
Cannabis Lexicon
For decades, the high-pressure sodium (HPS) lamp was the gold standard for high-performance indoor lighting. Today, it is primarily interesting because it clearly illustrates how far plant lighting has evolved from wattage toward PPFD, DLI, µmol/J, and uniform canopy coverage.
HPS lamps explained: spectrum, PPFD, light penetration, heat, power consumption, and a comparison with modern LEDs – including the question of whether this former gold standard is still relevant today.
Definition
An HPS lamp is a high-pressure sodium lamp belonging to the HID (High-Intensity Discharge) family. Light is produced by an electrical gas discharge within an arc tube; the system requires a compatible ballast, a socket, and in grow applications, typically a reflector.
Spectrum
Warm, characterized by yellow-orange tones, low blue light, and, depending on the bulb, relevant red and far-red proportions.
Strength
High photon output, proven technology, and still capable of high flower yields.
Weakness
Lower photon efficiency than modern LEDs, higher heat load, and the need for regular bulb replacement.
In this article
Key Takeaway
When comparing HPS and LED, wattage alone is rarely helpful. The decisive factor is how many photons consistently reach the canopy – and how much electrical energy is required to achieve this.
The high-pressure sodium lamp, commonly known as HPS in the grow community, is one of the most definitive lighting systems in indoor cannabis culture.
For decades, it was effectively the benchmark for high-performance indoor lighting. Entire generations of grow rooms were built using 250, 400, 600, or 1000-watt HPS systems, long before modern LED grow lights achieved their current levels of efficiency.
Technically, however, the HPS is not a specialized "cannabis light" but a high-intensity gas discharge lamp. Its strengths lie in high light output, proven technology, and a spectral distribution that plants can utilize. Today, its weaknesses primarily involve energy efficiency, heat management, light distribution, maintenance, and lifespan.
At the same time, modern cannabis research shows that HPS can still deliver very good yields. However, the former notion that sodium vapor is inherently the best light for flowering can no longer be scientifically supported.
An HPS lamp belongs to the group of HID lamps – High Intensity Discharge.
In this system, light is not generated by a semiconductor as it is in an LED. Instead, an electrical gas discharge occurs within a small arc tube.
In high-pressure sodium lamps, this arc tube contains elements including:
The electrical discharge excites these substances and produces the lamp’s characteristic intense radiation. HPS lamps also require a suitable ballast and, depending on the design, an igniter.
A classic grow HPS system therefore consists of more than just the bulb itself.
The system typically includes:
Bulb + Socket + Reflector + Ballast + Wiring
This technology differs significantly from modern LED lights, where the light source and electronics are usually designed as a complete fixture.
As indoor cannabis cultivation grew, HPS offered a highly attractive combination at the time:
HPS was already established in professional greenhouses and could provide supplemental light to relatively large areas.
It was therefore logical to use the same technology for light-intensive indoor crops.
Modern LEDs had to measure up to HPS for years as a result.
Typical HPS light is often simply described as red light.
That is inaccurate.
High-pressure sodium lamps usually possess a very warm-looking spectrum with strong portions in:
and, depending on the lamp, also portions of:
Blue, by contrast, is comparatively weakly represented.
In a controlled cannabis study, the HPS system examined contained only about 4% blue photons, while the LED systems tested delivered between approximately 10% and 20% blue.
Another horticultural measurement characterized an HPS spectrum as roughly 6% blue, 50% green, 34% red, and 7% far-red. However, such values depend heavily on the specific bulb used.
It is therefore more accurate to say:
HPS typically possesses a warm, yellow-orange dominated spectrum with little blue and relevant red/far-red components.
In classic indoor growing, this led to a simple division:
MH for growth
HPS for flowering
Metal halide lamps typically provide more short-wave or blue radiation. HPS provides significantly less blue and larger portions in the yellow-orange-red range.
