english
中文
A buyer sends the same request to three suppliers: a 600 × 300 mm panel, 660 nm and 850 nm, dimmable, remote control, CE and RoHS documentation, 500 units. Three quotes come back. Two are close. One is 35% higher. All three spec sheets list the same headline numbers.
The usual next move is to treat the middle quote as the market price and the high one as an outlier. Sometimes that is right. But a spec sheet only carries the values that are easy to print. The differences that decide what the panel is like in year two — and how many of them come back as warranty claims — are not on it.
Here are six places those differences hide, and the questions that surface them.
"660 nm" is a label, not a single value. LEDs leave the factory sorted into bins: wavelength bins, radiant flux bins and forward voltage bins. Two reels can both be sold as 660 nm and differ by several nanometres, with measurably different output at the same drive current.
The cheap approach is to buy whatever bin is available and least expensive that week. The controlled approach is to specify a bin window, purchase against it, and treat a bin change as an engineering change the customer is told about.
Why it matters beyond the first order: if your production panel uses one bin and the replacement or second-order panel uses another, the light looks different to your customer, and any dose figure you published stops matching the hardware you ship.
Ask: Which bin code is specified on the LED datasheet? Do you buy to that bin, or from distributor stock? If the bin changes, who tells me, and when?
Binning costs money twice, in rejected reels and in sorting labour. That is one honest reason two quotes differ.
Every LED loses output as its junction temperature rises. The effect is gradual and it is real: a panel measured cold can show a comfortable number, then settle several percent lower once the heatsink is saturated. Wavelength also drifts as the die heats — a fraction of a nanometre per degree, in the direction that moves the peak away from its nominal value.
Two panels with identical LEDs and identical drivers can therefore perform differently, because one has a thin aluminium-backed PCB, a minimal thermal pad, a small heatsink and no airflow, while the other has a thicker substrate, a proper interface and a fan moving air through a fin stack.
The cheap way to reach a number on paper is simple: overdrive the LEDs, measure immediately, and never publish the warm reading.
Ask: At what ambient temperature was output measured, and after how long? What is the output at 30 minutes as a percentage of the output at minute one? What is the surface temperature at 30 minutes? Which thermal interface material is used, and how thick is the metal-core PCB?
Thermal design is a safety and lifetime question as much as a performance one. Heat is what dries out electrolytic capacitors and degrades solder joints. It is one of the most common reasons a cheap batch clears incoming inspection and fails in the field eight months later.
A bare LED emits across roughly 120 degrees. A secondary lens concentrates that into 30, 45, 60 or 90 degrees. The narrower the beam, the higher the centre reading, and the smaller the area that receives the dose you are advertising.
This is where two honest spec sheets can disagree without either supplier lying. One quotes irradiance at 15 cm from the centre of the panel; another quotes 30 cm from the middle of the array; a third quotes the average across the whole treatment area. Three different numbers, all correctly measured, describing three different products.
For a panel sold into a professional setting or a home, the number that matters is not the peak. It is the area that reaches the target dose at the distance the user will actually work at, and how evenly that dose is spread.
Ask: Can you send a coverage map for the centre and the corners, at my working distance? What is the lens angle, and what is the LED pitch? At 30 cm, what percentage of the nominal centre irradiance do the corners still receive?
Two panels can use the same LEDs and produce different light, because the driver decides how those LEDs are actually run.
A constant-current driver holds current steady as the LEDs heat and as mains voltage moves. A cheaper constant-voltage design with series resistors lets current wander with temperature and supply — so the panel gets brighter or dimmer depending on the building it is plugged into.
Dimming is the second half. Reducing current, known as analogue dimming, changes brightness smoothly. Switching the LEDs on and off rapidly (PWM) is cheaper, and at low frequencies produces flicker that some users notice, that cameras certainly notice, and that shows up in slow-motion video your customer may well shoot for social media.
Ask: Is the driver constant current? What is the measured current ripple? Is dimming analogue or PWM, and at what frequency? Do you have a flicker measurement in the modes I will actually ship — not only at 100%?
An LED datasheet describes a component on a laboratory bench. Your customer receives a finished assembly: lens, PCB, driver, enclosure, fan, firmware. The only number that describes that assembly is a measurement of the assembly.
Ask four things: Was the finished unit measured, or only the LEDs? At what distance, in which mode, with what instrument? Who signed the report, and can I see the raw data rather than the summary table? And how is a production unit compared against the approved sample — 100% test, or AQL sampling, and at what level?
That last question is the one most buyers forget. A good sample and a good report tell you what the factory can do. A stated production control tells you what it will do on your order, on a Friday, in the week before a holiday.
Two panels of the same size can still differ in aluminium extrusion thickness and finish, acrylic versus polycarbonate diffuser, fan bearing type, hinge and stand hardware, cable gauge and strain relief, remote control quality, packaging foam and carton grade, and whether the unit is drop-tested in its retail box.
