Every check passes. Nothing here has a locked rotor to publish an inrush figure for, so the absence of one is a fact about the hardware rather than a hole in the library.
Official spec + Derived by WattPair · verified 2026-08-08 · US 120 V unit × LED TV
Nominal capacity4,096 Wh
Surge headroom10×+
Running108 Wof4,000 Wcontinuous output
0 – 4,000 W across the rails above
No start-up rail: this load has no locked rotor to publish an inrush for, so no start-up demand was checked.
Two numbers decide this, and readers use them interchangeably: watts are the ceiling on what the inverter holds at any instant, watt-hours are how much energy the battery holds. A machine can have plenty of one and not enough of the other.
The verdict is a calculation, not a lookup. What it will not do is print a runtime, and the reason is specific: we read 4 of EcoFlow’s own published documents on 2026-08-08 and found no depth of discharge or inverter efficiency for the US 120 V build in them, so there is no usable-capacity figure to divide.
Where those numbers come from
A blank is never left blank. A field we have found no figure for says so, and a field whose
sources disagree — two documents, or one document with itself — says that instead of picking
one.
No usable-capacity figure for this machine
EcoFlow states a nameplate watt-hour number for it, and we read 4 of EcoFlow’s own published documents on 2026-08-08 and found no depth of discharge or inverter efficiency for the US 120 V build in them. Dividing the nameplate by this load would be wrong by 20–35% in the flattering direction, so this page shows the nameplate for what it is and refuses to turn it into hours.
Missing: station.usable_capacity_wh
Energy rate instead of a runtime
108 Wh per houraveraged over this load’s duty cycle
This number needs no battery figure at all, which is why this page can give it when it cannot give hours. Divide any usable-capacity figure you trust by it and you have your own runtime, with our arithmetic visible instead of hidden.
What an LED TV demands
Running power107.5 W Official spec · verified 2026-08-01 · high confidence
Start-up duration50 ms Official spec · verified 2026-08-01 · low confidence
Duty cycle100% Official spec · verified 2026-08-01 · high confidence
What the EcoFlow DELTA Pro 3 offers
Continuous output4,000 W Official spec · verified 2026-08-01 · high confidence
Surge ceiling8,000 W Official spec · verified 2026-08-01 · high confidence
Nominal capacity4,096 Wh Official spec · verified 2026-08-01 · high confidence
Inverter waveformPure sine wave Official spec · verified 2026-08-01 · high confidence
BLUETTI Elite 30 V2 (US 120 V) reaches the same conclusion on an LED TV at 600 W continuous, and it is $239. Cheaper is only cheaper when the answer is the same — this one is. See that pairing →
Yes. 108 W of running draw against 4,000 W of continuous output — 2.7% of it. The verdict is computed from running watts, and inverter waveform — the working is in the evidence section above.
How long will an EcoFlow DELTA Pro 3 run an LED TV?
We will not put a number on it, and that is a finding rather than a gap. We read 4 of EcoFlow’s own published documents on 2026-08-08 and found no depth of discharge and no inverter efficiency for the US 120 V build in them. What it does state is a nameplate watt-hour rating. Dividing that nameplate by this load produces a figure that is typically 20–35% too high while looking exactly as authoritative as a real one. What this page gives instead is the load's energy rate in watt-hours per hour, which needs no battery figure at all.
Why is there no runtime on a page that says yes?
Because "yes it runs it" and "yes for this long" are two different claims with two different evidence requirements, and this machine only supports the first. The verdict needs a continuous rating and a surge ceiling, which EcoFlow publishes. Hours need a usable-capacity figure, which it does not. Twenty of the twenty-six machines in our library are in that position, and the alternative — nameplate watt-hours divided by label watts — is the single most common wrong number on this topic.