a “Regenerating” label, so operators don’t mistake normal ΔP and temperature changes for faults.
• Show communication status. A frozen value that looks normal is more dangerous than an obvious “COMMS FAIL”.
• Link to trends with one click, especially ΔP, back-pressure and outlet temperature.
• Show reference values from the acceptance test, such as “Reference ΔP at 50% load: 44 mbar”, so operators can judge readings in context.

Trends That Pay for Themselves

Trend logs turn individual readings into maintenance intelligence:

• ΔP at a standard load over weeks and months shows soot and ash build-up. A sudden unexplained fall can reveal a crack. See can RECD efficiency change between new and aged conditions.
• Regeneration frequency shows whether duty-cycle changes are affecting the device.
• Back-pressure against load confirms the engine stays within its limit. See how exhaust back-pressure affects a DG set.
• Outlet temperature during regeneration helps detect abnormal events.

Set trend intervals sensibly. One-minute samples during running are usually enough for RECD data. Make sure the BMS keeps the history long enough to compare across seasons, ideally at least a year.

Alarm Routing

Decide in advance who receives each alarm, and how:

Alarm priority	Example	Route
Critical	High back-pressure above engine limit; high outlet temperature	BMS alarm + SMS/call to shift engineer; audible in control room
Warning	High ΔP approaching cleaning threshold	BMS alarm + maintenance work order
Advisory	Regeneration overdue; trend deviation	BMS event log; weekly report
System	Communication failure; controller fault	BMS alarm to BMS/IT team and DG maintenance

Avoid sending every RECD event as a high-priority alarm. Too many alarms train operators to ignore them. The design of priorities and responses is covered in detail in how to design an RECD alarm response matrix for operators.

Power and Network Considerations

Power

The RECD controller, sensors, any gateway and the BMS controller reading them must be powered whenever the DG is running. Since the DG usually runs because grid power has failed, connecting monitoring equipment only to a grid-fed circuit is a classic mistake. Monitoring goes blind exactly when it’s needed. See RECD monitoring during a power failure: what should stay available.

Network and security

• Modbus RTU runs on a local RS-485 bus. It’s simple and physically contained, but needs correct termination and cabling.
• Modbus TCP and BACnet/IP run on Ethernet. Keep them on the building’s operational technology (OT) network, separated from the corporate IT network and the internet, following good practice such as the ISA/IEC 62443 series.
• Remote access for the RECD supplier, if needed, should go through a controlled, secure route, not an open port. See does an RECD need internet access to operate.

Turning BMS Data Into Useful Reports

Live screens help operators. Reports help managers. Once RECD data is in the BMS, a simple monthly report can serve maintenance, EHS and management at the same time. A practical report includes:

• DG running hours in the month, by set, with the number of starts
• Average and peak load during running
• ΔP at the standard check condition, compared with the acceptance reference value and the previous month
• Peak back-pressure, compared with the engine maker’s limit
• Number of regenerations, and any failed regenerations
• Alarms raised, by priority, with actions taken and time to close
• Communication availability, meaning the percentage of DG running time during which RECD data reached the BMS
• Recommended actions, such as ash cleaning due, sensor checks or a retest

The communication availability figure is easily overlooked but revealing. If RECD data was missing for a third of DG running hours, nobody was really watching the device during those periods.

These reports also support compliance and sustainability reporting. Running hours and load data feed directly into estimates of particulate avoided. See estimating kilograms of PM avoided after an RECD retrofit. Over time they show whether the device is ageing normally or needs attention, which helps decide when an RECD should be retested.

Who Does What: A Responsibility Matrix

Integration projects often stall because each party assumes another is responsible. Agree this up front:

Task	RECD supplier	Electrical contractor	BMS contractor	Owner/FM
Provide register map and point list	✔
Provide outputs (contacts, analogue, comms port)	✔
Cable from RECD controller to BMS panel/gateway		✔
Gateway supply and configuration			✔
BMS points, scaling, graphics, trends			✔
Alarm limits and response matrix	Advise		Configure	Approve
Backed-up power for monitoring	Specify load	✔		Approve
Point-to-point testing	Attend	Attend	✔	Witness
Operator training	✔ (RECD)		✔ (BMS)	Arrange

See customer and installer responsibilities in an RECD project and operator training in the RECD purchase order.

