Independent Lab Verification
Single-chip vs.
Dual-chip LEDs:
Independent Lab Testing
Settles the Debate.
We built two red light therapy panels that were identical in every respect except one: the LED package. Same housing. Same drivers. Same fans. Same optics. Same 178 watts drawn from the wall. Then we sent both to LightLab International and asked them to measure exactly what each panel produces. The results, taken directly from LightLab's signed test reports, are decisive.
01
The Marketing Claim vs. The Physics
For years, certain corners of the red light therapy market have promoted dual-chip LEDs as a feature upgrade. The pitch is intuitive: two emitter dies inside a single LED package should produce more light than one. Twice the chips, twice the output — right?
It turns out the physics doesn't cooperate with the marketing. To find out exactly how much it doesn't cooperate, we commissioned controlled, side-by-side, third-party lab testing.
02
Test Methodology
Two panels were constructed using the identical housing, glass enclosure, aluminum mounting plate, fan assembly, and dual True Full LED power supplies. Both panels drew approximately 178 W from a 120 V outlet. The only design variable was the LED package itself.
Both panels were tested by LightLab International (Allentown, PA, USA) — an NVLAP-accredited photometric testing laboratory — using a Labsphere 2 m integrating sphere with spectroradiometer for total radiometric and quantum measurements (4π geometry, per IES LM-79-24, LM-78-20, LM-58-20, ANSI C78.377-2017, and TM-30-24), and a Gigahertz-Optik BTS2048-VL-TEC spectral irradiance meter on a 49-point cartesian grid for application-distance intensity measurements at 12 inches (per IES LM-91-22, LM-79-24, and LM-58-20).
Test ambient temperature: 24.5–25.5 °C. All four reports were issued and digitally signed by Michael L. Grather between December 22, 2025 and January 7, 2026. Full reports are available for download below.
03
Headline Results
All metrics measured by LightLab International. Both panels at ~178 W wall draw.
Source: LightLab International, reports LLIA002849-001 / -002 / -004 / -005
| METRIC | SINGLE-CHIP | DUAL-CHIP | ADVANTAGE |
|---|---|---|---|
| Wall power consumed | 177.5 W | 177.0 W | Identical input |
| Total Radiant Flux (350–1050 nm) | 69.21 W | 53.60 W | +29.1% |
| Wall-Plug Radiant Efficiency | 39.0% | 30.3% | +8.7 pp |
| Total Photon Flux | 424.8 µmol/s | 335.0 µmol/s | +26.8% |
| Photon Flux Efficacy | 239.3 µmol/J | 189.3 µmol/J | +26.4% |
| Luminous Flux | 2,451.2 lm | 1,849.1 lm | +48.6% |
| Avg. Irradiance @ 12 in (350–1050 nm) | 63.02 mW/cm² | 48.65 mW/cm² | +29.5% |
| Avg. Irradiance @ 12 in (Red 600–700 nm) | 35.62 mW/cm² | 27.10 mW/cm² | +31.4% |
| Avg. Irradiance @ 12 in (NIR 780–1050 nm) | 23.22 mW/cm² | 21.35 mW/cm² | +8.7% |
| Target-Area Radiant Power (350–1050 nm) | 14.84 W | 11.30 W | +29.6% |
| Target-Area Radiant Power (Red 600–700 nm) | 8.27 W | 6.29 W | +31.5% |
| Radiant Energy delivered in 19.5 min | 83,540 J | 65,151 J | +18,389 J |
04
Total Radiant Flux: 69.21 W vs. 53.60 W
In an integrating sphere, every photon a panel emits — in every direction — is captured and counted. This is the most fundamental "how much light is the device producing" measurement, and the standard against which radiant efficiency is calculated.
The Single-Chip Panel produced 69.21 W of total radiant flux. The Dual-Chip Panel, drawing essentially the same wall power, produced only 53.60 W — a 15.6 W shortfall.
That is energy the customer paid the wall for that the dual-chip design simply failed to convert into usable light.
The Single-chip panel converts 39.0%.
Same input. One-third more output.
05
Photon Flux: it's the Photons that do the Work.
Photobiomodulation, at the cellular level, is a counting game. The relevant biological effects — cytochrome c oxidase activation, nitric oxide release, mitochondrial ATP production, modulation of reactive oxygen species — are driven by the number of photons of the right wavelength that reach target tissue. More photons of the right energy means more activated chromophores, means more biological signal.
LightLab's photon flux numbers:
That is nearly 90 micromoles of photons per second of difference — every second the device is on. Over a single 20-minute session, this gap compounds dramatically.
Source: LightLab reports LLIA002849-001 / -004
By a typical 20-minute session, the single-chip configuration has delivered 18,389 more joules of radiant energy than the dual-chip configuration. That is dose, missing.
06
Application Intensity: Where the Light Actually Lands
Total flux measures every photon emitted in every direction. What matters at the body is irradiance at the target — the milliwatts per square centimeter actually landing on tissue at the recommended treatment distance.
