The complete manual for learning laser engraving properly, and building a real business on top of it.
How the beam actually works, and which type of machine you need before you spend a cent.
A cheap machine and a professional-looking result rarely coexist. In laser engraving, they do, and that's the whole reason this hobby keeps turning into a business.
A £350 diode laser sitting on a garage workbench can turn out a wedding gift that looks like it came from a proper workshop. That's the reason this has become such a popular thing to turn into a business: the barrier to entry is low, but the output doesn't look cheap.
This guide treats learning the craft and starting the business as one path, because in practice that's what happens. Almost nobody sets out to build a laser business on day one. You buy a machine because it looks fun, you make some coasters, you post them somewhere, a friend asks if you'll make one for their dad's birthday, and six months later you're calculating whether you can quit your job.
Read Parts One and Two before you spend any money. The buying decision in Chapter 4 only makes sense once you understand Chapters 2 and 3. Part Three is something you'll come back to constantly once you own a machine, don't try to memorise it up front. Part Four you'll mostly ignore until something breaks or your machine racks up a hundred-odd hours, at which point it becomes the most important section in the book. Part Five is worth reading straight through once you're serious about selling, then dipping back into as a checklist.
A laser engraver points a focused beam of light at a material. Wherever the beam hits, it either burns or vaporises the surface (engraving, doesn't go all the way through) or vaporises straight through (cutting). The machine steers that beam using two settings you'll be adjusting for as long as you own the machine: speed (how fast the head moves) and power (how strong the beam is). Nearly everything else in this guide is detail layered on top of that one idea.
A laser produces a beam of light at a specific wavelength, focused down through a lens to a spot that can be smaller than a millimetre wide. At that focal point the energy is concentrated enough to heat the material almost instantly, burning it, melting it, vaporising it, or in some cases just changing its colour without removing anything. That last effect is how you get a dark mark on light wood, or a contrast mark on some metals and plastics, without physically cutting into the surface.
The head moves that focused point across the material in one of two ways.
Most real jobs use both. A photo gets rastered, the cut line around the edge of the piece gets vectored.
| Term | What's happening | Typical setting |
|---|---|---|
| Engrave | Beam removes or darkens the surface without going through | Fast speed, moderate power, several raster passes |
| Cut | Beam goes all the way through the material | Slow speed, high power, one vector pass (or two or three on thick stock) |
| Score | A single light vector pass, not meant to cut through | Fast speed, low power |
| Mark | Colour change on the surface, no material removed | Depends heavily on the material |
Scoring vs. cutting trips people up constantly when they're new. Scoring is deliberately weak: you're using it for fold lines on cardboard, registration marks, or fine decorative lines you don't want cut all the way through. If your "cut" line isn't going through and you didn't mean to score it, that's not a scoring feature working, that's your power too low or your speed too high for the material thickness. Check your settings before you assume the machine is broken.
DPI (dots per inch), sometimes called LPI on CO2 machines, controls how many lines the laser draws per inch as it rasters, and how finely it can turn on and off along each line. Higher DPI means finer detail and smoother gradients but a much longer engrave time, and more heat going into the material, which on wood especially means more charring and darker edges even where you didn't want them.
Every result a laser produces comes down to how much energy actually lands on the material, and that's a function of power and speed together.
There is no universal setting sheet you can copy and trust completely, and anyone who tells you otherwise hasn't cut enough different sheets of the "same" plywood. Settings depend on your machine's real power output (not the number on the box, more on that in Chapter 4), the lens you're using, the material, its thickness, and even the weather. This is why Chapter 10, on building your own test grids, is not a skippable chapter. It's the actual method professional shops use instead of guessing.
Wood is hygroscopic: it absorbs and releases moisture from the air constantly. A sheet of birch plywood that cuts cleanly through on a dry day can refuse to cut through on the exact same settings a few weeks later if the humidity's gone up and the wood has taken on moisture. Damp wood needs more energy, some of it goes into boiling off water before the wood itself will burn.
If your machine is already near its power ceiling for a given thickness, this is often the real reason a cut that used to work suddenly needs an extra pass, not a dirty lens or a dying tube (though check those too, see Chapter 14). Leather and cardboard behave similarly. If your workshop isn't climate controlled, expect your settings library to need small seasonal adjustments rather than treating it as fixed forever.
The lens focuses the beam down to its smallest, most powerful point at one specific distance from the material, called the focal point. Get this wrong and everything suffers: cuts don't go through cleanly, engraves come out fuzzy instead of crisp, and you'll find yourself cranking power up to compensate for a problem that isn't actually a power problem.
Three ways machines handle this:
Focusing slightly above or below the true focal point widens the beam spot, which can be useful for bold, less-detailed engraves on signage, or for reducing visible line structure on some materials. Not something to reach for as a beginner, but worth knowing once normal focusing feels routine.
A sealed glass tube filled with a CO2 gas mixture gets excited by an electrical current and produces an infrared beam, usually around 10.6 microns wavelength. Mirrors bounce that beam through the machine to a focusing lens at the head.
CO2 is the workhorse choice for a business doing wood and acrylic signage, awards, cutting boards, coasters, and ornaments, anything where you're cutting a shape out and engraving detail into it in the same job.
A semiconductor diode module, related to the same tech in a laser pointer but far more powerful, emits a beam directly. No gas tube, no big power supply, no warm-up. Most are blue light around 450nm, though some newer modules add an infrared diode alongside the blue one to help with darker or reflective materials.
"20W" on a diode laser's box can mean the optical output (the actual beam strength) or the power the module draws from the wall, and those are very different numbers. A module can draw 40W from the wall and only put out 10W of real optical power. Two machines marketed identically can cut and engrave nothing alike. Look for a manufacturer that publishes actual optical output, and treat vague specs with suspicion. More in Chapter 4.
Diode is the sensible entry point if you're starting from zero, testing whether you actually enjoy this, or working from a spare room or garage where a full CO2 setup isn't practical yet. A lot of small businesses run diode first and add CO2 once volume and revenue justify it.
If you're on a diode machine and can't cut clear acrylic for coloured accents, acrylic paint pens (Posca and Molotow are the common brands) let you fill engraved text or line detail on wood with colour after the fact, giving the layered, high-contrast look people associate with acrylic without ever touching a sheet of it. Full technique in Chapter 13.