Since blue light can, among other things, promote more compact morphologies, while low blue portions often allow for stronger stretching, HPS seemed well-suited for the reproductive phase.
Cannabis trials confirm at least part of this observation.
At the same PPFD, plants under HPS were found in several studies to be taller or more stretched than plants under certain LED spectra.
However, this should not lead to the conclusion:
Red light forces cannabis into flowering.
For photoperiodic cannabis, it is primarily the photoperiod or length of the dark phase that is decisive for reproductive development. The spectrum can influence morphology and plant physiology, but it does not replace photoperiodic control.
No.
The well-known 12/12 cycle is not related to the high-pressure sodium lamp.
It is a classic photoperiod strategy for photoperiodic cannabis.
Newer cannabis research even shows that 12 hours of light and 12 hours of darkness do not necessarily represent the yield optimum for every genotype. In one study, some chemovars were able to be developed further reproductively even with 14 hours of light, with flower mass and cannabinoid concentrations changing differently.
HPS and photoperiod are therefore two different levels:
HPS = Light source
12/12 = Photoperiod
The so-called penetration of the HPS is one of the most persistent terms in the grow community.
In this context, several things are often conflated.
An HPS lamp is a relatively concentrated light source. Therefore, very high PPFD values can be generated directly under a powerful lamp.
Subjectively, this can create the impression that the light "penetrates deeper" through the plant.
However, the actual amount of light within a canopy depends on, among other things:
A modern LED bar light, for instance, distributes the same number of photons over a much larger area. As a result, the peak intensity directly under the light source may be lower, but the uniformity across the entire canopy can be better.
Newer research on cannabis lighting explicitly describes the shift to LED as also providing more homogeneous lighting of the plant canopy.
Therefore, the statement:
HPS generally has better penetration than LED
is too generalized.
An additional factor is the relatively high green-yellow proportion of many HPS spectra.
Green photons are less strongly absorbed by upper leaf areas than a portion of the red and blue light, allowing them to penetrate deeper into leaves or plant canopies.
However, this does not mean that only HPS possesses this advantage.
Modern full-spectrum LEDs can also generate relevant amounts of green light.
The question of canopy penetration should therefore be considered through:
Spectrum + Light distribution + Plant architecture
rather than through the technology alone.
For plants, wattage alone is not very meaningful.
A 600-watt HPS lamp only tells you something about the electrical power of the system at first.
More interesting is the PPFD – Photosynthetic Photon Flux Density that actually reaches the plant canopy.
It describes the number of photosynthetically active photons hitting a square meter per second:
µmol/m²/s
This makes it much easier to compare different lighting technologies.
A comparison should therefore not be:
600 Watt HPS versus 600 Watt LED
but rather:
What PPFD and light distribution do both systems produce at what power consumption?
In addition to PPFD, the Daily Light Integral – DLI is interesting.
The DLI takes the following into account:
PPFD × Photoperiod
A plant under 600 µmol/m²/s for 18 hours therefore receives significantly more daily photons than the same plant at 600 µmol/m²/s for 12 hours.
This is particularly important when comparing the vegetative and flowering phases.
A high PPFD alone does not describe the total daily amount of light.
Cannabis can utilize relatively high light intensities.
A controlled study with a THC-dominant cultivar examined average canopy PPFD levels of:
The highest light level produced about 1.6 times more dry inflorescence mass than the lowest. The cannabinoid concentration itself did not change significantly.
A more recent study with CBD-rich hemp also found increasing floral biomass and cannabinoid yield with increasing LED light intensity between 200 and 600 µmol/m²/s.
This shows:
For yield, the actual amount of photons is often more important than the historical question of “HPS or LED?”
The comparison is more complex than marketing from both sides might suggest.
A cannabis study compared HPS with two LED spectra at approximately the same light intensity.
The HPS-grown plants were:
However, the THC concentration of the flowers was lower under HPS than under the two LED treatments.
In terms of total cannabinoid yield per plant, there were no significant differences between the lighting systems.
