Here's my verdict before anything else: on this mat, the light you cannot see is the part you have no way to audit at home. The BON CHARGE Infrared PEMF Collection manual (V1.0, May 2026) says some LEDs emit visible red light, and other LEDs emit near-infrared light, which is not visible to the human eye. A dark stretch across the panel is consistent with that build. Brightness proves nothing about whether an emitter fires, and it proves even less about how much energy reaches your skin.
One thing would leave that verdict worthless. If the manual that shipped with your exact model lists a different emitter layout, that document wins and none of this applies to your unit. Everything below stays inside the collection manual and the model it covers, so treat your own paperwork as the final word on your own mat.
Read the Panel First on Day One
Before you power anything up, read the emitter description in your model's manual and note which part of the surface is supposed to glow and which part is supposed to stay dark. Do it in that order, because the order is the whole point. Get it backwards, starting with a long stare at the panel and a guess about what looks wrong, and your first week goes into chasing a light that was never designed to appear.
The cost of skipping this is specific. You either return a working unit because half the panel looks dead, or you keep a genuinely faulty one because the red segment you can see looks bright and cheerful. Neither call is reachable from the glow alone. What the surface gives you is one honest signal, the visible red channel, and nothing else on the panel reports itself to your eye.
Four Layers, Four Different Jobs
The manual lists four things layered together: pulsed electromagnetic fields, red and near-infrared light, far-infrared heat, and a crystal layer. The last of those is a blend of amethyst, onyx, tourmaline and quartz, and the manual frames it as a stable, grounding surface — not a light source and not a heat source.
This matters at the buying stage, where the crystal story and the LED story get braided into one glossy paragraph. Pull them apart and ask about each on its own. What does the crystal layer do in your model's documentation? A surface description, nothing more quantitative than that. What does the LED section say about layout and operation? Those are two separate answers, and if you accept one story covering both, you lose the ability to check either of them later, which is exactly when you will want to.
Set Up Session One So Later Comparisons Hold
Session one is where you decide whether the next month produces evidence or noise. Fix the variables you control — the same room, the same spot on the floor for the mat, the same window of the day, the same number of clothing layers, and the duration your own model's manual specifies. Write that baseline down before you switch the mat on, not afterward.
Baseline first. Judgment later. Everything after depends on it. Without that written baseline, week one hands you a pile of sensations with nothing to compare them against, and you usually notice the problem after the week is gone. With one, any later "this felt different" gets checked against room, time and clothing before it gets blamed on the hardware. I am not handing you minute counts or temperature settings here. Those belong to your model's documentation, and a guessed figure would be worse than sending you back to it.
Log Only What You Can Actually See in Week One
Imagine a sheet with three columns and nothing else: session start time, what the panel looked like when you switched it on, and what you can or cannot confirm about how you felt afterward. One row a day. That is a hypothetical worksheet, not a test result, since nobody measured anything to produce it, but the shape of it does real work.
Two effects show up fast. The first is separation: the moment you write "panel — one bright band, rest dark" beside "felt — warm, relaxed, unsure," you stop merging the two into a single story. The second arrives within a few days, when the sheet shows you what your eye is actually able to confirm — the visible red channel, and only that. Everything else on this mat is either thermal or invisible, and neither one gets recorded by looking.
At Three Weeks Compare Sessions Not Sensations
By the third week you have enough rows for the only comparison your setup supports: identical settings against identical settings. Same time window, same duration, same room. If the matched rows still swing from great to nothing, the variable is your Tuesday, not the mat.
Then the boundary, plainly. At home you cannot measure near-infrared output. Your log can demonstrate consistency. It cannot demonstrate intensity, and it cannot demonstrate dose. Anyone who reads consistent entries as clinical effectiveness has walked past the document into something else entirely. Keep the sheet as a record of what you did, never as proof of what the device delivered.
Decide at One Month to Keep or Return
At the one-month mark, the call needs two things at once. The log shows a pattern tied to your session window, and that pattern repeats. The seller can also point to a clause in the manual for your specific model and region rather than a general brochure or a landing page. This clause is the only piece of the product you can verify from your chair.
If only the first holds, you have a habit and no documented product. If only the second holds, you have paperwork and no personal pattern. That is not a knock on the mat. It is an honest read of what you would be paying to keep. And if the only thing you can say at the end of the month is that the lights look good in a dim room, send it back.
Ask These Questions When the Glow Looks Wrong
When a section of the panel looks wrong, the useful move is documentation, not diagnosis. Ask for the manual page labeled with your model and region that shows the visible red and near-infrared emitter layout. Ask for the construction description covering the four layers. Ask for the operating and warning clauses written for that model. Three asks, all answerable on paper by anyone selling you the thing.
Then set the limit. There is no independent clinical or bench report for this product on record, and the manufacturer material supports construction, operation and warnings, nothing beyond that. The question "is the output strong enough" cannot be settled with a manual, and a seller who answers it with a number dressed in documentation language is improvising. Take the page and decline the number.
Which leaves the question sitting under all of this: did you waste the money? Looking at the panel will not answer it, and a good month of moods will not answer it either. The uncomfortable part is that nobody at home can settle it with a measurement. What you can settle is whether anything is on paper. That is where the return window comes in, and it is the only lever you have while it is still open. Read your own sheet, read the page with your model number on it, and stop asking the panel to confess something it was never built to say.