Fiber lasers use a doped fiber optic cable as the lasing medium, producing a much shorter wavelength (around 1064nm) that metal absorbs well and wood mostly doesn't. They're the standard tool for marking bare, uncoated metal: stainless steel, aluminium, titanium, brass. Jewellery, tools, pet tags, industrial parts. They're not a wood or acrylic tool, the beam largely reflects off or passes through most organic material.
Entry-level fiber marking machines start around £1,200 to £2,500. A common growth pattern is starting with CO2 or diode, building a customer base, then adding a fiber laser specifically to serve metal engraving requests the first machine can't handle (more in Chapter 24).
| If you want to… | Go with |
|---|---|
| Do the widest range of gift, decor, and signage work on one machine | CO2, 40–60W |
| Start cheap and find out if you actually like this | Diode, 10–20W class |
| Cut clear acrylic regularly | CO2 (diode can't) |
| Mark bare metal, jewellery, tools, tumblers | Fiber (neither CO2 nor diode does this well) |
| Run a home signage or gift business from day one | CO2 50–80W, or diode first and add CO2 as revenue supports it |
| Keep the smallest possible footprint | Diode |
Buying the right machine, building a workspace that won't smoke you out or burn down, and getting the software sorted before your first real project.
This is the single most misunderstood spec in the whole hobby, worth spending real time on. When a manufacturer lists "power," ask what they're actually measuring, because at least three different things get called "power" depending on the brand.
Two diode lasers both marketed as "20W" from two different brands can produce noticeably different cutting depth and speed, because one might mean 20W optical and the other might mean 20W input power drawing maybe 5–6W optical. Look for actual optical power figures where published, read independent reviews that test cut depth on standard material rather than trusting the box copy, and treat any number without a clear "optical output" label with suspicion. CO2 machines are more consistent here, tube wattage has been a standard industry spec for decades.
Bigger isn't automatically better, it's a tradeoff. A larger bed lets you do bigger signs and batch multiple smaller items in one job (huge for efficiency once you're selling), but the machine costs more, takes up more floor space, and for CO2 in particular, larger beds generally mean you need more power to actually reach cutting depths across the full working distance.
For a business doing gifts, coasters, small signage, and personalised items, a bed in the 12×20 inch to 16×24 inch range covers the vast majority of orders. If you know you'll be doing large wall signs or furniture-scale work regularly, size up from the start, retrofitting a bigger machine in later is more expensive than buying right the first time.
Enclosed machines (Glowforge, xTool's higher-end models, most name-brand CO2 units) contain the beam entirely, often include built-in extraction or filtration, and are generally safer and quieter to run in a shared living space. Open-frame machines (most budget diode setups, many DIY-style CO2 kits) are cheaper for the same working area and power, but you're entirely responsible for beam containment, ventilation, and keeping pets, kids, or curious housemates away from an active laser.
If you're running this from a garage or dedicated workshop with the door closed and nobody wandering through, open-frame is a completely reasonable and much cheaper way to start. If you're in a shared house, apartment, or anywhere people or animals might walk near an active machine, pay the premium for an enclosure, or build one.
Brands and specific models change fast in this space, so treat the following as categories to research rather than a fixed shopping list.
| Category | Price range | What you get |
|---|---|---|
| Entry diode | Under £400 | 5–10W optical, small bed (8×8 to 16×16 in) |
| Mid-range diode | £400–£1,000 | Larger bed, genuine 10–20W optical, air assist available |
| Entry CO2 (glass tube) | £800–£2,500 | 40–60W tube, open frame or basic enclosure |
| Enclosed all-in-one CO2 | £2,500–£5,000+ | Built-in filtration, camera placement, app-based control |
| Industrial / production CO2 | £5,000+ | 80–150W+, built for continuous production use |
Ventilation is not optional. Treat this as the most important chapter in the equipment section, and the one new hobbyists most underestimate before their first cut.
Cutting and engraving produces smoke, fine particulates, and depending on the material, seriously toxic gases. This isn't a "nice to have if you're sensitive to smells" issue, it's a real air quality and fire safety issue.
Anything containing PVC (many cheap vinyl and pleather-style products) releases chlorine gas when lasered. That's properly dangerous to breathe and corrosive to your machine's metal components. Never laser PVC, vinyl of unknown composition, or anything you can't confirm the material of. If you don't know exactly what a scrap of plastic is, don't put it in the laser. A test scrap that fills the room with acrid black smoke and a chemical smell within seconds is a sign to stop immediately and ventilate the space fully.
The gold standard. A dedicated exhaust fan (an inline duct booster fan, commonly 4 or 6 inch depending on your machine's port size) pulls air from the machine's exhaust port through flexible or rigid ducting straight out through a wall, window, or vent to the outside. This actually removes smoke and fumes from your space rather than filtering and recirculating them. If you're setting up a permanent workspace, especially a garage or shed, plan for this from day one. Cutting a proper vent hole through a wall is far nicer than a duct hanging out of a cracked-open window.
A cheaper, less permanent version of the above: duct runs from the machine to a window insert or a cut piece of board sealed into an open window, with the fan pushing air outside. Works well, less tidy, and loses some efficiency compared to a wall-mounted vent, but a completely reasonable setup for renters or anyone not ready to commit to permanent ducting.
Standalone units with multi-stage filters (usually a pre-filter for particulates, then activated carbon for odours and gases) that pull air in, clean it, and recirculate it back into the room instead of venting outside. This is what enclosed all-in-one machines like Glowforge often rely on, or what people running in an apartment with no venting option use. They do help, but they aren't as thorough as venting outside, filters need regular replacement (an ongoing cost), and heavy daily cutting will still leave some smell in the room over time.
Not adequate for anything beyond the occasional very light engrave test. If this is your plan for a business doing regular cutting, expect health complaints, smoke-damaged walls and curtains, and smoke alarms going off constantly.
Leave clearance around the machine for the exhaust ducting to run without sharp bends. Sharp bends reduce airflow efficiency noticeably, and a fan straining against a kinked duct is a common, easily avoidable reason ventilation underperforms even after you've bought the right equipment.
This rule gets repeated throughout the guide because it matters more than any setting or technique. Flare-ups happen, especially on wood and with air assist off, and an unattended flare-up is how workshops burn down. Stay in the room. Every time.