This is a good example of why:
flower weight,
cannabinoid concentration
and
cannabinoid yield
are not the same thing.
In another study, PPFD was strictly controlled between the lighting treatments.
The tested HPS lamp had the lowest blue content and produced slightly higher floral yields per area than some of the tested LED spectra.
But:
The HPS light achieved only about 1.7 µmol/J, while the most efficient LED tested produced about 2.5 µmol/J.
The result was remarkable:
The LED produced slightly less per area, but due to its higher electrical efficiency, it achieved about 27% more yield per dollar of electricity spent within this experimental model.
The decisive metric was therefore not just the spectrum.
It was the photon efficiency of the entire fixture.
The µmol/J metric describes how many photosynthetically active photons a light produces per joule of electrical energy used.
The higher this value, the more efficiently electricity is converted into usable plant light.
This makes it much more helpful for comparing modern grow lights than:
A high-quality double-ended HPS typically lies in the range of 1.7–1.8 µmol/J.
Modern LED technology can be significantly higher.
This is exactly where the decisive technological advantage of LED arose.
Yes – but this point also requires precision.
Every electrical grow light ultimately contributes to the heat load of the room.
So an LED is not a cold light.
The key difference lies in how much electrical power is needed for the same amount of photons and how the heat is dissipated.
HPS lamps have very hot burners and emit a comparatively large amount of long-wave heat radiation toward the plants.
LED systems dissipate a larger portion of their waste heat through:
For comparable PPFD, leaves under HPS can therefore become warmer.
An energy-balance study under typical indoor conditions found approximately 1.3 °C higher leaf temperatures under HPS than under LED, whereby water status, air movement, and transpiration can have an even greater influence on leaf temperature.
Under HPS, special attention must therefore be paid to the following factors:
A room with a 600-watt HPS usually requires different climate management than a more efficient LED system, which generates the same PPFD with significantly less electrical power.
This also explains why a direct watt-for-watt comparison makes little sense.
The heat dissipation of HPS is not exclusively a disadvantage under all conditions.
In a:
additional radiant heat can sometimes be desirable.
Horticultural research shows that plants under HPS can have warmer leaves due to higher long-wave heat radiation.
The supposed efficiency disadvantage can thereby partially replace heating energy in a cold production system.
In an already warm indoor room, however, the opposite happens:
The heat must be removed again using exhaust air or air conditioning.
Due to the combination of high spot PPFD and heat radiation, insufficient lamp distances can become problematic.
Possible consequences are:
A blanket distance table is not very helpful here.
Crucial factors are:
A PAR meter therefore provides more information than the pure centimeter measurement between the light source and the tip.
Before full-spectrum LEDs were widespread, indoor growing often involved switching between two HID systems:
stronger blue content and cooler-looking light.
less blue, significantly warmer spectral distribution.
The classic practice was therefore:
MH in the vegetative phase → HPS in flowering
In professional horticulture as well, different discharge lamps or mixed systems were used to better adapt the spectrum to cultivation goals.
Today, a single LED light can combine such spectral ranges much more flexibly.
Not in the sense that only red light could produce flowers.
Cannabis uses photons across a broad range of photosynthetically active radiation.
The spectrum can influence:
However, the results are highly genotype- and experiment-dependent.
Cannabis studies show, for example, different reactions of morphology and cannabinoid profile to different blue-to-red ratios.
The simple rule:
Red = Flower
is therefore more of a growing tradition than complete plant physiology.
An important disadvantage compared to many modern LED systems is lamp degradation.
HPS bulbs lose light output with increasing operating time.
Michigan State University describes replacement after approximately 12,000 operating hours as a typical guideline for HPS, when performance has dropped to about 85–90% of the original output.
At the same time, the bulb continues to consume approximately the same electrical power. As a result, efficiency decreases with age.
Also, dirty:
can further reduce the actual photon output reaching the plants.
