The dc numbers, in one place, for the first time.
Article 310 is new. It does not describe a piece of equipment — it sets the dc hazard thresholds that the rest of Chapter 3 leans on. Batteries, capacitors, supercapacitors and photovoltaic systems all now point back to a common set of numbers instead of each carrying its own.
310.1 identifies dc electrical hazard thresholds for the implementation of electrical safety-related requirements, and it applies only to the systems covered in Chapter 3. It is a reference article, not a work-practice article — the work practices stay in 320, 360, 370 and 380.
§ The three thresholds
310.2 requires that risk control be selected and applied from the hierarchy of risk control if any of the following are exceeded. Note that two of the three are two-part tests — voltage alone does not trigger them.
| Hazard | Threshold | The trap |
|---|---|---|
| Contact thermal | Power ≥ 1000 W | Single-part test. No voltage term at all — this one can be exceeded by a string that is far below the shock threshold |
| Electric shock | Voltage ≥ 100 V dc and current > 40 mA | Both conditions. A 480 V dc bus with no available path above 40 mA does not meet it |
| Arc flash | Voltage > 150 V dc and incident energy > 1.2 cal/cm² | Both conditions. 1.2 cal/cm² is the same onset-of-second-degree-burn value used throughout the standard |
§ Why contact thermal is listed first
The shock and arc flash thresholds both carry a voltage term. The contact thermal threshold does not. That single asymmetry is the most important thing on this page.
A 48 V telecom string is under the 100 V shock threshold and under the 150 V arc flash threshold. It is not under 1000 W. Available short-circuit power on a stationary string is routinely tens of thousands of watts — and the injury mechanism is a tool, a ring or a watch band heating to failure across the terminals. Run a string through the check →
Every dc system in a critical facility now resolves against these three numbers: the UPS dc bus and its link capacitors, the battery string behind it, any supercapacitor ride-through module, and campus PV if you have it.
The practical effect is consistency. Before 2027 a technician could be told one number in the battery room and a different number at the PV combiner. Article 310 makes the baseline the same everywhere, and the equipment articles add on top rather than starting over.
The 1000 W contact thermal threshold is the one to socialize. It is the threshold most likely to be exceeded by equipment your team currently treats as low-hazard, and it is the only one of the three that does not care about voltage.
§ What to update, in order
Any dc procedure that states a voltage-only threshold. If it says "below 100 volts, no PPE required," it is now wrong — contact thermal has no voltage term.
Battery room and UPS dc work packages — confirm the available short-circuit power is calculated and recorded, not assumed.
Training material — the three thresholds belong on a slide together, with the two-part tests shown as two-part.
PV and BESS procedures — these now inherit a common baseline they did not have before.
The two-part tests are read as one-part tests in the field almost every time. "Over 100 volts means shock PPE" drops the 40 mA condition; "over 150 volts means arc flash PPE" drops the incident energy condition. Both readings are conservative, so they rarely cause an injury — but they train people to ignore the second half of a requirement, and on the contact thermal threshold there is no second half to fall back on.
The full Chapter 3 map →
Every article that moved in 2027, verified against NFPA LiNK.
Article 360 — batteries →
Where these thresholds get applied first.
Battery threshold tool →
Put a real string through the numbers.
NFPA 70E 2024 → 2027 transition training
Article 310 did not exist before. A focused session covers what is new in Chapter 3 and what your documents have to say now.
See the schedule Bring it onsite