What Should a Useful RECD Operating Dashboard Show?

Many RECD dashboards fail in one of two ways. Some show almost nothing: a single differential pressure number with no context, which nobody knows how to interpret. Others show everything: twenty gauges, flashing icons, raw register values and colourful charts, so that the one reading that matters is lost in the noise.

A useful dashboard sits between the two. It shows the right information, for the right person, in a way they can understand in seconds. This article sets out what a Retrofit Emission Control Device dashboard should show, organised by who is looking at it, with the key indicators defined, design rules to follow and a list of things to leave out.

Start With the Questions the Dashboard Must Answer

Before choosing a single chart, agree what the dashboard is for. For most sites, it must answer three questions quickly:

1. Is the DG running, and is the RECD working normally right now?
2. Is anything trending towards a problem?
3. Is the monitoring itself trustworthy? In other words, is the data current and complete?

Every element on the dashboard should help answer one of these. If it doesn’t, it probably belongs somewhere else, or nowhere.

Three Audiences, Three Layers

Different people need different views. Rather than one crowded screen, design three layers that link to each other.

Layer	Who	Main question	Time horizon
Operating view	Shift operators, BMS staff, on-call technicians	“Is it OK right now? Do I need to act?”	Now / last hour
Maintenance view	DG maintainers, RECD service engineers	“Is it ageing normally? What’s due?”	Weeks to months
Management view	Facility managers, EHS leads, auditors	“Is it performing and compliant? Are we on top of it?”	Month / quarter / year

Layer 1: The Operating View

This is the screen an operator glances at during a DG run. It should be readable in under ten seconds.

What it should show

DG status strip

• Running / stopped, with time since start
• Load in kW (and as % of rating)
• Supply mode, such as “on DG, grid failed”

RECD health block

• Differential pressure (ΔP) across the RECD, with the normal band for the current load shown alongside, not just a raw number
• Engine back-pressure, with the engine maker’s limit marked
• RECD inlet and outlet temperatures
• Regeneration status: idle, requested, active, failed
• Heater status, if fitted

Active alarms

• Only current, unacknowledged and acknowledged-but-active alarms, sorted by priority
• Each linked to its response guidance. See how to design an RECD alarm response matrix for operators

Data freshness

• “Last update: 8 seconds ago”, or a clear STALE / COMMS FAIL indicator if data has stopped

Why “normal band for current load” matters

A ΔP of 60 mbar may be perfectly normal at 90% load and a warning sign at 30% load. Showing the raw number alone forces the operator to remember reference values. A better approach is to show the reading against an expected range for the current load, for example as a small bar with a shaded “normal” region. Expected ranges come from acceptance-test reference values and routine data. See why RECD test results may differ from day-to-day site readings.

Short trends on the operating view

A small one-hour trend for ΔP, back-pressure and outlet temperature helps the operator answer: is it rising, falling or steady? A sudden step is very different from a slow drift.

Layer 2: The Maintenance View

This view supports planning, diagnosis and ageing analysis. See can RECD efficiency change between new and aged conditions.

What it should show

Long-term ΔP trend at a standard condition. Raw ΔP varies with load, so the most useful long-term chart plots ΔP only at a standard check condition (for example 50% load, fully warm, no regeneration), or ΔP normalised for exhaust flow. That chart shows ash and soot build-up clearly.

• A slow rise indicates ash build-up, so plan cleaning
• An unexplained step down suggests a crack or bypass, so inspect

Regeneration history

• Number of regenerations per week or month
• Success / failure count
• Average duration and peak temperature, for active systems

Back-pressure versus load scatter plot. Each point is a minute of operation. Over time, the cloud shifts upwards as the device loads with ash. Points near the engine limit stand out immediately.

Hours and counters

• DG running hours, and hours since the last RECD cleaning or service
• Hours since the last full emission test
• Next planned maintenance, linked to the RECD maintenance guide

Alarm history

• Alarms by type over the last 90 days, highlighting repeats

Sensor health

• Last calibration or check date for each sensor
• Any sensor faults, frozen values or out-of-range readings

Layer 3: The Management View

Managers don’t need live values. They need a few well-defined indicators, ideally monthly, with trends.

Suggested key performance indicators

KPI	Definition	Why it matters
DG running hours	Total hours in the period	Context for all other figures
Monitoring data availability	% of DG running minutes with valid RECD data	Shows whether the device was actually being watched
ΔP index	ΔP at standard condition ÷ acceptance reference ΔP	Tracks device loading and ageing on one scale
Peak back-pressure margin	Engine limit − highest recorded back-pressure	Engine protection
Regeneration success rate	Successful ÷ attempted regenerations	Health of regeneration system
Critical alarms	Count, with mean time to acknowledge and to clear	Responsiveness
Maintenance compliance	Planned RECD maintenance tasks done on time, %	Discipline
Days since last full emission test	Calendar days	Compliance currency
Estimated PM avoided	kg in the period (see method)	Environmental benefit

The PM-avoided figure should follow a documented method. See estimating kilograms of PM avoided after an RECD retrofit.