Most CO2 tubes need active cooling to avoid overheating and shortening tube life. Smaller machines often use a simple water-cooling loop with a small reservoir or an external chiller unit. Check your coolant level and temperature regularly. In cold workshops during winter, plain water can freeze in an unheated space overnight, most people run a coolant mix with an antifreeze-like additive rated safe for laser chillers, or make sure the workshop stays above freezing.
You'll typically work across two types of software: something to create or edit your artwork, and something to control the laser itself.
Once you import or draw a design, LightBurn's Cuts/Layers panel lists each colour used in your file as a separate row. For each row you set the operation (cut, scan/engrave, or a few other modes), speed, power (usually as a percentage of max, sometimes with min/max power for dynamic power on curves), number of passes, and for engraving, the DPI or line interval.
A typical simple job might have a black layer set to engrave at 300 DPI, high speed, moderate power, and a red layer set to cut, slow speed, high power, one or two passes. Getting comfortable reordering these layers matters too: generally you want engraving to happen before cutting, so the piece doesn't shift or fall through the honeycomb bed before the detail work is finished.
| Layer colour | Operation | Speed | Power | Passes |
|---|---|---|---|---|
| Black | Scan (engrave) | 100% | 35% | 1 |
| Red | Line (cut) | 8% | 90% | 2 |
You don't need to design everything from scratch, and honestly most small laser businesses don't. Marketplaces selling laser-ready SVG bundles, fonts, and decor designs are a normal part of this business. Sites like Creative Fabrica, Design Bundles, Etsy design shops, and similar are where a lot of sellers source a chunk of their catalogue, especially for seasonal or decorative pieces.
Licences vary a lot between "personal use only," "small business use up to X units sold," and full commercial licences with no cap. A file being cheap or bundled in a "500 SVG pack" doesn't automatically mean you can sell unlimited physical copies made from it, some bundles are personal-use-only unless you buy a separate commercial add-on. Keep a simple record (a spreadsheet is fine) of which designs you've bought, where from, and what the licence actually permits. "I bought it in a bundle" isn't a defence, the actual licence terms are. The same applies to fonts: a free download doesn't automatically come with commercial rights to sell items using it.
Settings below are starting points, not gospel. Your machine, lens condition, and material batch will shift these. Always test first (Chapter 10).
The most common laser material by far, and thickness is the main variable that decides whether you're even able to cut through, separate from any settings tweaking.
As a rough starting reference on a well-tuned machine: a 40–60W CO2 will comfortably cut up to about 6–10mm plywood in one or two passes, and up to roughly 12–19mm hardwood in multiple passes. A 10–20W class diode is realistically limited to around 3–8mm in most woods before pass count and edge charring make it impractical for production work, not because it "can't" go thicker but because the time and edge quality stop being worth it commercially.
Genuine vegetable-tanned leather engraves beautifully, giving a rich dark brown mark without an unpleasant smell. Chrome-tanned leather (much more common in cheap goods) can release more unpleasant and potentially harmful fumes when lasered, extraction matters, and it's worth asking suppliers which tanning method they use if you're buying leather specifically for laser work. Faux/PU leather varies a lot: some are laser-safe, some contain PVC and should never go near the machine. Always check with the supplier.
CO2 lasers don't cut glass but can create a frosted, etched-looking mark on the surface, popular for wine glasses and mirrors. A wet paper towel or masking tape on the surface during engraving helps reduce chipping and improves mark consistency. Slate coasters and tiles engrave with excellent, high-contrast results and are a very popular, low-cost, low-fuss product for beginners and small shops alike.
Bare metal needs either a fiber laser (Chapter 3) or a marking spray/paste like Cermark/Thermark applied before engraving on a CO2 machine, the compound reacts with the laser heat to leave a permanent black mark bonded to the metal surface. Anodised aluminium is the exception, both CO2 and some diode lasers can mark it directly by burning away the anodised layer to reveal the metal or a colour change underneath, no coating needed. This is one of the most accessible ways to do "metal engraving" without buying a fiber laser.
Felt engraves and cuts cleanly on CO2 with almost no fraying, a popular material for patches and appliqué pieces. Cotton and other natural fabrics cut reasonably but can scorch visibly at the cut edge unless settings are dialled in carefully. Cardstock and paper cut and score extremely easily at low power, useful for invitations, gift tags, and packaging inserts.
PVC, vinyl of unknown composition, ABS plastic, any plastic you can't positively identify, polystyrene/styrofoam, fibreglass, treated or painted wood of unknown origin (old furniture, reclaimed pallets unless you know exactly what's on them), and carbon fiber (releases fine conductive dust that can damage electronics). If in doubt, don't.
Most CO2 and higher-power diode lasers used in this hobby are Class 4, the highest hazard class, capable of causing skin burns and starting fires from stray or reflected beams, not just direct eye contact. Treat the machine with that level of respect even once it feels routine. Familiarity is when accidents actually happen.
First aid basics. Keep a basic burn kit nearby. Minor burns happen even to experienced users, usually from touching a just-cut piece of material or a hot honeycomb bed, not from the beam itself.
Everything you'll actually be doing at the machine, from dialling in settings to the finishing work that makes a piece sellable.
By now the core ideas (speed, power, DPI, frequency, focus) have all been introduced. This chapter is about actually setting them in software, using LightBurn as the reference since it's what most people end up on.
Say you're making a birch plywood coaster with an engraved logo and a cut-out edge. In your design file you'd draw the logo in one colour, say black, and the cut outline in another, say red. Import into LightBurn and you'll see two rows in the Cuts/Layers panel, one per colour. You set the black layer to "Scan" (LightBurn's term for raster engraving) at something like 300 DPI, 100% speed (or close to your machine's max), and a moderate power, maybe 30 to 50% depending on your machine and how dark you want the mark. You set the red layer to "Line" (vector) mode at a much slower speed, maybe 5 to 15% of max, and higher power, often 80 to 100%, with one or two passes, until the coaster drops free.
Order matters: put the engrave layer above the cut layer in the operation order, so the design detail is engraved into the material while it's still securely attached, rather than after it's a loose cut-out that can shift mid-engrave and ruin your registration.
A common beginner mistake is treating power as the only dial and leaving speed fixed. In reality you're always balancing both, and for any given material and desired result there's usually a range of speed/power combinations that all work reasonably well, not just one exact number. Slower with less power and faster with more power can sometimes produce a similar depth of cut, but they won't necessarily look the same, the slower pass often gives a cleaner edge with less charring because the beam has more time to fully vaporise material rather than pushing through it forcefully.