LightLab measured both panels on a 7×7 cartesian grid (49 points) at 12 inches, the standard photobiomodulation application distance. Three wavebands were captured.
Average Irradiance Across the 49-point grid
| WAVEBAND | SINGLE-CHIP | DUAL-CHIP | ADVANTAGE |
|---|---|---|---|
| Total (350–1050 nm) | 63.02 mW/cm² | 48.65 mW/cm² | +29.5% |
| Red (600–700 nm) | 35.62 mW/cm² | 27.10 mW/cm² | +31.4% |
| Near-IR (780–1050 nm) | 23.20 mW/cm² | 21.35 mW/cm² | +8.7% |
Total Radiant Power Delivered to the Treatment Area
| WAVEBAND | SINGLE-CHIP | DUAL-CHIP | ADVANTAGE |
|---|---|---|---|
| Total (350–1050 nm) | 14.64 W | 11.30 W | +29.6% |
| Red (600–700 nm) | 8.27 W | 6.29 W | +31.5% |
| Near-IR (780–1050 nm) | 5.39 W | 4.96 W | +8.7% |
In every measurement that matters most for surface and shallow-tissue photobiomodulation — total irradiance, total target-area power, and red-band intensity — the single-chip configuration delivers more. The red band, which is the dominant photobiomodulation wavelength range and the one most heavily marketed by every manufacturer in the category, shows the largest gap. The dual-chip panel delivers roughly one-third less red light to the treatment area than the single-chip panel built on the same chassis.
07
Why dual-chip LEDs underperform: the Optics Problem
The performance gap is not an accident. It is a predictable consequence of basic optical geometry.
A conical reflector or refractive optic — the small clear lens you see seated over each LED — is engineered to collect light from a point source at its focal point and redirect it forward into a tighter, more useful beam. When the emitter sits dead-center under the optic, the light is focused along the intended axis. Off-axis emitters produce off-axis beams.
A single-chip LED has one emitter die. That die sits at the optical center, and the optic does the job it was designed to do.
A dual-chip LED has two emitter dies in the same package. They cannot both sit at the optical center. They sit side-by-side, each offset from the focal point by roughly half the spacing between them. The optic above them now has two off-axis sources, each producing a tilted, defocused beam that escapes outside the intended treatment cone.
That escaped light still gets counted in some marketing claims (total wattage of LEDs installed, for example), but it doesn't reach the target. It is lost into walls, ceilings, and floors.
This is exactly what LightLab's measurements show: lower wall-plug efficiency despite identical input, lower target-area radiant power in the band that matters most, and worse red-band uniformity (Max/Min ratio of 1.46 vs. 1.34) — and the red dies are the ones most affected, because they are the most laterally displaced from the optical center.
There is also a thermal compromise inherent to the dual-chip approach: two emitter dies sharing a single thermal pad and a single junction-temperature limit must each be driven softer than an equivalent single-chip die would be driven, or operated at higher junction temperatures with corresponding loss of efficiency and useful life. The lab data is consistent with the former — the dual-chip dies are simply not producing as many photons per watt as their single-chip counterparts.
08
What This Means for Treatment
If two devices draw the same wall power and one delivers 30% more total radiant flux and 31% more red-band irradiance to the target, the practical consequences compound on every session:
- Shorter sessions for an equivalent dose. A target dose of 60 J/cm² in the red band is reached roughly 24% sooner with the higher-output device.
- More dose per kilowatt-hour purchased. Same wall draw, more usable light — meaning more dose per unit of electricity, every session.
- Tighter, more uniform treatment field. Less photon loss to scatter outside the intended treatment zone, especially in the red band where the dual-chip's off-axis geometry hurts most.
- Honest specifications. The number on the spec sheet matches what reaches the body. In a category where customers are asked to take performance claims on faith, that point matters most.
09
The Platinum Therapy Lights position
Platinum Therapy Lights uses single-chip LED architecture across our panel lineup. We did not arrive at this decision through marketing — we arrived at it through bench testing of both configurations and through the optical physics of how secondary optics actually work.
When we elected to validate that decision with independent third-party testing, we sent both designs to LightLab International on the same protocol, on the same chassis, with identical drivers, optics, and thermal management. The reports are reproduced as LightLab issued them. They are signed, dated, and downloadable below in their unaltered form. Anyone — customer, competitor, journalist, regulator — is welcome to verify the numbers.
The marketing for dual-chip LEDs sells the appearance of more. The lab data shows that "more chips" does not translate to more light at the patient. It translates to lower efficiency, lower output, and a less uniform red-band field — at the same energy cost per session.
We chose the design that delivers what we say it delivers. The data is on the table.
10
The Complete Lab Reports
All four LightLab International test reports referenced throughout this article, available in their original unaltered form. Each is digitally signed by Michael L. Grather and dated December 2025 / January 2026.
