Quick answer: Use CTO gels to warm light toward tungsten, CTB gels to cool it toward daylight, and ND gels to reduce brightness without changing colour. Match the filter to its published starting and finishing colour temperatures, not just its name. LEE's Full C.T. Orange, Full C.T. Blue and ND filter data describe these different jobs.
This guide covers choosing a correction, allowing for lost light and interpreting heat figures. Use the LEDask lighting calculator hub for broader lighting calculations; the examples below focus on what happens after adding a filter.
What lighting gels actually do
Lighting gels change the light passing through them. According to Wikipedia's colour gel overview, they are transparent coloured sheets used in theatre, events, photography and filmmaking, commonly made from thin polycarbonate, polyester or other heat-resistant plastics and placed in a fixture's beam.
Start with the result you need: an artistic colour, a warmer or cooler white, or less brightness. Colour temperature describes the appearance of light in kelvins; it is separate from the physical temperature of the filter. Choosing an orange correction is therefore a colour decision, while choosing a suitable filter material is a separate heat decision.
Which gel should you choose?
Choose the correction direction first, then the manufacturer's specific strength. These published examples show why a name such as “full CTO” does not guarantee the same endpoint across brands.
| Your goal | Filter example | Published change |
|---|---|---|
| Cool tungsten toward daylight | LEE 201 Full C.T. Blue | 3,200 K to 5,700 K, per LEE |
| Warm daylight toward tungsten | LEE 204 Full C.T. Orange | According to LEE, 6,500 K to 3,200 K |
| Make a smaller cooling correction | Rosco half CTB | According to Rosco's filter guide, 3,200 K to 4,100 K |
| Warm Rosco's daylight reference toward tungsten | RoscoSun three-quarter CTO | According to Rosco's filter guide, 5,500 K to 3,200 K |
| Make a stronger warming correction from that reference | RoscoSun Full CTO | According to Rosco's filter guide, 5,500 K to 2,900 K |
The starting temperature matters as much as the destination. A practical selection sequence is:
- Identify the source's colour temperature and the colour you want to match.
- Choose the correction direction from the table.
- Compare the specific filter's published conversion with that starting point.
- Budget for transmission loss before choosing how much source output you need.
Do not treat a fractional name as a promise to remove that fraction of the kelvin difference. Use the listed conversion or mired shift to compare corrections.
How much light do CTO and CTB gels absorb?
Full blue can cost substantially more light than full orange in the published examples here. LEE 201 lists 35% transmission in its tungsten data row. According to LEE 204's data, full orange transmits 62.8% in that same row category. Those figures describe the listed reference conditions; they are not universal transmission values for every lamp.
Worked example: assumed output and power
Assumed inputs: a source producing 1,000 lumens while drawing 10 W, with transmission matching the cited tungsten-row figures. Assume power stays at 10 W after adding either gel. These are planning assumptions, not a measurement of an LED fixture.
Filtered lumens = unfiltered lumens × transmission percentage ÷ 100.
- Blue: 1,000 × 35 ÷ 100 = 35,000 ÷ 100 = 350 lumens.
- Orange: 1,000 × 62.8 ÷ 100 = 62,800 ÷ 100 = 628 lumens.
- Blue light loss: 100% − 35% = 65%.
- Orange light loss: 100% − 62.8% = 37.2%.
The luminous efficacy calculator uses efficacy = lumens ÷ watts. With these assumed inputs, unfiltered efficacy is 1,000 ÷ 10 = 100 lm/W; after blue it is 350 ÷ 10 = 35 lm/W; after orange it is 628 ÷ 10 = 62.8 lm/W.
Enter the filtered output to estimate the efficacy of the source-and-filter combination. This calculation budgets brightness; it does not predict whether a particular LED spectrum will produce the desired colour correction.
What does a mired shift mean?
A mired shift expresses correction using reciprocal colour temperature. The relationship is mired = 1,000,000 ÷ kelvins, as described in Wikipedia's mired explanation. Use the destination value minus the starting value: warming produces a positive shift and cooling a negative one.
Worked example: assumed tungsten-to-daylight correction
Assumed inputs: a starting temperature of 3,200 K and a target of 5,700 K, chosen to match LEE 201's published conversion.
- Starting value = 1,000,000 ÷ 3,200 = 312.5 mired.
- Target value = 1,000,000 ÷ 5,700 ≈ 175.4386 mired.
- Shift = target − start ≈ 175.4386 − 312.5 = −137.0614 mired.
- Rounded to a whole number, the result is −137 mired, matching LEE's listed shift.
This gives a numerical way to compare correction strength instead of assuming similarly named filters are interchangeable.
When should you use neutral-density gels?
Use ND when you want less light while keeping the colour unchanged. LEE's 209 0.3ND is specified as a one-stop reduction, with 50% transmission in its tungsten row. Its 210 0.6ND is specified as two stops, and its 211 0.9ND as three stops.
Keep the purpose clear when choosing a sheet: ND is for reducing brightness; the CTO and CTB examples above are for changing colour temperature. Their loss of light is part of the budget, but it does not make them substitutes for neutral density.
Will lighting gels melt?
Gels can deteriorate under absorbed energy. Wikipedia describes fading and possible melting; Rosco's filter guide explains that most blocked radiant energy is absorbed as heat.
According to Rosco's Roscolux material specifications, its polyester and body-coloured polycarbonate filters have different thresholds:
| Roscolux material | Softening temperature | Melting temperature |
|---|---|---|
| Polyester | 257°F / 125°C | 572°F / 300°C |
| Body-coloured polycarbonate | 320°F / 160°C | 425°F / 220°C |
These are Rosco's material figures, not approved operating temperatures for every gel installation. Polycarbonate has the higher softening threshold in this comparison, while polyester has the higher melting threshold. Calling either material simply “more heat resistant” hides that distinction. Use the fixture and filter manufacturer's mounting guidance rather than treating a melting point as a safe working limit.
Are there gels specifically for LEDs?
Yes. According to LEE's Zircon range description, these LED filters use 180-micron material, which LEE describes as more than double normal lighting-filter thickness. LEE markets them as longer lasting and includes warm-amber and minus-green corrections for LED casts.
That is a manufacturer description, not a measured lifetime for your fixture. Choose the correction for the problem you see, and distinguish a published material thickness from a promise about service life.
FAQ
What is the difference between CTO and CTB?
CTO is orange correction toward a warmer white; CTB is blue correction toward a cooler white. Compare the starting and finishing temperatures in LEE's CTO data and CTB data when selecting a filter.
Does full CTO always give the same result?
No. According to Rosco, RoscoSun Full CTO converts 5,500 K to 2,900 K; according to LEE, its Full C.T. Orange converts 6,500 K to 3,200 K.
Can I use an ND gel without making the light blue?
Yes. LEE describes its 0.3ND filter as reducing light by one stop without changing colour.
Does a high melting point mean a gel cannot deform?
No. According to Rosco's specifications, its polyester softens at 125°C but melts at 300°C, so those thresholds answer different questions.
Do gels last forever on LED lights?
Do not infer an unlimited life from LED compatibility. LEE describes Zircon as longer lasting than standard filters, not permanent.
Jack Shi
Founder & editor, LEDaskJack Shi builds and writes LEDask, an independent LED-lighting tools project operated by clooms. He designs the calculators, checks their formulas and reference values against published engineering data, and writes the guides across the site.