Rather than cranking power to the maximum to force a cut through thick material in one go, multiple passes at a more moderate, controlled power setting almost always give a cleaner result with less charring and less risk of a flare-up. It costs you time, but time is cheaper than wasted material and scorched edges you then have to sand or explain away to a customer.
As a rule of thumb, if you're pushing past about 90% power and still not cutting through cleanly, that's usually a sign to add a pass rather than push power any higher.
Lower frequency means more energy packed into each individual pulse, which tends to give a rougher, more aggressive cut, useful when you're trying to push through thicker material. Higher frequency gives smoother, finer results and is generally what you want for detailed engraving work. Most people leave this at a sensible default and only start adjusting it once they're troubleshooting a specific problem, like rough cut edges on thick acrylic that speed and power alone aren't fixing.
If your machine has air assist, treat turning it on or off as part of your settings, not a fixed always-on accessory. For most cutting work you want it on, it clears debris and smoke from the beam path and reduces flare-ups. For some engraving work, particularly certain photo engraves on wood where you want a slightly darker, richer tone, some people deliberately turn air assist down or off, since the airflow can cool the material slightly and lighten the resulting mark. Test both ways on scrap before deciding.
Guessing settings material by material wastes far more time and stock than doing this properly once per material and thickness combination you regularly use.
Most laser software includes a built-in test grid or "material test" tool (LightBurn has one under Tools), which generates a grid of small squares, each engraved or cut at a different speed/power combination, with the values labelled directly on the piece. Run this once on a scrap piece of whatever material and thickness you're planning to use regularly, then physically inspect the results: which squares engraved with good contrast and no charring, which cut lines went cleanly through without excess burn, which combinations wasted time doing nothing useful.
Write the winning combination down immediately, next to the physical test piece if you can, memory is unreliable and "I'll remember" is how people end up re-testing the same material three separate times over a year.
A simple spreadsheet works fine: columns for material, thickness, machine (if you ever run more than one), operation (engrave/cut/score), speed, power, passes, DPI, and any notes (like "damp batch, needed extra pass" or "supplier changed, retest"). This becomes one of the most valuable documents in your business once you're taking custom orders regularly, because a new customer asking for a material you haven't used stops being a guessing exercise and becomes a five-minute lookup, or a quick fifteen-minute retest if it's actually new.
Update your library when things change: a new material batch from a different supplier, a new lens, a machine that's aged and lost some power, or a seasonal humidity shift (see Chapter 2). It's a living document, not a one-time setup task.
Cutting needs vector paths, clean, closed shapes the laser can trace as an outline. Engraving can use either, vector for crisp line art and text, raster for anything with shading, gradients, or photographic detail.
If you've got a raster image (a logo saved as a JPG, say) and need a clean cut line around it, you'll need to trace it into a vector shape first. Illustrator's Image Trace, Inkscape's Trace Bitmap, or a dedicated vectorising tool all do this, converting the pixel edges of your image into smooth vector paths. It rarely comes out perfectly clean automatically, budget time to manually clean up stray nodes and merge paths, especially on logos with fine detail or gradients that don't trace well as flat shapes.
Photo engraving on wood or similar materials generally works best converted to greyscale first, then run through a dithering process (many laser software packages have this built in, sometimes called "Jarvis," "Stucki," or similar dithering algorithms) which converts continuous tone into a pattern of dots the laser can actually reproduce, since the laser is fundamentally an on/off (or variable power) tool, not something that paints continuous grey values the way ink does. Higher DPI does help photo detail specifically, this is one of the cases where it's worth the extra engrave time, 400 DPI or higher is common for anything meant to read clearly as a photo rather than an abstract pattern.
Contrast matters more than most people expect going in. A flat, low-contrast photo will engrave as a flat, muddy-looking piece, boosting contrast and sometimes brightness in your photo editor before sending it to the laser noticeably improves the final result.
Once you're making the same design repeatedly for sale, nest multiple copies onto one sheet of material rather than running single pieces one at a time, this saves both material (better use of sheet space) and time (fewer job setups, fewer refocus checks). Keep a template file per product with the cut and engrave layers already colour-coded and settings pre-assigned, so a repeat order is "open file, hit start" rather than rebuilding the job from scratch each time.
Standard raster engraving fills an area line by line at a fixed or image-driven power level. For bold, solid-colour-look fills (common on painted or coated materials, or where you want maximum contrast on wood), running two passes, sometimes rotated 90 degrees between them, can produce a cleaner, more even fill than a single pass, particularly at lower DPI where a single pass can leave faint visible banding.
Some software supports greyscale-mapped depth engraving, where lighter greys in your source image engrave shallower and darker greys engrave deeper, building up a genuine 3D relief effect rather than a flat mark. This works best on CO2 with decent power headroom, since you need enough depth range to actually show a visible relief, and it's slow, often the slowest kind of job you'll run, but the results on things like portrait plaques or topographic maps can be seriously impressive and justify a premium price.
A rotary attachment replaces (or works alongside) the flat bed with a set of rollers that spin a cylindrical object, tumblers, glasses, bottles, so the laser can engrave around a curved surface as if it were flat. This opens up an entire product category, personalised tumblers and glassware are one of the most consistently popular laser business products, and it's worth budgeting for a rotary attachment fairly early if that's a product line you want to offer, rather than treating it as a much later upgrade. Software handles the geometry conversion but you'll want to test carefully on scrap cylindrical stock, cheap tumblers of the exact size and coating you plan to sell, before committing to a batch, since coating thickness and material vary between suppliers even for visually similar blanks.
Because this trips people up early on and it's worth reinforcing separately from the settings chapter: a cut is meant to separate material completely, a score is a deliberately shallow line that does not go through. You'd use scoring for fold lines on cardboard packaging, for fine internal detail lines within a larger cut piece that you don't want to fall apart, or for light registration marks. In LightBurn, this is purely a settings difference on a vector layer, faster speed and lower power for a score, slower and higher for a cut, there's no separate "scoring mode" toggle, it's the same Line/vector operation type just tuned weaker.
Some products combine layers, a walnut base with an acrylic overlay, for example, or a two-tone wood inlay where a cut piece from one wood is glued into a matching cut recess in another. This requires cutting each layer as a separate job, on its own material, then assembly and gluing afterward. Registration between layers matters here, using small alignment marks cut into a corner of each piece, then trimmed off after assembly, is a simple, reliable way to keep layered pieces aligned.