For plant lighting, lux is only suitable to a limited extent, because the value is weighted according to the sensitivity of the human eye.
HPS in particular has large emissions in wavelength ranges to which the human eye reacts strongly.
Consequently, an HPS lamp can appear extremely bright in lux, without this value being directly comparable to plant-relevant PPFD.
For grow light, it is therefore better to use:
are more suitable.
You can find more on this under Lux vs. PPFD.
A classic HPS lamp requires a compatible ballast.
It limits the current and, in conjunction with the ignition technology, establishes the electrical conditions necessary for startup.
HID systems also require a certain warm-up time.
Following a power interruption, hot HPS lamps cannot always reignite immediately. A "restrike" can only occur after the burner has cooled down sufficiently.
This clearly distinguishes them from LEDs, which can be turned:
practically instantaneously.
An HPS lamp operates with:
Therefore, the following should be designed for the appropriate power rating:
Improvised wiring, overloaded power strips, or unsuitable sockets are particularly unwise solutions for high-performance HID systems.
In an HPS system, the reflector is a central component of the lighting setup.
Since the lamp emits light in many directions, a portion of it must be directed back onto the plant area.
The shape of the reflector therefore influences:
A poor HPS setup can produce extremely high PPFD in the center while simultaneously delivering significantly too little light at the edges.
A modern, large-area LED bar design can reduce these discrepancies.
This makes it clear that:
Light quality is not just spectrum – it is also light distribution.
A cooltube, or air-cooled hood, was developed to dissipate some of the heat from an HPS system directly at the light source.
Air is drawn past the hot lamp and transported directly out of the grow area.
This can:
At the same time, the glass has a disadvantage:
It reduces some of the usable light output.
Scientific HPS/LED trials also account for this effect; a glass pane in front of an HPS lamp can measurably reduce photosynthetic photon flux.
Technically:
Yes.
Economically and practically:
It depends on the setup.
An existing HPS system can still be interesting if:
A functioning HPS system does not suddenly become useless just because better LEDs exist.
For completely new purchases, there are now many reasons for choosing modern LED systems:
The cannabis study by Westmoreland et al. demonstrates perfectly why efficiency is so important: while HPS achieved slightly higher biomass per area in their trial due to its low blue light content, the more efficient LED produced significantly more flower mass relative to electricity costs.
The more sensible question today is not:
Which technology is fundamentally better?
But rather:
How many photons reach the canopy uniformly – and how much energy does that cost?
Important metrics are:
Followed by:
This makes a comparison much more robust than "600 Watt HPS versus 600 Watt LED".
A classic example:
A 600-watt HPS requires not only the nominal lamp but also a ballast with its own losses.
The actual power consumption of the system can therefore be slightly above the value stated on the lamp.
If an LED produces the same usable amount of photons with less electrical power, two efficiency advantages arise:
Especially with long lighting periods, this difference adds up significantly.
Another myth is:
HPS produces fundamentally better flowers.
There is no general scientific basis for this either.
A controlled study found:
Other studies, in turn, show that low blue light content can promote flower yield without significantly changing THC or CBD concentrations.
The result therefore depends on:
No single lighting technology possesses a universal quality bonus.
There is no simple rule for terpenes either.
The light spectrum can influence the composition of secondary metabolites, but the reaction is highly dependent on the genotype.
Therefore, a specific HPS lamp does not guarantee:
Just as little does a full-spectrum LED automatically guarantee a superior profile.
Genetics remain the starting point; light is one of several environmental factors that influence how this potential is expressed.
HPS lamps do not belong in household waste.
Gas-discharge lamps can contain mercury and, in Germany, are subject to the Electrical and Electronic Equipment Act.
The Federal Environment Agency points out that gas-discharge lamps must be collected separately and can be handed in, for example, free of charge:
The lamp should be transported in a break-proof manner if possible.
In this respect, the disposal of an old HPS lamp also differs significantly from that of a classic incandescent bulb.
