# Read a LightBurn material test grid and choose the next test [Read as HTML]() Start in LightBurn at Laser Tools → Material Test; the built-in generator creates the grid, so you do not need a downloaded test file. Before interpreting a completed sample, recover what that grid actually varied and what stayed fixed. Then judge the cells against a result you chose first. A visible top mark is not proof that a cut separated. ## Recover the grid's real setup In LightBurn's documented Power/Speed example, speed labels run down the left and power labels along the bottom. Do not assume every grid uses that pairing or orientation: the Param fields select the tested variables, and the labels can be disabled. Open Laser Tools → Material Test and load the original saved Material Test preset (.lbmt), if available. Inspect each axis's Param, Count, Min, and Max, then open Edit Material Setting to see the settings shared by the cells. The dialog shows its current configuration; if it has since changed and the original test was not saved, today's values do not prove what produced an older grid. Printed labels recover only the settings they show; mark any setting missing from both the retained record and the labels unknown. Record the exact material and thickness, job goal, tested axes and units, and shared settings. LightBurn warns that speed units matter; do not convert a remembered number by guess. Use only material and operating ranges approved by the laser maker, and follow its placement, ventilation, supervision, and fire-safety instructions. Preview and frame a new test as LightBurn documents; never leave a running job unattended. ```text Machine/controller: [ ] Material maker/type and thickness: [ ] Original Material Test preset (.lbmt) or retained setup record: [ ] Goal: engraving / cutting; success criterion: [ ] Row Param / units / Min–Max: [ ] Column Param / units / Min–Max: [ ] Shared settings from Edit Material Setting: [ ] Cell labels and physical result checked: [ ] ``` ## Compare like with like For engraving, compare surface appearance against your stated goal. For cutting, use a cutting test and record what you physically checked on the completed sample. A top-side still can help read labels or compare visible edges; a separate underside view can show its appearance, but physical release needs the operator's check. Separate a known failed result from an unknown one: if the operator checked the finished sample and it remained attached, record that observed failure; if only the top is visible, separation is unverified. Do not force a piece free or infer why it stayed attached. | Completed result | Decision it supports | | --- | --- | | Engraving mark meets the stated appearance goal | Keep the labeled cell as an engraving candidate; it says nothing about cutting through | | Operator checked the finished cut sample: clean release and acceptable edge | This cell passes those recorded cutting criteria for this sample and setup | | Operator checked the sample and it remained attached | This cell failed the release criterion; the observation alone does not identify the cause | | Only a top-side image is available | Record appearance, but leave backside quality and separation unverified | ## Worked example: retain the fastest cell that passes Hypothetical example, not a machine test or settings recipe. The original labeled grid holds power constant in one column and shows three speeds S1 < S2 < S3; these symbols stand for the machine-specific values printed on that grid. Before comparing, the operator sets the goal as clean release with an acceptable edge, choosing the fastest cell that meets both. At S1, the sample releases but has excess charring, so it fails the edge criterion. At S2, the checked sample releases cleanly and its edge meets the goal. At S3, the top has a cut mark, but physical inspection finds the sample still attached, so it fails release. S2 is therefore the fastest of these three cells that passes both criteria; the top mark at S3 does not override the observed failure. If a faster acceptable result is worth another test, and the machine maker's permitted range allows it, a narrower test between S2 and S3 can check that boundary while keeping the same material, power column, and shared settings. Record and inspect the new completed sample by the same criteria. This comparison does not establish why S3 remained attached, guarantee the next result, or transfer to a different material or setup. If the physical check is unavailable, S3 remains unverified rather than a known failed release. ## Use a still only when it helps compare the sample LightBurn's printed labels, original .lbmt preset, and saved setup notes are the first evidence to check. For a second opinion on visible cell differences, attach one clear image showing the whole labeled sample; a close-up is useful only when a label or edge cannot be read. A direct attachment is enough for a one-off comparison. When comparing successive completed samples, LetItSee can request a new view from the selected camera. The operator supplies the physical release result. It does not watch a running laser or control LightBurn. ## Prompt to adapt > Here is my completed LightBurn material test and, if needed, a still of the labeled sample. First identify the axes, displayed units, and shared settings from the original .lbmt preset, retained setup record, or printed labels; mark settings absent from those records unknown and distinguish missing context from values in today's changed dialog. My goal and criteria are: [ ]. Compare the reported physical outcomes against those criteria, separating known failures from results that are unverified in the image. Recommend whether any labeled cell qualifies and, only if useful and within my machine maker's guidance, what bounded next test would clarify. Do not infer a cause or invent settings. ## Source documentation - [LightBurn: Material Test]() - [LightBurn: First Material Test]() ## Connect your source Add https://letitsee.com/mcp using remote MCP, Streamable HTTP, OAuth, and image results. Sign in with the same account and choose your source. [Use a phone camera with an MCP assistant]() or [Use a webcam with an MCP assistant]() cover setup and browser requirements. ## Set up without starting a capture [Set up your MCP client](https://letitsee.com/for-agents) with the full tool reference. The separate [capture page](https://letitsee.com/live) can replace the active capture page for your account. It does not take a photo by itself. ## Related guides [All camera and screen guides](https://letitsee.com/guides)