Wood and acrylic both leave a layer of smoke residue and light surface soot around engraved and cut areas. On wood, a damp cloth or masking tape applied before engraving (removed afterward) reduces residue significantly, or you can wipe it off after with a barely damp cloth, being careful not to soak bare wood. Isopropyl alcohol on a cloth works well for stubborn residue on sealed or painted surfaces, and for acrylic, which often shows residue as a light haze around cut edges. Test any cleaning method on a scrap piece first, some finishes and dyed materials can be affected by alcohol.
Applying a layer of transfer tape, painter's tape, or purpose-made laser masking film to your material before engraving protects the surface from smoke residue and, on some materials, gives a cleaner overall finish once removed. It's standard practice on acrylic and increasingly common on wood pieces meant for direct sale, since a customer opening a box to a clean piece rather than one with visible smoke haze around the design makes a real difference to perceived quality, even though the underlying engrave is identical either way.
This is an underused technique worth knowing well, especially relevant if you're on a diode machine and can't cut clear acrylic for coloured accents.
Acrylic paint pens (Posca and Molotow are the two most commonly used brands) let you fill engraved text or line detail on wood with colour after engraving, wiping away the excess from the surrounding raised surface before it fully dries, leaving colour sitting only in the recessed engraved lines. This gives you the two-tone, high-contrast look people associate with layered acrylic accents, without needing to cut acrylic at all. It works especially well on darker woods like walnut, where a metallic gold or white paint pen fill makes engraved text pop dramatically. It's slower per piece than a straight engrave, factor that into your pricing (Chapter 19), but it noticeably raises the perceived value of a piece for a relatively small material cost.
Raw engraved wood will pick up dirt, moisture, and general wear over time, especially on anything handled often like cutting boards or coasters. A food-safe mineral oil or a beeswax/mineral oil blend is standard for cutting boards and anything that'll contact food. A clear polyurethane, lacquer, or wax finish works well for decorative pieces not intended for food contact, and also helps protect and slightly darken or enrich an engraved design. Whatever you choose, be consistent and mention it in your product listings, "food-safe finish" and "sealed for durability" are both things customers specifically look for and will ask about if you don't mention it upfront.
A light sand with fine grit sandpaper (320 to 600) after engraving removes any raised, slightly rough char texture left on wood, giving a smoother finish before oiling or sealing. Don't sand so aggressively that you remove the darker engraved contrast itself, a light pass is usually enough, and always sand with the grain.
What to clean, how often, and how to work out what's actually wrong when a job doesn't come out right.
Treat maintenance as a scheduled task, not a "when something goes wrong" reaction. A machine that's cleaned and checked regularly holds its settings library accuracy far longer, breaks down less, and simply produces better, more consistent results.
The focusing lens sits directly in the path of every bit of smoke and debris your material produces, and it gets dirty faster than most beginners expect, sometimes within a single long engraving session on a smoky material like MDF. A dirty lens doesn't just look bad, it actively absorbs some of the laser energy meant to be focused on your material, converting it into heat in the lens itself. That means less power reaching your work (so you compensate by cranking power up, masking the real problem).
In bad cases, heat buildup in a dirty lens is enough to crack or permanently damage it. Check the lens before every significant job, not just when something seems off. Remove it following your machine's specific instructions, inspect it against a light for haze, spotting, or visible residue, and clean gently with a lens-safe solution and a lens cloth or cotton swab, wiping in one direction rather than scrubbing. Never use regular glass cleaner or a rough cloth, both can scratch the coating. If you're engraving smoky materials daily, a quick check and clean every few hours of runtime is normal, not excessive.
CO2 machines with external mirrors (as opposed to fully sealed optical paths) typically have two or three mirrors directing the beam from the tube to the head. Same principle as the lens: smoke residue builds up and degrades beam quality. Clean on the same rough schedule as the lens, using the same gentle lens-safe method. A quick way to tell your mirrors or lens need attention: if cut quality has dropped or you're needing more passes than your settings library says you should, at settings that used to work fine, check optics before assuming the tube is dying.
On open-frame CO2 machines, the beam path runs through multiple mirrors before reaching the head, and if a mirror has shifted even slightly (from vibration, a bump, or just time), the beam can end up misaligned, meaning it's not hitting the centre of the lens correctly at every position across the bed. Symptoms include inconsistent power or cut quality that changes depending on where on the bed you're cutting, sometimes strong in one corner and weak in another. Most machines have a documented alignment procedure involving firing a test pulse onto masking tape at each mirror stage and adjusting small set screws until the burn mark lands centred. It's fiddly the first time and oddly satisfying once you've done it a couple of times. Fully sealed-optic machines (Glowforge and similar) don't have this issue at all, one of the real tradeoffs for paying more and getting less user access.
Check belt tension periodically, a loose belt causes visible ghosting or slight misregistration in engraves, especially noticeable as a faint double-edge on text or fine detail. Keep rails clean and lightly lubricated per your manufacturer's specific recommendation, dust and debris buildup on rails causes rougher movement over time and, eventually, positioning inaccuracy. Periodically check that the gantry is still square to the bed, an out-of-square gantry shows up as designs that engrave slightly skewed even though your file is straight.
The honeycomb cutting bed collects burnt residue over time, which can transfer onto the back of lighter materials and cause backside marking you don't want. Clean it periodically, most can be soaked or scrubbed, and replace it outright once it's heavily charred and no longer supporting material flat and evenly.
If your machine has an air assist pump, check the tubing for kinks or blockages periodically and check the pump's own filter/intake if it has one, a partially blocked air assist line reduces airflow at the nozzle without being obviously broken, and you'll just notice worse cut quality and more charring without an obvious cause until you actually check it.
