Solar street light autonomy in Morocco: the calculation

Solar street light autonomy in Morocco is sized on December, the weakest month: 2.58 kWh/m²/day in Tangier, 2.99 in Casablanca and 3.81 in Ouarzazate on a horizontal plane (PVGIS-SARAH2 2005-2020, JRC). A flat 100 Wp panel then harvests ≈ 215 Wh per day in Casablanca and ≈ 274 Wh in Ouarzazate at a system efficiency of 0.72, for a winter night of 14.1 h. The calculation fits on three lines: what the panel charges in December, what the LED draws at the chosen dimming profile, and how many days of reserve the battery keeps for overcast weather.

Why size a solar street light on December rather than on the annual average?

Because a light that switches off at 2 a.m. in December is a failed light in the eyes of the commune or the residents' association, even if it runs all night in July. In Morocco, December delivers only 49 to 64 % of the annual daily mean depending on the city (2.99 against 5.38 kWh/m²/day in Casablanca), while June and July sit between 7.3 and 8.2 kWh/m²/day everywhere. Sizing on the annual average means counting on 1.6 to 2 times the energy actually available at the very moment the night is longest. The southern regions exceed 2 200 kWh/m² per year, which is why the same product behaves so differently in Tangier and in Laâyoune.

How many peak sun hours does each Moroccan city get in December?

Peak sun hours (PSH) are simply the daily irradiation expressed as equivalent hours at 1 000 W/m²: 2.99 kWh/m²/day = 2.99 PSH. In December, Morocco ranges from 2.58 PSH in Tangier to 3.97 PSH in Laâyoune on a horizontal plane, which is what the flat panel of an all-in-one solar street light sees. A separate panel tilted 32° towards the south changes the picture: 4.32 PSH in Tangier, 4.93 in Casablanca and 6.37 in Ouarzazate, a gain of +62 to +68 % in December at the same Wp (Laâyoune: 27° and +44 %). On the other side of the ledger, the night to be lit lasts 14.1 h in Casablanca on 21 December (astronomical calculation), against 9.4 to 10.1 h on 21 June.

Irradiation from PVGIS-SARAH2 2005-2020 (JRC); night length by astronomical calculation.

How do you turn panel Wp into Wh available each night?

The formula we use for every quotation is deliberately simple: daily energy (Wh) = Wp × December PSH × system efficiency. The system efficiency, between 0.70 and 0.75, bundles panel temperature, soiling, MPPT controller efficiency and battery charge and discharge losses; it is an engineering assumption we state openly, not a measurement. At 0.72:

• Casablanca, 100 Wp flat: 100 × 2.99 × 0.72 ≈ 215 Wh/day.
• Ouarzazate, 100 Wp flat: 100 × 3.81 × 0.72 ≈ 274 Wh/day.
• Casablanca, 100 Wp tilted 32° (split kit): 100 × 4.93 × 0.72 ≈ 355 Wh/day.

Those Wh are the budget for the following night. An LED driven at 40 W full flux exhausts them in 5.4 h in Casablanca, which is exactly why no serious solar street light runs at 100 % all night in winter, and why the dimming profile is part of the sizing just as much as the Wp.

What does the LED really draw: nominal watts or average watts?

The power printed on the datasheet ("120 W LED") is the module rating, not what it consumes. Our modules deliver 180 to 220 lm/W; a luminaire driven to ≈ 6 000 lm therefore draws in the order of 6 000 ÷ 200 ≈ 30 W of electrical power at full flux (LED efficacy, an optimistic assumption: the luminaire as a whole is less efficient), and far less in standby. What matters for autonomy is the average power over the night, set by the controller: full flux in the early evening, reduced flux after midnight, a step up towards dawn, and motion detection (PIR) on lightly used lanes. We unpack the difference between nominal watts, Wp and Wh in the guide A "300 W" solar street light: nominal versus real power.

Which dimming profile survives a 14-hour December night?

The table below takes the 100 Wp panel and an LED drawing 40 W at full flux, over Casablanca's 14.1 h night. The balance compares the energy used per night with the December daily harvest.