No. Cannabis studies show both strengths of HPS and advantages of modern LEDs. Decisive factors are PPFD, spectrum, genetics, and fixture efficiency.
Not as a universal technology trait. Light distribution, spectrum, and plant architecture determine how much radiation reaches deeper canopy areas.
Only if this actually results in more usable photons and the plant can utilize them.
For photoperiodic cannabis, the length of the light or dark phase is the decisive factor.
False. LEDs also convert a portion of electrical energy into heat. However, they are typically more photon-efficient and transfer less long-wave heat radiation directly to the plants.
No. Light output decreases over the operating life.
No. In a direct cannabis study, the THC concentration under HPS was actually lower than under the two LED spectra tested.
NDL stands for sodium vapor lamp. In the context of cultivation, this usually refers to a high-pressure sodium vapor lamp, or HPS.
HPS stands for High Pressure Sodium.
Yes. High-pressure sodium lamps belong to the category of High-Intensity Discharge lamps.
Yes. A classic HPS requires a compatible ballast and the corresponding ignition technology.
Typical HPS systems have a warm, yellow-orange dominated spectrum with relatively little blue, as well as relevant red and some far-red components.
No. Cannabis can also grow vegetatively under HPS. However, due to the low blue light content, the morphology may end up more stretched than under bluer spectra.
Metal Halide typically has a higher blue content, which can promote a more compact plant morphology.
Not fundamentally. Modern LEDs usually have a clear advantage in photon efficiency and can distribute light more evenly. HPS can still achieve high yields, however.
In some individual trials, HPS produced slightly more biomass than LED spectra with higher blue content at the same PPFD. However, in terms of electrical energy used, more efficient LEDs were more economical.
No. A direct comparative study found lower THC concentrations under HPS than under the LEDs tested. Conversely, the total cannabinoid yield did not differ significantly.
HPS generates more directed long-wave heat radiation towards the canopy. At a comparable PPFD, leaves can therefore become slightly warmer. The actual difference depends heavily on air circulation and transpiration.
PPFD describes the density of photosynthetically active photons hitting a surface per second.
This key figure describes how many plant-relevant photons a light fixture generates per joule of electricity consumed.
Light output decreases as operating hours increase. In commercial horticulture, it is therefore recommended to replace old HPS bulbs before they fail completely.
No. In Germany, gas discharge lamps must be collected separately as waste electrical equipment; they may contain mercury.
The sodium vapor lamp is one of the most important technologies in the history of indoor cannabis cultivation. For decades, HPS was the benchmark for powerful artificial lighting—and for good reason. The technology can generate high PPFD, reliably guide cannabis through a complete cultivation cycle, and deliver very high bud yields under suitable conditions.
Many of the classic arguments for the NDL must be viewed more precisely today, however. Its spectrum is not simply "red," but is heavily characterized by yellow-orange and is comparatively poor in blue. A special "penetration effect" is not a universal property of the technology. And even regarding cannabinoids, controlled comparisons show no fundamental quality advantage over LED.
The biggest difference today lies in efficiency. Modern LED systems can generate the same amount of plant light with less electrical energy and distribute heat differently. As a result, depending on the setup, both power and climate control requirements decrease. HPS, on the other hand, can even use the additional heat radiation to its advantage in cold rooms.
Therefore, in 2026, the NDL is neither "bad" nor technically insignificant. An existing, well-functioning system can continue to serve its purpose. However, for a complete new design, energy efficiency, light distribution, dimming, and maintenance requirements now mostly point much more strongly toward modern LED technology.
The NDL was not successful because cannabis necessarily requires sodium vapor. It was successful because, for decades, it was one of the most efficient ways to provide a large amount of photons to an indoor plant area. Today, LEDs can perform this exact task more efficiently—and with that, the former gold standard becomes one thing above all: a functional classic with clearly defined strengths and equally clear technical limitations.