AFDD in the Data Center: Beyond Hardware Circuit Protection
The $1.86B AFDD market focuses on hardware arc-fault interrupters. Data centers need the analytics layer: continuous fault detection at the circuit level.
Are you writing AFDD into your data center specifications right now? If the answer is "we already have arc-fault breakers installed," you are covering one layer. There is a second layer that most RFPs leave out entirely, and it is the one that determines whether you catch equipment degradation before or after it becomes an outage.
The AFDD market is projected at $1.86 billion with 8.2% CAGR (MarketsAndMarkets, 2024). Most of that spend goes to hardware devices: AFCIs, GFCIs, and combination devices that meet NEC 210.12 and IEC 62606 arc-fault detection requirements codes. These do their job. They detect arcing or ground faults and trip the circuit. They are reactive by design, activating at the moment of failure rather than ahead of it.
If you are in procurement, that distinction should change how you write your specifications.
Two layers of AFDD: what your spec should cover
You need both layers. Here's how they differ, and why your RFP probably addresses only one.
Hardware AFDD is binary protection. It detects a specific condition (arc, ground fault) and takes a specific action (trip the circuit). It operates per device, gets installed per code requirements, and runs independently of any monitoring platform. You probably already specify this. Your electrical contractor handles it.
Analytics AFDD is continuous detection across your entire monitored infrastructure. Instead of waiting for a fault to trigger a binary response, it analyzes electrical waveforms to identify the degradation patterns that come before failures. This is the layer most procurement specs miss.
| Dimension | Hardware AFDD (AFCI/GFCI) | Analytics AFDD (Verdigris) |
|---|---|---|
| Detection scope | Single fault type per device | Multiple degradation modes per circuit |
| Response | Circuit trip (binary) | Early warning with diagnostic context |
| Timing | At fault occurrence | Before fault occurrence |
| Coverage | Per-device installation | Fleet-wide, per-circuit |
| Diagnostics | Fault/no-fault | Root cause analysis, trend data |
| Standards reference | NEC 210.12, IEC 62606 | IEC 61557-12 (insulation monitoring) |
Your hardware AFDD is the last line of defense. Analytics AFDD is the detection layer that keeps you from needing that last line of defense.
What analytics AFDD catches in a data center
Here's where this becomes concrete for your facility. Verdigris provides continuous fault detection through 8 kHz waveform analysis. These are the four equipment categories where you get the earliest detection.
UPS rectifier degradation. Capacitor loss elevates specific harmonic components, particularly 5th and 7th order THD. Verdigris Signals tracks these shifts against per-unit baselines and fleet-wide norms. You get early warning before the rectifier drops below operational thresholds. In field deployments, this approach has identified degrading units that no threshold alarm had flagged, with weeks of advance warning.
Compressor faults. Your CRAC and CRAH compressors degrade in ways that change their inrush current profiles and harmonic content. Bearing wear, refrigerant loss, and motor winding degradation each have distinct electrical signatures. Continuous waveform analysis separates them. You know which mode is failing, not merely that something triggered an alarm.
Transformer insulation stress. Winding insulation degrades under thermal and electrical stress. That degradation appears in high-frequency waveform data before it produces a measurable temperature change. If transformer monitoring stops at temperature, the early window is invisible.
Phase and load anomalies. Unbalanced loading, loose connections, neutral conductor degradation. Each produces measurable changes in phase relationships and harmonic distortion patterns. These conditions stress your distribution equipment and raise failure risk across the entire power path.
Where IEC 61557-12 fits into your spec
If you are writing specifications, IEC 61557-12 — covered in the broader power-quality monitoring framework — gives you the framework for continuous electrical monitoring devices. It defines accuracy classes, measurement functions, and communication protocols for power quality monitoring in low-voltage distribution.
Verdigris EV2 hardware delivers measurement capabilities aligned with IEC 61557-12, including harmonic analysis, power quality assessment, and continuous measurement at the circuit level. The platform then adds automated analysis on top of those measurements: Signals with diagnostic context, trend data, and fleet-wide comparisons that the standard does not require but your operations team needs.
When you cite IEC 61557-12 in your specification, you establish the measurement baseline. Analytics AFDD builds on that baseline to deliver fault detection beyond what measurement alone provides.
How it deploys (without touching your existing systems)
This matters for procurement because deployment complexity affects your project timeline and risk. Here's the architecture.
Hardware layer. Verdigris EV2 meters install non-invasively using split-core current transformers. No downtime required. No modifications to existing circuits. Each meter captures at 8 kHz with sub-second data transmission. If you have concerns about installing new equipment on production circuits, this is the answer: the CTs clamp on. Nothing gets disconnected.
Analytics layer. Verdigris Signals applies harmonic signature analysis, baseline comparison, and fleet-wide anomaly detection to the continuous waveform data. Detection models are tuned per equipment type: UPS rectifiers (shipped), CRAC compressors, transformers, and motor loads (expanding through Signals Lab).
Integration layer. Alerts route to your existing operational tools. ServiceNow integration for automated work orders. MQTT and REST APIs for DCIM and BMS integration. Email and Slack for direct notification. You do not need to replace your existing systems. Analytics AFDD sits between your hardware protection (which reacts to faults) and your BMS polling (which checks thresholds at 1 Hz).
Your analytics AFDD specification checklist
When you write your next facility RFP, include these requirements alongside your standard hardware AFDD specifications.
| Specification | What to require | Why it matters |
|---|---|---|
| Sampling resolution | Minimum 4 kHz for harmonic analysis through 25th order | Below 4 kHz, you miss the harmonic content that indicates early degradation |
| Coverage | Per-circuit monitoring across UPS, distribution, and mechanical systems | Fleet-level averages hide individual unit degradation |
| Detection capability | Degradation trends in addition to threshold alarms | Threshold alarms fire at the problem, not before it |
| Diagnostics | Root cause context with each alert | "Something is wrong" is not actionable. "Rectifier 47 shows elevated 5th harmonic THD" is |
| Integration | API access for BMS/DCIM, work order system connectivity | Your operations team will not adopt a tool that lives outside their workflow |
| Standards alignment | IEC 61557-12 measurement functions, IEC 61000-4-7 harmonic analysis | Established standards, not proprietary benchmarks |
| Deployment | Non-invasive installation, zero operational downtime | If it requires a maintenance window to install, you will defer it indefinitely |
Stop specifying half the solution
Your hardware AFDD devices protect against acute electrical hazards. Your analytics AFDD layer detects the gradual degradation that leads to those hazards. A specification that stops at the first layer pays for protection without paying for prevention.
Get a technical briefing from Verdigris engineering. We can walk through how analytics AFDD maps to your specific facility requirements and existing procurement framework.