The short answer
- LED exposes single-part emulsion in around 8–15 seconds, dual-cure in 40–55
- Metal halide runs 30 seconds to two minutes, and needs warm-up time
- LED draws about 100 watts; a metal halide unit around 10,000
- LEDs last 10,000–50,000 hours; halide lamps are a consumable
- Metal halide with an integrator still gives the deepest, most forgiving cure
Exposure is the step that decides whether a screen holds detail, washes out cleanly and survives a run. It happens before any ink is mixed, it takes under a minute, and it is responsible for a surprising share of the problems blamed on artwork, mesh or the press.
The choice of light source shapes all of it, and the two technologies differ more than the marketing suggests in both directions.
What is exposure actually doing?
Emulsion coated on the mesh hardens where UV light reaches it. The film positive blocks light where the design is, so that area stays soft and washes out, leaving open mesh for ink to pass through. Everything else hardens into a stencil.
Two failures follow from getting it wrong. Underexposure leaves emulsion soft below the surface, so it breaks down during the run and detail washes away too readily. Overexposure hardens emulsion under the film as well, so fine detail never opens and small text fills in.
The window between those is the latitude, and a light source with good latitude is more forgiving of the inevitable variation in coating thickness, humidity and mesh count.
How much faster is LED?
Substantially. Typical LED exposure runs around 8 to 15 seconds for single-part emulsions, and 40 to 55 seconds for dual-cure or diazo. Metal halide commonly runs 30 seconds to two minutes with the right emulsion, and most SBQ emulsions under either source want 30–90 seconds.
| LED | Metal halide | |
|---|---|---|
| Single-part emulsion | 8–15 sec | 30 sec – 2 min |
| Dual-cure / diazo | 40–55 sec | Longer again |
| Warm-up | None | A few minutes |
| Power draw | ~100 W | ~10,000 W |
| Source life | 10,000–50,000 hr | Lamp is a consumable |
The warm-up row matters as much as the exposure time in a shop making screens through the day. A metal halide unit needs a few minutes before it is usable; an LED is ready instantly, so a single screen at four in the afternoon is a fifteen-second job rather than a five-minute one.
A hundred watts against ten thousand, for the same job. That is the headline, and it is not the whole story.
Power, lamp life and running cost
The difference is stark. Most UV LED units draw around 100 watts against roughly 10,000 for a high-efficiency metal halide — a factor of a hundred, before accounting for the halide running during warm-up and between screens.
LEDs also have a working life of 10,000 to 50,000 hours depending on the source, which for most shops means the unit is unlikely to need a new emitter in its practical lifetime. Metal halide lamps degrade with use, shift in output as they age, and are a recurring cost — and an ageing lamp changes your exposure times without announcing it, which is a subtle source of inconsistency.
That degradation is worth planning for. If you run halide, log lamp hours and retest exposure periodically, or you will slowly drift into underexposure and blame the emulsion.
Where does metal halide still win?
Quality of cure, and it is a real advantage rather than nostalgia. It is well recognised in the trade that a metal halide lamp with a built-in integrator gives the best cure for screen printing emulsions — the spectrum overlaps well with emulsion sensitivity and peaks at 405nm, which penetrates deeply and delivers optimum cure and latitude.
Depth of cure is what makes a stencil durable through a long run, and latitude is what makes the process forgiving. A shop running high-detail work, halftones or very long production runs benefits from both in ways a stopwatch does not capture.
The integrator matters too. It measures light output rather than time, so a lamp that has aged still delivers the same total energy — which is precisely the drift problem that otherwise makes halide inconsistent.
The emulsion has to match the light
This is the part that catches shops moving to LED, and it is the most common reason a new unit disappoints. LED output is narrow-band, so full exposure in seconds only happens when the emulsion is sensitised to the range your LED emits.
Pair an LED unit with an emulsion outside its range and you get long, unpredictable exposures and poor cure — the same symptoms as a fault, on new equipment. Check the emulsion is specified for LED before blaming the unit, and buy them as a system rather than separately.
Establish exposure properly, once per combination
- Run a step wedge on your actual mesh, emulsion and coating method
- Do it again for each mesh count you stock — finer mesh exposes differently
- Repeat after any emulsion change, including a new batch
- Record the result where the whole team can see it
- On halide, re-test as lamp hours accumulate
Which should you buy?
| Situation | Choose |
|---|---|
| Many screens a day, mostly spot colour | LED — speed and no warm-up dominate |
| High detail, halftones, long runs | Metal halide with an integrator |
| Electricity is a real cost | LED, decisively |
| Occasional screens, small shop | LED — instant readiness matters most |
| Existing halide performing well | Keep it; log lamp hours and retest |
For most decorating shops the answer is LED, and the reason is not the energy saving — it is that exposure stops being a scheduled activity. When a screen takes fifteen seconds with no warm-up, screens get made when they are needed rather than batched, which removes a genuine constraint from the day.
Whichever you choose, cost it the same way as any other machine: fixed monthly cost divided by realistic volume, plus the variable per screen including emulsion and, for halide, lamp replacement. The method is in calculating cost per print before you buy, and if you are looking at a used unit, a halide lamp near the end of its life is a cost the seller is unlikely to volunteer. The rest of the equipment archive assumes you match the emulsion to the light.



