Productization
Energy

Energy Service voltage

Delivered voltage at the service, trended against the nominal band. Sustained under- or over-voltage overheats motors and shortens the life of electronics; flagging it protects equipment and tells a real grid problem apart from an internal fault — avoiding a wasted truck roll.

01 Metric Definition

What it measures and how it is calculated

What it measures

The voltage delivered to the site at the service entrance, sampled from the meter and trended against the equipment's nominal voltage band. Service voltage is a power-quality metric: it tells you whether the grid is delivering clean power, or whether sustained under- or over-voltage is quietly shortening the life of every motor and electronic load behind the panel.

How it is calculated

Service voltage is read directly from RMS voltage at the service entrance and compared, interval by interval, against the nominal band for the site's supply. The metric is a deviation from that band, not an absolute number — 208V is healthy on a 208V service and a fault on a 240V one, so the band is set per service before the metric is activated.

Availability depends on power-quality metering at the service. This is the Optimized differentiator for Energy: the same submetering that reads per-circuit current also reads service voltage. Without it the metric is not produced — a site on a standard revenue meter alone does not surface voltage trends.

Two sources anchor the read: the meter telemetry that produces the voltage trend, and the utility bill, which confirms the tariff and the service the customer is paying for. There is no weather or occupancy normalization in V1 — voltage is a grid-side signal, so a sustained out-of-band trend is flagged for a walkthrough, not auto-attributed to a load pattern.

The exact band and tolerance window vary by client, supply voltage, and equipment mix. Calibrate per client before activating compliance reporting.

Reference thresholds

RangeClassificationInterpretation
Below the lower band (e.g. more than ~8% under nominal)Sustained undervoltage — flagChronic low voltage forces motors to draw more current for the same work, overheating windings and shortening compressor and fan-motor life. The service, transformer, or neutral is the utility's responsibility.
Within the nominal band (ANSI C84.1 Range B ≈ ±8–10% / EN 50160)In range — healthyDelivered voltage is inside the band the equipment is rated for. No action; continue trending.
Above the upper band (e.g. more than ~5% over nominal)Sustained overvoltage — flagSustained high voltage shortens the life of LED drivers, POS electronics, and refrigeration controls. Also a grid-side condition for the utility to correct.
Brief excursions inside startup / inrush windowsExpected — ignoreShort voltage sag at motor start or transient spikes are normal. Only sustained out-of-band trends across consecutive intervals are flagged.

Reference bands only, anchored to ANSI C84.1 Range B and EN 50160 as examples. Calibrate the band, the tolerance, and the consecutive-interval window per client and per supply voltage before activating compliance reporting. Out-of-band voltage is the utility's responsibility — the band defines when to file, not what to repair on site.

Portfolio compliance target

There is no fleet compliance percentage for service voltage in the same sense as the comfort metrics — healthy is simply “in band, all intervals.” Where a portfolio voltage-health target is useful for reporting, define it with the client (a reference starting point is 95% of sites in band); this is a starting point only, calibrated per client.

02 Impact

How this metric moves the customer value drivers

The table below shows how moving Service voltage impacts each customer value driver the product is designed to improve — the metric page explains the mechanism; the product pages express the magnitude.

Value driverImpact strengthHow Service voltage moves this lever
Asset lifespanDirect, primarySustained under- or over-voltage is one of the most damaging grid conditions for the equipment behind the panel. Low voltage overheats motor windings; high voltage degrades LED drivers and electronics. Catching an out-of-band trend early and pre-empting the damage — a compressor check before burnout, protecting sensitive POS and refrigeration electronics — is the most direct lever this metric pulls. Energy Intelligence reads the meter; it does not command it, so the protective action runs through scheduled maintenance and the customer's other Keedian products, never through the meter.
Avoided truck rollsDirect, strongWhen equipment trips, the voltage trend is the signal that separates a real grid problem from an internal fault. Attributing a trip to a documented power-quality event prevents a wasted diagnostic dispatch chasing a fault that was never on the site — and redirects the response to where it belongs: a power-quality claim with the utility, or a coordinated electrician visit, rather than a truck roll to ‘fix’ something the site cannot fix.
03 Detection

How it surfaces and when it is reviewed

How it surfaces

Alarm
Configured to fire when RMS voltage at the service sits below the lower band, or above the upper band, for several consecutive intervals — the sustained trend, not a single sag or spike.
Equipment
Voltage shown out of band in the site's power-quality view, with the direction of the deviation (under or over) and the affected window visible.
Site
A voltage event correlated against equipment current and trips, so a grid-side cause can be told apart from an internal fault before anyone is dispatched.
Portfolio
Sites with recurring or chronic out-of-band voltage surfaced in the portfolio view — the set worth pushing the utility on, or worth a power-quality claim.

Review cadence

Real-time
Alert triggered when voltage holds out of band across consecutive intervals, where power-quality metering is present and the alarm is configured.
Weekly
Operations reviews sites with active or recent voltage events, and reconciles any equipment trips against the voltage trend.
Monthly
Reviewed in the MBR — chronic and recurring out-of-band sites, and the status of any open utility power-quality claims.
04 Alarms

Principal alarms derived from this metric

The table below summarizes the alarms that fire directly from Service voltage. Each row links to the full operational detail (trigger, preconditions, action plan, human role, escalation, prevention) in the SOPs catalog.