Monitoring data availability deserves special attention. If the dashboard shows a clean, healthy picture but data was missing for a third of the running hours, the picture isn’t trustworthy. See detecting missing data in RECD monitoring records.

Design Rules That Make Dashboards Work

Many principles from established guidance on human–machine interfaces, such as the ISA-101 series, apply directly to RECD dashboards.

1. Make normal look calm

Use neutral colours, such as greys and muted tones, for normal states. Reserve strong colours for abnormal conditions. If everything is bright green, red alarms don’t stand out.

2. Use colour consistently, and not alone

Pick one colour for each priority (for example, amber for warning, red for critical) and use it everywhere. Pair colour with text or shape, so the dashboard still works for colour-blind users and on poor screens.

3. Show values with context

Every analogue value should show its units, its normal range and, where relevant, its limit. A number with no context forces the reader to remember thresholds.

4. Prefer trends to gauges

Dial gauges look impressive but show only the current value. A small trend line shows direction and rate of change, which usually matters more.

5. Put the exhaust path in order

Lay out RECD readings in the order the exhaust flows: engine → back-pressure → RECD inlet → ΔP → RECD outlet → stack. Operators understand the system faster when the screen mirrors the physical layout.

6. Show time and freshness everywhere

Every page should show the current time, with time zone, and how recent the data is. A frozen screen that looks normal is dangerous. See how to align RECD sensor timestamps with generator runtime logs.

7. One click to detail

From any summary tile, the user should reach the underlying trend, event log or alarm guidance in one click.

8. Design for the smallest screen it will be viewed on

On-call technicians will often open the dashboard on a phone at night. The operating view should be legible on a phone, with the most important information at the top.

Setting the “Normal” Bands Shown on the Dashboard

Showing each value against a normal range is only useful if the ranges are realistic. A practical way to set them:

1. Start with the acceptance-test fingerprint. During the acceptance test, note ΔP, back-pressure and temperatures at each tested load. See how to set up a witnessed RECD performance demonstration.
2. Collect four to eight weeks of normal operation. Use only data from runs where the DG was fully warm and no regeneration was active.
3. Group the data by load band, for example 0–25%, 25–50%, 50–75% and 75–100% of rated kW.
4. Calculate a typical range for each band. For example, use the middle 90% of observed values, leaving out the top and bottom 5%.
5. Set the dashboard band for each load range from these values, and show the band matching the current load.
6. Review after cleaning, substrate replacement or engine work, because normal values shift. See when should an RECD be retested after changes to the DG set.

Normal bands aren’t alarm limits. Alarm limits come from device and engine limits and the alarm matrix. Normal bands help operators see at a glance whether a value is typical for the current conditions, before it gets anywhere near an alarm.

Dashboard Mistakes Seen on Real Sites

• “All green, all the time.” Every tile shows green, so a slowly rising ΔP never stands out. Normal should be neutral, and only abnormal should draw attention.
• Gauges with default scales. A ΔP gauge running 0–1,000 mbar for a device that normally reads 40 mbar makes every change invisible.
• No load shown. Operators can’t judge whether readings are normal.
• Last values shown after a comms failure. The screen looks healthy while nothing is being received. See detecting missing data in RECD monitoring records.
• Different units on different screens. mbar on one page and kPa on another, or a temperature shown in one place with an extra decimal.
• Engineering pages used as operating pages. Raw registers, diagnostic counters and configuration values on the main screen confuse operators.
• No owner. Nobody is responsible for keeping limits, bands and links up to date, so the dashboard slowly drifts out of step with the device.

Each of these is easy to fix once spotted. A short review with operators every few months usually finds them.

What to Leave Off the Dashboard

Removing clutter is as important as adding information. Leave out:

• Raw register values or internal codes, such as “REG 40011 = 452”. Show engineering units only. See RECD Modbus data: units, scaling and register mapping explained
• Decorative 3D graphics of the RECD that take space but carry no information
• Every available data point “just in case.” If nobody acts on it, move it to an engineering page
• Readings while the DG is stopped presented as live, such as ΔP or temperatures with no flow. Grey them out or label them “DG stopped”
• Unqualified efficiency percentages. A live “RECD efficiency: 92%” calculated from opacity or a single sensor isn’t a measured particulate reduction. Don’t show one unless it has a sound, documented basis. See smoke opacity vs particulate mass
• Duplicated alarms: the same condition shown in three places with different wording

A Worked Dashboard Specification for One DG Set

The specification below is an illustrative example for a 250 kVA set with a heater-assisted RECD, designed for a BMS or web platform.