Reading the table: in Casablanca, only the dimmed profiles are sustainable in December with a flat 100 Wp panel; the same panel tilted 32° (355 Wh/day) supports the "100 % for 4 h, then 40 %" profile. In Ouarzazate, the flat panel copes with the 40 % profile and every lighter one. This is why we steer coastal and northern projects towards the tilted split panel, and accept the all-in-one format more readily in the south and south-east.

How many days of reserve does the battery really give?

Reserve is the battery capacity divided by the daily deficit, not by the total consumption: an overcast day does not yield zero, and the controller reduces the flux as soon as the voltage drops. Formula: days of reserve ≈ battery Wh ÷ (Wh used per night − Wh harvested on an overcast day). On our all-in-one solar street lights the 3.2 V LiFePO4 battery is sized for "up to 12 h per night, 3–5 days of reserve at the default profile" and recharges in ≈ 8 h of full sun from empty; autonomy depends on irradiation (sized on December) and on the dimming profile. The LS-A60 carries 60 Ah, i.e. 192 Wh; the LS-A120, 120 Ah or 384 Wh; the LS-A140, 160 Ah or 512 Wh. A split kit LS-S60 houses 30 to 50 Ah at 12.8 V at the base of the pole, i.e. 384 to 640 Wh, which doubles or triples the reserve on unattended rural tracks.

Casablanca versus Ouarzazate: the complete worked example

Take an all-in-one LS-A60 (60 Wp, 192 Wh, ≈ 6 000 lm) and an LS-S60 kit with a separate 100 Wp panel tilted 32° and a 12.8 V 40 Ah battery (512 Wh). December harvest at an efficiency of 0.72:

At ≈ 30 W full flux, an LS-A60 set to 40 % all night uses 12 W × 14.1 h ≈ 169 Wh: in Ouarzazate the December balance is even (165 Wh harvested); in Casablanca the PIR profile (≈ 133 Wh) only just breaks even with the 129 Wh harvested; a tilted LS-S60 set for the same flux (≈ 30 W at full flux) holds the "100 % for 4 h, then 40 %" profile (≈ 241 Wh) with a margin of more than 40 %. In July both configurations are limited by the battery, not by the sun: the night lasts only 9.4 to 10.1 h.

What does the NM 06.8.001 standard require on autonomy?

The draft Moroccan standard NM 06.8.001 (public enquiry EP 52/2024, IMANOR) sets a clear test condition: 16 h of continuous operation at no less than 40 % of the luminous flux, at 25 °C, with a battery and a controller rated for a service life of at least 5 years. It also requires a controller with 0–100 % dimming, an astronomical clock and automatic weather-adaptive dimming based on the computed night length and the energy left in the battery: that adaptive behaviour is what makes our December balances hold in practice. The clause numbers quoted come from the September 2024 draft; the homologated text is purchased from IMANOR. One last point, often forgotten: temperature. In January the mean air temperature is 12.2 °C in Casablanca and 7.9 °C in Ouarzazate (ERA5 2005-2020, PVGIS); a cold panel produces slightly better than in summer, and that effect is already included in the 0.72 efficiency.

Key takeaways

• Size on December: 2.58 to 3.97 kWh/m²/day depending on the city, i.e. 49 to 64 % of the annual mean, for a night of 13.6 to 14.3 h.
• Daily energy = Wp × PSH × 0.70–0.75; a flat 100 Wp gives ≈ 215 Wh/day in Casablanca and ≈ 274 Wh/day in Ouarzazate.
• The dimming profile is a sizing parameter: at 100 % all night, no reasonable panel survives the winter.
• A separate panel tilted 32° harvests +62 to +68 % in December: the right choice for the coast and the north.
• The 3 to 5 days of reserve are computed on the daily deficit at the default profile, never on the total consumption.
• NM 06.8.001 tests 16 h at ≥ 40 % flux at 25 °C and requires weather-adaptive dimming: ask for that behaviour in your tender specification.

Sources

• PVGIS 5.2, SARAH2 database, 2005-2020, JRC (European Commission): re.jrc.ec.europa.eu/api/v5_2/MRcalc
• IMANOR, NM 06.8.001 catalogue entry: imanor.gov.ma/Norme/nm-06-8-001/ and draft EP 52/2024: PDF
• IMANOR, guide on applying standards in public procurement, February 2026: PDF
• Wikipedia (fr), "Énergie solaire au Maroc": fr.wikipedia.org