TB-5526 · REV F · Technical newsheet
Electronic TestDevice profile
Why You Can't Buy a Safety-Certified 120-V Regenerative E-Load
A 1,000-W regenerative e-load on a shared power strip can overload strip wiring while the breaker sees only net power — which is why NRTLs won't certify 120-V plug-in regenerative loads.
By Amara Osei5 min read924 words
Features
- Dissipative e-loads run at 0% efficiency, converting all absorbed power to heat; regenerative units return energy to the grid via a bidirectional inverter stage.
- Regenerative e-loads become economically attractive above 10 kW; most connect via hardwired three-phase 208 V or 480 V AC dedicated circuits.
- A 1,000-W regenerative load on a power strip with 2,400 W of supplies yields 1,400 W net draw — no breaker trip — while internal currents can exceed the strip's rating.
- NRTLs withhold certification under the 'reasonably foreseeable misuse' principle; warnings are considered the weakest form of risk control.
- Bob Zollo, Keysight solution architect for battery testing, concludes no reputable vendor will offer a safety-certified 120-V AC low-power regenerative e-load.

A 1,000-W regenerative electronic load plugged into the same power strip as 2,400 W of supplies keeps the breaker happy while overloading the strip's wiring. That arithmetic, laid out by Keysight's Bob Zollo, is why no reputable vendor will sell you a safety-certified 120-V AC regenerative e-load. The breaker never trips. The wiring still overheats. Fire hazard, delivered quietly.
Zollo, solution architect for battery testing in Keysight's Electronic Industrial Solutions Group, walks through the physics and the safety-standard logic behind the absence of these instruments in a recent technical article. His verdict is blunt: 120-V AC low-power regenerative e-loads will not get certified, and therefore will not exist on the legitimate market.
What is an e-load, and why does the dissipative kind run at 0% efficiency?
An electronic load (e-load) is a programmable instrument that draws controlled current, voltage, resistance, or power to characterize a device under test (DUT) — power supplies, batteries, chargers, DC-DC converters, solar panels, fuel cells, and server power systems among them. The common type is dissipative: power transistors and heatsinks convert everything the load absorbs into heat dumped into the air.
By the metric of power out over power in, a dissipative e-load runs at 0% efficiency. All of it becomes waste heat, which drives heatsink, fan, and HVAC costs that grow painful as power levels climb. Simplicity and low acquisition cost keep it the default below 10 kW.
Takeaway: the dissipative load is cheap to buy and expensive to run forever.
How does the control loop actually regulate?
Basic e-loads operate in constant current (CC) or constant voltage (CV); constant resistance, constant power, and constant impedance modes generally derive from those two. A CC load adjusts the RDS (drain-source on-resistance) of a FET. A series shunt resistor senses actual current; that signal is amplified, compared against a DAC-set programming voltage, and fed back to the gate.
Too much current? The loop raises RDS. Too little? It lowers RDS. If the load can't reach the programmed value, the gate turns on fully and the FET becomes a near short. Maximum current then follows I = V/R, where R is RDS(on) plus the shunt resistance — the load's specified "minimum on-resistance." Beyond that point the load is unregulated.
Takeaway: the loop regulates until it saturates, then it's just a resistor.
Where does regeneration pay off?
A regenerative e-load performs the same test function but pushes the absorbed energy through an inverter stage back onto the utility grid as usable AC power, operating internally as a bidirectional converter. Above 10 kW, and on up to megawatt levels run for long durations, energy recovery and reduced cooling offset the higher purchase price quickly. Below 10 kW, dissipative still wins on price and simplicity.
The catch: a regenerative e-load isn't a load on the grid. It's a power source, and that changes everything — interconnection rules, installation complexity, safety compliance, and facility infrastructure. Most operate above 5 kW, connecting through three-phase 208 V or 480 V AC, hardwired per building electrical code onto a dedicated circuit with its own breaker.
Takeaway: regeneration saves money only where the grid connection is engineered, not plugged in.
Why does a line cord kill certification?
Here is the failure scenario. A 500-to-1,500-W regenerative e-load returning under 20 A could theoretically use a standard 120-V outlet and line cord — a lovely usability feature, plugging in anywhere. The problem is human, not electrical.
A power strip's breaker sits between the strip's outlets and its AC cord. It protects against overload because power normally flows one way: grid, through cord, through breaker, to the plugged-in devices. Insert a regenerative load and current can flow between devices on the strip without ever reaching the breaker.
Zollo's example: three 800-W supplies (2,400 W total, over the strip rating) alone would simply trip the breaker. Add a 1,000-W regenerative load and the net draw the breaker sees drops to 1,400 W — no trip — while internal strip current can exceed its amperage rating. That's a fire hazard with the protection fully intact and fully blind.
Takeaway: the breaker meters net power; the wiring carries gross current.
What does 'reasonably foreseeable misuse' mean for certification?
The standard risk-management term for predictable user actions outside the intended instructions is reasonably foreseeable misuse. During design and safety evaluation, manufacturers must mitigate such actions through inherently safe design, protective measures, or — weakest of all — warnings. Documentation requiring a dedicated branch can't physically stop anyone from using a power strip, and plugging into one is foreseeable even when the manual forbids it.
A Nationally Recognized Testing Laboratory (NRTL) will withhold certification when the misuse is reasonably foreseeable, the hazard is serious (fire, shock, injury), and the manufacturer can't adequately mitigate the risk. All three conditions hold here.
Takeaway: a warning label is the last line of defense, not a design.
The adoption question
Regenerative loads above 10 kW are becoming standard for battery and high-power test floors, with engineered three-phase interconnections. The open question for test-lab planners: as DUT power levels fragment downward, will instrument makers introduce hardware-level anti-islanding (grid-sensing that blocks backfeed onto shared branch circuits) to make low-power regeneration certifiable — or will the 120-V outlet remain a one-way street by design?
via Electronic Design (Source)
Filed under
- electronic-load
- regenerative-load
- power-supply-testing
- battery-testing
- safety-certification
More from Amara Osei
Show full bio
Senior reporter covering people, launches and day-to-day practice in Instruments, test and measurement at Testbench Report.
124 articles
Application notes
- Yokogawa Ships Split-Core AC/DC Current Sensors Aimed at EV and AI Data Center Power Work
- Martindale Rounds Out PAT Range With Top-of-Range Portable Testers
- EV Scale Puts Tensile Testing Standards Back on the Bench
- NIST CTL Advances Power Metrology Program for Grid-Scale Measurement
- Waste Heat Could Power Industrial Sensing in ECS4DRES Project