CO2 tubes lose output power gradually over their lifespan, well before they fail outright, which is why a settings library needs occasional retesting as a machine ages, a setting that cut cleanly through 6mm plywood a year ago might need an extra pass now, and that's expected, not a sign of a fault necessarily. Diode modules generally degrade less dramatically but do lose some output over thousands of hours of use, and are also vulnerable to damage from running dirty or from physical shock.
| Frequency | What to do |
|---|---|
| Every session / few hours of heavy use | Check lens for visible residue, wipe if needed |
| Weekly (regular use) | Clean lens and mirrors properly, check air assist tubing, clear bed debris |
| Monthly | Check belt tension and rail cleanliness, inspect honeycomb bed, check chiller coolant (CO2) |
| Every few months / after a bump | Check mirror alignment and gantry squareness |
| Problem | Likely cause | Check first |
|---|---|---|
| Cut doesn't go through, used to work fine | Power dropped, dirty optics, material moisture, focus drifted | Clean lens/mirrors, refocus, add a pass before blaming the tube |
| Engrave looks fuzzy or blurry | Focus is off | Re-check focus height for the material thickness |
| Uneven quality across the bed | Mirror misalignment (CO2) or gantry not square | Run alignment procedure, check gantry squareness |
| Excess charring around edges | Power too high for speed, air assist off/blocked, too many passes | Lower power or raise speed, check air assist, try fewer passes |
| Faint double-edge/ghosting on text | Loose belt | Check and re-tension belts |
| Smoke marks on back of material | Dirty honeycomb bed | Clean or replace honeycomb insert |
| Flare-up or small fire mid-cut | Air assist off, flammable material, power too high | Never leave unattended, keep air assist on, extinguisher within reach |
| Diode won't cut clear acrylic | Not a settings problem | Known diode limitation, beam passes through (see Chapter 3) |
| Cuts in the wrong place | Job origin mismatch or material shifted | Check job origin settings, secure material properly |
| Strong chemical smell, heavy dark smoke | Wrong material, likely PVC/unknown plastic | Stop immediately, ventilate fully, identify the material first |
| Software won't connect to machine | Driver, USB, or Wi-Fi issue, not hardware | Check cable/connection, restart both, check manufacturer's guide |
Legal setup, pricing, sourcing, selling, marketing, fulfilment, and growing past a one-person operation.
Yes, with real caveats.
Laser engraving is one of the more accessible small manufacturing businesses to start: low equipment cost relative to output value, strong demand for personalised goods, and a wide range of price points from £8 keychains to £250+ custom furniture pieces. It's also become a lot more competitive over the last several years precisely because it's so accessible, plenty of people have bought the same machine you're considering.
That means the businesses that do well tend to differentiate on something beyond "I have a laser": design quality, a specific niche, solid finishing and presentation, reliable turnaround, or strong local relationships (wedding vendors, corporate gifting, local shops wanting signage). "I'll engrave anything for anyone" is a much harder business to build a following around than a shop known specifically for, say, walnut anniversary gifts or pet memorial pieces.
Material costs are usually low relative to selling price for most laser goods, which is exactly why competition on price alone is a race to the bottom. Time is usually the real cost that new sellers underprice: design time, test time, actual machine run time, finishing, packaging, and customer communication for custom orders all add up to far more than the twenty minutes the laser itself was actually running. Chapter 19 covers pricing properly, but the short version going in: if you're only accounting for material cost plus a bit extra, you will underprice yourself and burn out.
You don't need a fifty-page formal document, but write down, even just for yourself: what you're actually selling and to whom, how you'll reach those customers, what it costs to make each item versus what you'll charge, and what your first three to six months of realistic sales might look like. This isn't bureaucracy for its own sake, it's the thing that stops you from discovering six months in that your pricing never actually covered your time.
Rough UK ranges in GBP, adjust for what you already own.
| Item | Budget-conscious | More established |
|---|---|---|
| Machine | £350 (diode) | £2,000–£5,000 (CO2) |
| Ventilation setup | £80–£160 | £250–£500 |
| Software licence (LightBurn) | ~£50–£100 one-time | Same |
| Safety gear | £40–£80 | £40–£80 |
| Initial material stock | £120–£250 | £400–£800 |
| Design assets/fonts (licences) | £40–£120 | £150–£400 |
| Packaging and shipping supplies | £80–£160 | £250–£500 |
| Photography setup | £40–£80 | £150–£400 |
| Business registration | £40–£120 | £120–£400 |
| Rough total to launch | ~£800–£1,200 | ~£3,200–£7,200 |
You do not need to spend at the top of this range to start, plenty of successful small laser businesses began with an entry diode laser and grew the equipment budget from actual sales rather than upfront investment. What you shouldn't skip regardless of budget tier: proper ventilation, safety gear, and correctly licensed design assets, all three are corners that cause real problems (health, fire, legal) if cut.
Most people starting this kind of business self-fund from savings or by starting really small (a diode machine, minimal initial stock) and reinvesting early sales into better equipment and materials. Given the relatively low entry cost compared to most physical product businesses, taking on debt to start isn't usually necessary or advisable unless you're going straight for a production-scale CO2 setup before you've validated demand.
Most small laser businesses in the UK start out as a sole trader, simply because it's the default, no-paperwork option when you start selling, you just need to register with HMRC once you're actually trading. As revenue grows, forming a limited company is worth considering for the personal liability protection it provides, separating your personal assets from business liabilities, especially relevant here because you're running equipment with genuine fire risk and selling physical products that could theoretically cause harm (a poorly finished item, an allergic reaction to a finish, splinters). Talk to an accountant once you're doing consistent revenue, the sole trader vs. limited company decision affects tax quite differently depending on how much you're actually earning.
Beyond that, it's mostly paperwork you can look up as it becomes relevant rather than something to front-load: keep an eye on the VAT threshold as sales grow, and if you're running the laser from a garage or shed, a quick check with your local council on planning and business rates is sensible before committing to a permanent setup. None of it is difficult, it's just easy to put off. Do it early rather than retroactively once you're already invoicing customers.
A standard home or contents insurance policy typically does not cover business equipment or business liability, running a laser (real fire risk) as an uninsured home business is a genuine exposure, and could also void your existing home policy if you haven't declared the business activity. Public liability insurance, and separately business contents/equipment insurance covering the machine itself, are both worth pricing out once you're operating regularly, costs are often lower than people expect for a home-based small business policy. If you sell at in-person markets or craft fairs, many venues now require proof of public liability insurance before you can even book a table, so this can become a practical requirement, not just a "nice to have," fairly early.
Be thoughtful about what you sell and how you describe it, especially anything food-adjacent (cutting boards, coasters) or for children. Use food-safe finishes and say so explicitly, avoid vague claims you can't back up, and keep records of what finishes and materials you used on what batches, useful both for your own settings library and, in the unlikely event of a complaint, to show you used a reasonable, documented process.