Operating view (one screen)

Zone	Content
Top strip	Site / DG2 · RUNNING 1 h 42 m · 118 kW (59%) · On DG (grid failed) · Updated 5 s ago · 14:32 IST
Left panel (exhaust path)	Back-pressure 41 mbar (limit 60) · Inlet 395 °C · ΔP 49 mbar (normal at this load 40–55) · Outlet 388 °C · Regeneration: idle · Heater: off
Right panel	1-hour trends: ΔP, back-pressure, outlet temperature
Bottom	Active alarms (none) · Link: alarm guidance · Link: maintenance view

Maintenance view

Panel	Content
ΔP at standard condition, 12 months	Weekly points; reference line from acceptance test; cleaning events marked
Back-pressure vs load, 90 days	Scatter, with engine limit line
Regenerations, 12 weeks	Bar chart: successful vs failed
Counters	DG hours; hours since cleaning; hours since last full test
Sensor health	Last check dates; current faults
Alarm history, 90 days	Table, grouped by type

Management view (monthly)

KPI	This month	Last month	Target
DG running hours	64	41	—
Data availability	97.8%	88.2%	≥ 95%
ΔP index	1.08	1.05	< 1.30
Peak back-pressure margin	11 mbar	14 mbar	> 5 mbar
Regeneration success	9/9	6/7	100%
Critical alarms (ack time)	0	1 (12 min)	—
Maintenance compliance	100%	100%	100%
Days since last full test	212	182	< 365
Estimated PM avoided	1.4 kg	0.9 kg	—

The previous month’s lower data availability prompted a check of the gateway power supply, which had been on a grid-only circuit. See RECD monitoring during a power failure.

Dashboards for Multi-Site Portfolios

Organisations with many sites, such as retail chains, banks, telecom operators and campuses, need a portfolio view above the site layers:

• A list or map of sites, each with a simple status: normal / warning / alarm / not reporting
• Sort by the worst condition first
• Columns for data availability, ΔP index and days since last test
• Click through to each site’s operating view

The “not reporting” status is critical. At unattended sites, a site that has gone silent should rank as high as one in alarm. See how to set up remote RECD alerts for an unattended site.

Where the Dashboard Lives

Common options:

• On the BMS, as part of the DG pages. See integrating RECD monitoring with a building management system
• On the RECD controller’s local display, which usually shows only the operating layer
• On a web or cloud platform, which is useful for remote sites and portfolios. See does an RECD need internet access to operate

Whatever the platform, keep the local display functional. If the network or platform fails, operators on site still need to see the basics.

Reviewing and Improving the Dashboard

After a few months of use, ask users:

• Which tiles do you look at, and which do you ignore?
• Did the dashboard help you spot a real problem early?
• Were there problems the dashboard didn’t show?
• Is anything confusing, inconsistent or hard to read on a phone?

Remove what isn’t used, sharpen what’s confusing, and add what was missing. A dashboard that’s never revised slowly fills with clutter. Record each change, so everyone knows why the screen looks the way it does.

Related Reading

• How to set up remote RECD alerts for an unattended site
• What should an RECD event log record?
• How to align RECD sensor timestamps with generator runtime logs
• Detecting missing data in RECD monitoring records
• How exhaust back-pressure affects a DG set

Frequently Asked Questions

Can a dashboard show the RECD’s particulate reduction in real time?

Not reliably. Particulate reduction is measured by gravimetric testing under controlled conditions. A dashboard can show indicators of health, such as ΔP, temperatures, back-pressure and regeneration, and the results of the last formal test, but a live “efficiency %” from simple sensors would be misleading.

What’s the single most useful chart on an RECD dashboard?

A long-term trend of ΔP at a standard operating condition, with the acceptance-test reference value marked. It shows soot and ash build-up, cleaning effects and possible damage on one chart.

Should the dashboard show readings when the DG is stopped?

It can show the last values, clearly labelled as from the last run, or grey them out. Showing them as live values is misleading, because ΔP and temperatures without exhaust flow don’t describe RECD health.

How often should the dashboard update?

For the operating view, every few seconds is plenty, since RECD values change slowly. Long-term views can use hourly or daily aggregates.

Should routine opacity readings appear on the dashboard?

Yes, if they’re taken regularly at a standard condition. Add them to the maintenance view as a trend with the acceptance-test reference value marked. Manual readings can be entered into the platform or BMS, labelled with the date, load and instrument used, so they sit alongside automatic data.

Can a dashboard replace routine site visits?

No. It reduces the need for unplanned visits and helps target the planned ones, but physical inspections are still needed for leaks, insulation, mountings, corrosion and anything sensors can’t see. Use the dashboard to decide what to look at during each visit.

Who should design the dashboard?

The BMS or platform provider builds it. The content should be agreed with operators, the DG maintainer and the RECD supplier, who knows which indicators and ranges matter for the device.

Show Less, Mean More

A useful RECD dashboard isn’t the one with the most data. It’s the one that lets each user answer their question quickly and trust the answer. Aceget supplies CPCB type-approved Retrofit Emission Control Devices for DG sets from 25 kVA to 2500 kVA. We can recommend indicators, reference values and alarm limits for your dashboard. See our certifications, or contact our team.