AlarmDescriptionSeverityTierAI executes?Value driversSOP
Sustained undervoltage Service voltage sits below the lower band for several intervals — chronic undervoltage overheats motors and shortens their life. High Optimized Hybrid Asset lifespan · Avoided truck rolls Open SOP →
Sustained overvoltage Service voltage sits above the upper band for an extended window — overvoltage shortens the life of LED drivers and electronics. Medium Optimized Hybrid Asset lifespan · Avoided truck rolls Open SOP →
More alarms in development

More alarms in development (single-metric): voltage imbalance across phases, and brown-out / sag-frequency trending. Composite power-quality alarms that combine voltage with per-circuit current and equipment trips for automatic grid-vs-internal attribution will appear in a future release.

05 Actions

What to do based on the alarm

The action plan for each alarm lives on its own SOP page in the SOPs catalog — with the diagnostic steps, human role, value drivers, escalation, and prevention specific to that alarm. The list below maps each alarm to its SOP.

  • Sustained undervoltage → — Service voltage sits below the lower band for several intervals — chronic undervoltage overheats motors and shortens their life.
  • Sustained overvoltage → — Service voltage sits above the upper band for an extended window — overvoltage shortens the life of LED drivers and electronics.

Power-quality metering not present or band not on file (metric-level — not a single alarm)

Service voltage is only produced where power-quality metering reads the service and the nominal band is on file. If either is missing, the metric cannot be trusted: do not report voltage health for the affected sites, and do not let an absent reading look like ‘in band.’ Confirm the metering is in place and the band is set for the site's supply voltage before any voltage alarm is treated as real.

Responsible
Keedian technical team
Urgency
Medium — without metering and a calibrated band, voltage health cannot be reported
Client approval
No

Recurring grid-side voltage across multiple sites in one area (metric-level — utility pattern)

When several sites on the same utility or feeder show the same out-of-band voltage in the same period, this is a grid pattern, not a site-by-site fault. Consolidate the evidence and raise it with the utility as one power-quality case rather than filing site by site. Present it in the MBR as a grid condition the customer's sites share, with the equipment-damage exposure quantified per the equipment behind each service.

Responsible
Keedian account manager (utility coordination) + technical team (evidence)
Urgency
Medium — escalates if equipment damage is accruing
Client approval
Yes — a utility claim is filed in coordination with the customer

Voltage event coincides with an equipment trip (metric-level — attribution before dispatch)

Before dispatching for an equipment trip, check the voltage trend for the same window. If a documented out-of-band event lines up with the trip, attribute it to power quality rather than chasing an internal fault — file or update the utility claim and pre-empt the damage (a compressor check, protecting sensitive electronics). This attribution step is what avoids the wasted truck roll; it reads across alarms and does not belong to any single one.

Responsible
Keedian operations team
Urgency
High — attribution prevents both equipment damage and a wasted dispatch
Client approval
No for attribution — Yes if an electrician or utility visit is coordinated
06 Escalation

When and how to escalate

Per-alarm escalation criteria live in the Escalation block of each SOP in the SOPs catalog. The patterns below are metric-level — read from the portfolio view, not from any single alarm firing.

Portfolio-level patterns — typically communicated in the MBR
  • Sites with chronic out-of-band voltage that persists across multiple months — the set where the utility should be pushed on the service, transformer, or neutral
  • A cluster of sites on the same utility or feeder showing the same voltage condition — consolidate into a single power-quality case rather than filing site by site
  • Open utility power-quality claims and their status — what has been filed, acknowledged, and resolved, with the equipment-damage exposure that motivated each
Cross-alarm urgency — typically requires out-of-cycle communication
  • A voltage event correlated with an equipment trip or damage in progress — attribute to power quality, file the claim, and pre-empt further damage without waiting for the next MBR
  • Sustained overvoltage threatening sensitive POS or refrigeration electronics across a site — the failure mode is silent until equipment dies, so surface it before it does
  • Power-quality metering or band-of-record dropping out across the portfolio — pause voltage reporting until the source is restored, so an absent reading is never mistaken for healthy voltage
07 Prevention

Controls to avoid recurring issues

Configuration controls

  • Set and validate the nominal voltage band and tolerance for every site's supply voltage before activating the metric — the band, not an absolute number, is what defines out-of-band
  • Confirm power-quality metering is reading the service at install — it is the Optimized differentiator that makes the metric possible at all
  • Record the utility, feeder, and service details per site so a grid-side pattern can be recognized and a claim filed against the right party

Monitoring controls

  • Configure alerts for sustained under- and over-voltage separately — they damage different equipment (motors vs. electronics) and carry different urgency
  • Correlate voltage events against per-circuit current and equipment trips, so a grid problem is told apart from an internal fault before anyone is dispatched
  • Track band-compliance per site month over month to surface chronic issues worth raising with the utility proactively

Reporting controls

  • Include sites with active or recurring voltage events, and the status of any open utility power-quality claims, in every MBR
  • Quantify the equipment-damage exposure behind each chronic site — the motors and electronics at risk — to justify the claim and the pre-emptive maintenance
  • Maintain a log of voltage events and their attribution (grid vs. internal) so the avoided-dispatch value is visible and so feeder-level patterns become obvious