Pricing based on material cost plus a bit extra, then wondering why the business doesn't feel worth the time, is the mistake almost everyone makes early.
Material is usually the smallest real cost in a laser item. A proper price accounts for:
This is the reason so many handmade sellers quietly earn well under minimum wage for their time without realising it until they actually do the maths. Your labour is a real cost, price it like one.
(Material + Machine time + Labour time + Packaging) × markup multiplier = price
A common starting markup multiplier for handmade goods is somewhere between 2× and 3× your all-in cost, adjust based on your niche, local market, and how differentiated your work is. Highly custom, well-finished pieces in a niche with less price competition can support a higher multiplier than a generic keychain in a crowded category.
Standard catalogue items (the same coaster design, made repeatedly) can be priced tightly since your time-per-unit is well known and low once the file and settings are dialled in. Custom or one-off orders (a specific name, a unique photo engraving, a bespoke design request) should carry a meaningful premium over your standard items, because the design and communication time is real and often underestimated, a "quick custom job" regularly takes three or four times longer than the standard version once you count messaging back and forth with the customer, proofing, and revisions.
Because material costs are so low across this whole hobby, it's tempting to win orders by simply being the cheapest listing. This is a race you can win short-term and lose long-term, since there's almost always someone willing to charge even less, often someone who hasn't yet worked out they're underpricing their own time. Compete instead on finish quality, presentation, reliability, niche focus, and customer experience, all things a race-to-the-bottom competitor usually can't or won't match.
Once a product line is proven to sell consistently, buying its core material in bulk lowers per-unit cost meaningfully, but ties up cash in inventory and requires storage space (and for wood, storage in stable humidity conditions, since moisture content affects your settings, as covered in Chapter 2). For new or untested product ideas, buy small quantities first, prove the design sells, then scale up material purchasing once demand is confirmed rather than betting on an unproven idea.
A supplier that's slightly more expensive but consistent in thickness, finish, and material composition batch to batch is often worth it over a cheaper supplier with variable quality, because inconsistent material means your carefully tested settings stop being reliable, and that shows up as wasted material and inconsistent finished product, both of which cost more than the price difference saved on the material itself.
Still one of the most common starting points, built-in buyer traffic looking specifically for personalised and handmade goods, relatively low barrier to listing. Downsides: real fees (listing fees, transaction fees, payment processing), significant competition within laser-engraved product categories specifically, and you don't own the customer relationship or platform, policy and algorithm changes are outside your control.
More setup work and you have to drive your own traffic, but you keep more margin, own the customer data and relationship, and have full control over branding and presentation. Often works best as a second channel once you already have some following from Etsy, social media, or local sales, rather than as your only channel from day one when you have no existing audience to draw on.
Genuinely valuable for a laser business specifically because people can see and touch the quality in person, something photos alone don't always convey well. Also a strong source of direct customer feedback and word-of-mouth referrals. Costs a table fee and your time, and as mentioned in Chapter 18, increasingly requires proof of liability insurance to book a table at established markets.
Restaurants wanting branded coasters or menus, local shops wanting signage, wedding venues wanting a preferred vendor relationship, corporate clients wanting awards or branded gifts. Often lower volume per client than online retail but higher order value and more repeat business once a relationship is established, and referrals from one satisfied local business client can be a real driver of growth.
Instagram and TikTok specifically have become direct sales channels for handmade goods, not just awareness tools, through built-in shopping features. Short process videos, watching a design being cut and revealed, perform particularly well for laser work specifically, it's inherently satisfying, visual content, and this genre has driven real sales for a lot of small laser businesses.
Once established, selling in bulk at a wholesale price to boutiques, gift shops, or other retailers can provide more predictable, larger-volume income than one-off retail sales, at a lower per-unit margin. Worth considering once you have a proven, repeatable product and the production capacity to fulfil larger orders reliably.
For a visual, gift-oriented product category, product photography quality has an outsized effect on sales compared to almost any other marketing effort. Natural light near a window, a simple uncluttered background, and a few different angles including a close-up on the actual engraved detail will outperform a cluttered, dim, phone-flash photo every time, and this is achievable without expensive equipment starting out. Show scale (a hand holding the item, or a common object beside it) since online shoppers can't otherwise judge size accurately, and this is one of the most common causes of return requests and complaints when skipped.
A consistent name, logo, and colour scheme across your packaging, social media, and listings makes a small operation look established and trustworthy, which matters a lot for a category (personalised gifts, often for significant occasions) where customers are trusting you with something they care about. This doesn't need to be expensive, a clean, simple, consistently applied logo beats an elaborate one used inconsistently.
If local sales matter to your business (signage, B2B, wedding vendor work), a Google Business Profile, consistent local keywords on your website, and simply being easy to find when someone searches "laser engraving [your town]" matters more than people expect. Word of mouth remains powerful in this category specifically, personalised gifts are the kind of purchase people talk about and recommend, make it easy for happy customers to tag you or share where they bought something, a small branded card or sticker included with orders asking customers to share a photo is a low-cost, effective nudge.
Process videos (the reveal moment when a design comes off the machine is compelling content on its own), behind-the-scenes workshop content, and customer testimonial/gift-reaction content (with permission) all tend to outperform generic product photos alone on social platforms. Consistency matters more than perfection, posting simple, real content regularly beats occasional highly polished posts.
A few pieces of equipment beyond the laser itself become essential quickly once you're actually shipping orders, and they're worth budgeting for from the start rather than discovering the gap mid-order.
Once you're shipping more than a handful of orders a week, printing shipping labels on a home inkjet and cutting them out by hand becomes a genuine time sink and looks unprofessional taped crookedly to a box. A dedicated thermal label printer (Rollo and Dymo are two commonly used brands) prints labels quickly with no ink cost, since thermal printers use heat-sensitive paper rather than cartridges, which also makes them cheaper to run long-term than they first appear. Pair it with a proper parcel scale, guessing weight leads to either overcharging customers or undercharging and eating the difference yourself, repeatedly.
Buy packaging supplies in a range of sizes, not just one box size for everything, oversized boxes waste money on shipping cost and increase the chance of an item shifting and getting damaged in transit.
This is easy to overlook until you've already designed a product line and then discover shipping it well is awkward. Before finalising a product's dimensions, think about what standard box or mailer size it'll actually ship in, a coaster set designed to fit neatly into a standard small box costs much less to ship and looks better packed than one that's an inch too big for every standard size and needs a custom oversized box for every single order.
