Selection & sizing

How to read an LED driver spec sheet

Field-by-field walkthrough of an LED driver nameplate and datasheet - input, output, current accuracy, power factor, THD, tc versus ta, inrush current - plus which missing numbers signal risk, and a receiving checklist you can run on site.

Last updated：2026-10-31

On this page

The one-sentence answer
Reading the nameplate field by field
Field-by-field check
The five numbers hiding between the lines
Three red flags
Receiving checklist you can run today

A datasheet is not there to confirm numbers you already expect. It is there to reveal the numbers that are missing. Anyone can print the easy figures; what decides whether a driver survives three years in your project is usually the five values that never made it onto the nameplate.

The one-sentence answer

Of the twelve nameplate fields, only four actually cause trouble in the field: output current accuracy, tc temperature, the lower input voltage limit, and inrush current. The other eight are a verification exercise. And the five things a datasheet does not print - derating curve, dimming floor, start-up time, the load point at which power factor was measured, and cable compensation range - are where the difference between two quotes really lives.

Reading the nameplate field by field

The figure below is a typical constant-current nameplate with each field annotated. Your unit may print them in a different order, but the field types are the same.

Six zones of a nameplate: input, output, mains-side quality, temperature, inrush, protection

Field-by-field check

Field	What to look for	Typical value	Where it goes wrong
Input voltage range	The lower bound matters more	100-240V / 180-264V	A wide-range unit at 100V usually loses 2-4 points of efficiency
Input frequency	50/60 Hz coverage	47-63 Hz	A 50 Hz-only unit needs separate confirmation for 60 Hz markets
Input current	Basis for breaker and cable sizing	0.45 A at 230 V	Size distribution with a 1.25 factor
Output voltage window (CC)	Sets how many LEDs per string	25-40 V	Running at either extreme hurts efficiency and ripple
Output current and accuracy	5% is the dividing line	3% / 5% / 8%	Parallel 8% units give visibly uneven brightness
Output voltage (CV)	12V or 24V nominal, +/-5%	24 V +/-5%	A measured 22.8 V halves the voltage-drop budget at the far end
Rated power	Output watts or input watts	14 W	Some quote input, some output - the difference is efficiency
Power factor	Note the stated load point	0.95 at full load	Units that fall below 0.85 at half load drag down a whole panel
THD	Check full and part load	below 20%	Dozens switching at once can push neutral current above phase current
tc	The only input to lifetime	85 C / 90 C	Every 10 C lower roughly doubles electrolytic capacitor life
ta	Ambient ceiling	45 C	Light boxes often reach 55-65 C, which forces derating
IP rating	Match it to the mounting position	IP20 / IP65 / IP67	IP does not cover condensation, UV or surge

tc is not ta. ta is the air around the driver; tc is the temperature at a defined point on the case. Lifetime curves are plotted against tc, never against ta. In a sealed luminaire the gap between the two is typically 15-25 C.

The five numbers hiding between the lines

They do not fit on a nameplate, but they decide field behaviour. If they are not in the datasheet, treat them as absent - and absent means risk.

Derating curve - how much load is allowed once ta exceeds the rating. With no curve published, be conservative: reserve 10% of output power for every 5 C above rated ta.
Dimming floor - is it 1% or 10%, and on which dimmer was it measured. A "1%" claim with no dimmer model named is marketing.
Start-up time - from power applied to full output. The difference between 0.2 s and 1.5 s matters in emergency transfer and sensor-triggered scenes.
Part-load power factor - if only full load is stated, estimate part load at roughly 80% of that figure.
Cable compensation range - whether a CV unit offsets voltage drop on the output leads, and by how many millivolts. Ask this on any strip run longer than 8-10 m.

Three red flags

Stop before ordering if you see any of these:

Input power quoted without output power - 14 W in at 88% efficiency is about 12.3 W out, and the LEDs see less still
"50,000 hours" with no tc stated - an hour figure without a tc condition cannot be verified or compared
Certification marks without certificate numbers - marks can be printed; numbers can be checked on the issuing body's site. Validity is always subject to local regulations and the certifier's current requirements

Receiving checklist you can run today

Nameplate versus datasheet - compare all twelve fields character by character; stop on any mismatch
Certificate numbers - copy them and look up status and covered models on the issuer's site
Measure output - for CC units, measure at rated load; anything outside the stated accuracy is a return
Measure tc - after two hours at full load in the final enclosure, put a thermocouple on the marked point; aim for at least 10 C below the rated tc
Inrush current - capture it with a clamp meter at switch-on and recalculate how many units a breaker can carry
Dimming floor - connect the dimmer actually specified for the job, ramp down slowly from 100%, and record where flicker starts and where the light drops out

Budget around 30 minutes per unit. Checking three units before mass production is far cheaper than a batch rework.

For constant-current dimming models with 3% accuracy and a published derating curve, look at H01 and H02 in the H series. For long constant-voltage strip runs, settle the cable compensation question before choosing between K01 and K02. When in doubt, photograph the nameplate and send it over.

Not sure about a datasheet? Send us a photo of the nameplate - we will flag every field that clashes with your project and suggest an alternative from 47 models. Contact →