Commissioning the Integration

Don’t treat integration as finished when values appear on the screen. Test it properly.

Point-to-point testing

For every point on the schedule:

1. Compare the BMS value with the local value on the RECD controller or gauge, at two or more operating conditions (e.g. DG off and DG running at load). They should match within the stated resolution.
2. Check units and scaling. A ΔP of 4.5 shown as 45 means a scaling error. See RECD Modbus data: units, scaling and register mapping explained.
3. Force or simulate each alarm where the controller allows, and confirm it appears on the BMS with the correct text and priority, and reaches the correct people.
4. Test the hardwired common alarm independently of the data link.
5. Break the communication link deliberately, and confirm the BMS raises a communication failure alarm and doesn’t keep showing the last value as if it were live.
6. Test during a simulated grid failure, with the DG running, to confirm monitoring stays powered and connected.

Record each result on a signed test sheet.

Integration into acceptance

Add the integration tests to the RECD project’s handover requirements. See closing the RECD punch list before handover and factory acceptance vs site acceptance tests for RECD projects.

Handover Documents to Request

• Final points schedule with point names, sources, units, ranges and alarm limits
• RECD register map (if Modbus) with data types and scaling
• Gateway configuration backup and mapping file (if used)
• Wiring diagrams for hardwired signals and communication cables
• Signed point-to-point test sheets
• Alarm response matrix, approved by the owner
• Screenshots of BMS graphics
• Contact details for RECD supplier and BMS contractor support

Keep these with your environmental compliance checklist records. They also show inspectors that the device is actively monitored.

Common Mistakes to Avoid

• Integrating values without load context, so operators can’t tell whether a ΔP of 60 mbar is normal or not
• No communication-failure alarm, so frozen values look normal
• Monitoring powered only from grid supply, so it’s blind during DG operation
• Too many high-priority alarms, which operators soon learn to ignore
• Scaling errors discovered months later, when trends don’t make sense
• No owner for alarm limits, which stay at default values forever
• Integration not included in the RECD order, leading to extra cost and delay later. See how to manage scope changes after an RECD order

Related Reading

• How to design an RECD alarm response matrix for operators
• RECD Modbus data: units, scaling and register mapping explained
• Does an RECD need internet access to operate?
• RECD monitoring during a power failure: what should stay available?
• RECD maintenance guide

Frequently Asked Questions

Can every RECD be connected to a BMS?

Any RECD with a sensor or controller output can provide at least some information, even if only a dry-contact alarm or a 4–20 mA ΔP signal. A purely passive device with only a local mechanical gauge would need a transmitter added. Check with your supplier what outputs are available, or specify them when ordering.

Is Modbus or BACnet better for RECD integration?

Neither is better in itself. Use what your BMS supports natively. Many RECD controllers offer Modbus, and a gateway can translate it to BACnet if the BMS needs it.

Do I still need a hardwired alarm if I have a Modbus link?

It’s strongly recommended for important sites. A hardwired common alarm still works if the communication link, gateway or BMS software fails.

How often should the BMS poll the RECD controller?

Polling every few seconds is common and more than enough for RECD values, which change slowly. Trend logging at one-minute intervals during running is usually sufficient.

Can RECD data be added to an existing BMS without replacing it?

Usually, yes. Most BMS platforms can read Modbus directly or through an add-on interface, and a gateway can translate protocols where needed. The work is mainly configuration, cabling and testing, not replacing the BMS.

Who should set the alarm limits?

The RECD supplier should recommend limits based on the device design and engine data, and the owner should approve them. Reference values from the acceptance test help set realistic warning levels.

Bring the RECD Into the Operator’s View

An RECD that’s monitored from the BMS is an RECD that gets attention before problems grow. Aceget supplies CPCB type-approved Retrofit Emission Control Devices for DG sets from 25 kVA to 2500 kVA. We can advise on monitoring outputs and integration requirements at the ordering stage. See our certifications, learn how an RECD works, or contact our team with your BMS details.