This matters even more for anything rotary-engraved like tumblers, which need specific protective packaging (often a dedicated tumbler shipping box with built-in padding) that you should source and test before you start selling the product, not after your first breakage complaint. Flat items (signs, plaques) ship far more cheaply and safely than anything with awkward protruding parts, if a product design can be adjusted slightly to pack flatter without compromising the design, that's often worth doing purely for the shipping and damage-rate benefit.
Even a simple spreadsheet tracking order date, customer details, product, custom text/details, status (in production, ready to ship, shipped), and any special notes prevents the surprisingly common and embarrassing mistake of a missed custom detail or a forgotten order once you're handling more than a handful of orders a week. As volume grows, dedicated order management or shop platform tools become worth the cost.
Be honest and slightly conservative in the turnaround time you promise, especially for custom orders where design, proofing, and production all take real time. It's far better to promise five to seven business days and deliver in four than to promise two days and consistently deliver in five, the second version generates complaints and bad reviews even though the actual production time might be identical. Build in buffer for busy seasons (wedding season, the run-up to major gifting holidays) when order volume and turnaround pressure both spike at once.
Do a final check on every item before packaging, checking for consistent engrave depth, no unexpected charring, correct spelling on any custom text (a common and entirely preventable source of costly redos), and a clean finish. A simple habit of reading custom text back once more against the original order before starting the job, not just after, catches a lot of the most expensive mistakes before any material or time is wasted on them.
The most common first scaling move is adding a second machine, often a CO2 machine if you started on diode, or a second unit of the same type to run jobs in parallel and increase throughput without extending your own working hours proportionally. This becomes worthwhile once order volume regularly means your single machine is the bottleneck, not your own available time.
As covered in Chapter 3, a common growth path is adding a fiber laser once you've built a customer base and started getting regular requests for metal engraving (jewellery, tools, metal tumblers, pet tags) that your CO2 or diode setup fundamentally can't do well. Worth tracking how often you're turning down or referring out metal requests, if it's frequent enough, it's a strong signal the investment will pay for itself reasonably quickly.
Some businesses add a UV printer alongside laser engraving to offer full-colour printed designs on similar blanks (tumblers, ornaments, phone cases), a properly complementary technology that opens up colour work a laser fundamentally can't do, since a laser only marks, it doesn't add colour of its own (aside from the paint-fill technique in Chapter 13). This is a bigger investment and a real second skill set to learn, worth considering once laser engraving alone has proven the business model.
Once you're consistently turning away orders or working unsustainable hours, hiring help, even a few hours a week for packaging, shipping, or basic production on proven, repeatable designs, frees you up for the design, sales, and customer relationship work that actually grows the business, which is much harder to hand off than production tasks are.
There's a real tension between adding new product lines and equipment (more revenue potential, more complexity) and staying tightly focused on what you already do well (easier to market, easier to be known for something specific, lower operational complexity). Growing what's already working tends to be a safer bet than adding a new equipment category purely because it seems interesting, unless there's a clear, proven demand signal (like the fiber laser example above) driving the decision.
The parts you'll actually flip back to: settings, terms, and checklists, in one place.
Starting points only, always test on scrap first (Chapter 10). Your actual machine, lens condition, and material batch will shift these.
| Material | Thickness | Operation | Speed | Power | Passes |
|---|---|---|---|---|---|
| Birch plywood | 3mm | Cut | Slow | 80–100% | 1 |
| Birch plywood | 6mm | Cut | Slow | 90–100% | 1–2 |
| Birch plywood | any | Engrave | Fast | 30–50% | 1 |
| Cast acrylic (clear) | 3mm | Cut | Slow–Med | 80–100% | 1 |
| Cast acrylic | any | Engrave (frosted) | Fast | 20–40% | 1 |
| Leather (veg-tanned) | 2–3mm | Engrave | Med–Fast | 30–45% | 1 |
| Slate | tile/coaster | Engrave | Fast | 50–70% | 1 |
| Cardstock | standard | Cut/Score | Fast | 10–25% | 1 |
| Anodised aluminium | sheet | Mark | Fast | 15–30% | 1 |
| MDF | 3mm | Cut | Slow | 85–100% | 1–2 |
| Material | Thickness | Operation | Speed | Power | Passes |
|---|---|---|---|---|---|
| Birch plywood | 3mm | Cut | Slow | 90–100% | 2–4 |
| Birch plywood | 6mm | Cut | V. slow | 100% | 4–8 (weigh the time cost) |
| Birch plywood | any | Engrave | Medium | 40–60% | 1 |
| Basswood | 3mm | Cut | Slow | 80–100% | 1–3 |
| Leather (veg-tanned) | 2–3mm | Engrave | Medium | 35–55% | 1 |
| Slate | tile/coaster | Engrave | Medium | 50–70% | 1 |
| Cardstock | standard | Cut/Score | Fast | 10–20% | 1 |
| Anodised aluminium | sheet | Mark | Medium | 20–40% | 1 |
| Clear acrylic | any | Not achievable, beam passes through (see Chapter 3) | |||
Air assist: a jet of air aimed at the laser's focal point during cutting, clearing smoke and debris and reducing charring and flare-ups.
DPI / LPI: dots per inch / lines per inch, controls raster engraving resolution and detail level.
Dithering: converting a greyscale image into a pattern of dots so a laser can approximate continuous tonal shading.
Focal point: the exact distance from the lens at which the beam is at its smallest, most concentrated, and most powerful.
Frequency (CO2): pulses per second the tube fires at, affects cut roughness/fineness.
Kerf: the width of material actually removed by the beam during a cut, relevant when designing parts meant to fit together precisely.
Passes: how many times the laser repeats the same path over a job, used instead of maximum power to cut thicker material more cleanly.
Raster: engraving mode where the head sweeps line by line, used for shading, fills, and photographs.
Rotary attachment: a set of rollers that spins a cylindrical object so the laser can engrave around its curved surface.
Score: a shallow vector pass not intended to cut all the way through.
Vector: engraving or cutting mode where the head follows the outline of a shape continuously, used for cutting and clean line art.
None of this needs to be figured out in advance. Buy the machine, ruin some plywood, build the settings library one test grid at a time, and let the business grow out of the craft the way it usually does for everyone who sticks with it long enough to get good.
Good luck.