None of this appears on a spec sheet. All of it appears in return rates. Ask for the bill-of-materials differences rather than the feature list. Most suppliers will tell you, if you ask specifically.
The reverse is worth saying too. A higher quote can reflect better components, more testing, a wider certification scope, a smaller MOQ, a stronger warranty — or simply a different margin. It can also reflect a slow factory with an inefficient line.
Price is evidence of cost structure, not of quality. The only way to turn it into quality is to make the supplier show you what sits behind the number: the bin code, the thermal measurement, the coverage map, the ripple figure, the production test record.
The cheap quote deserves the same treatment. Ask what it leaves out. Sometimes the answer is "nothing that matters for my market". Sometimes it is "the thermal pad, the constant-current driver and the 100% aging step".
And the purchasing decision does not end when the container arrives. It continues in customer support, replacement costs and repeat orders. A supplier who answers technical questions before the first order is usually the same supplier who answers them after it — and the one who sends a price with no explanation is usually the one who disappears when a batch goes wrong. The cheapest quote is not the cheapest purchase if you are the one absorbing the returns.
Most quote comparisons are not comparisons at all. They are three suppliers answering three slightly different questions, while the buyer compares the answers.
Write one requirement sheet and send the same document to everyone. At minimum:
- Intended use and target market
- Wavelengths, and target irradiance at a stated distance
- Treatment area and required uniformity
- Modes: continuous, pulsed, dimming range
- Voltage, plug type, and required certification by model and market
- Enclosure, ingress protection and mounting
- Packaging, labelling, and any retail or marketplace requirements
- Order quantity, delivery window and warranty terms
- Documents required with the shipment: test reports, declarations, batch records
Then ask each supplier to quote against it line by line, and to mark anything they cannot meet. What is left is a real price difference attached to a defined scope. Everything else is noise.
These questions are not hypothetical for us. They are the ones our engineers answer in every OEM project.
Founded in 2017, Shenzhen HSY LED Technology Co., Ltd. manufactures red light therapy panels and other red light devices in a 5,000 m² facility with six SMT pick-and-place machines and three assembly lines. Actual monthly average output is more than 15,000 units, and we hold three months of stock on key components so that a component shortage does not become your delay.
Quality control runs in five stages: IQC, IPQC, 100% aging of 8–24 hours on every unit, FQC and OQC, with AQL Level II 2.5/4.0 sampling. Warranty is set per product category (three years on our full-body panels, one year on all other product categories) and is stated on the quotation rather than implied.
Documentation is issued per model and per market: CE, RoHS, FCC and FDA registration, supported by ISO 9001 and ISO 13485 quality management. We do not claim FDA 510(k) clearance, and we say so plainly. A buyer who is told "FDA approved" without qualification has been told something that will not survive their own compliance review.
MOQ is 100 units. Custom work covering design, tooling, structure, controller, packaging and manuals is quoted per project. You can also read more about our factory before you ask.
Two panels. Same size. Same wavelength claim. Very different quotes. The only way to know what you are paying for is to ask what sits behind the headline specification, and to expect a specific answer rather than a brochure.
So: before you place your first order, what evidence do you ask a supplier for? Send us your requirement sheet and we will answer it line by line, including the parts where the honest answer is "here is the limit".
When comparing suppliers, ask more than "What is your best price?" Ask about technical data, application requirements, operating conditions and the test documentation behind every number. A low purchase price is attractive. A reliable panel is what keeps customers coming back.
And a reliable panel is not a lucky one. It is a panel whose wavelength bins, thermal behaviour and driver were specified before the order, tested after assembly, and documented in a file you can hand to your own compliance team. Ask for that file early. Suppliers who have it will send it without hesitation, and the ones who do not will tell you something useful by hesitating.
Because a specification sheet describes only a few headline values. Differences sit in LED bin selection, thermal design, lens optics, driver topology, testing depth and the rest of the bill of materials — all of which carry real cost but rarely appear in a short spec sheet.
For the exact model and configuration: a test report on the finished unit with measurement distance and mode, LED bin specification, CE and RoHS declarations, FCC documentation where applicable, and the factory's ISO certificates. For FDA, ask whether the supplier holds a registration or a clearance — these are different, and only one of them can usually be claimed.
Ask at what distance, in which operating mode, with what instrument, and whether the reading was taken cold or at thermal equilibrium. A single centre reading does not describe a panel; ask for a coverage map across the treatment area.
No. A higher price can reflect better components and more testing, or a less efficient factory. Price is evidence of cost structure, not quality. The same discipline should be applied to a low quote: ask what it leaves out.
A datasheet describes a component measured on a bench. A finished-panel report describes the assembled product — lens, PCB, driver, enclosure and firmware — which is what your customer actually receives and what your claims should be based on.
