The Elexpert LED’s feature the daylight harvesting feature.  When the ambient light level is very lower the LED will be running at full power.  As the level of ambient light increases the power output reduces accordingly.  At ambient light levels close to 1000 Lux, the power output level will reduce to less than 20% of full power thereby providing even more savings.  The % output at various levels of ambient light is shown in the table below.

On a cloudy day, where the ambient light in the building is about 100 Lux, the Led lights will substitute with 75% power.

Because South Africa is such a sunny country with more than 90% of the days having ample ambient light, consideration could be given to even consider reducing the light requirement slightly below the ideal requirement for these few day, obviously still within safety standards.

This feature will obviously not make a difference in areas where very little sunlight can come in.  The type of light fitting can also have a bearing on the effectiveness of this feature.

The LED used is the Luminus CXM-32 Cob Array.  The specific unit used mainly is the CXM-32-50-70-54-AC00-F2-3 which yields 140 Lumens per Watt at 150 Watt with a CRI of 70.  The LED can be selected based on customer’s requirements.  The table below shows the Lumen maintenance curve for a similar Luminus LED with actual test of 6000 hours.  The CXM 32 has only been released recently and therefore these results are not available yet.

The Luminus LED’s are well known worldwide and offer excellent Lumens yield.

The fan used is X-Fan RDL8025B4.  This is a 48 Volt fan.  The life cycle tests for this unit is shown below.

The working temperature for the fan would be less than 60 degrees C.  This will yield a mean time between failures of more than 160 000 hours.

The driver is a constant current supply which have been developed and are being made in South Africa. The technical specifications for this is shown below.

The life expectancy of the various key components of the product are as follows:

• The Power supply. – 10 years plus continuous.
• The fan. – 10 years plus continuous.
•   LED. – 10% failure after 50 000 hours usage.
• Other components. – At least 10 years.

The LED is so mounted that it can easily be replaced.  When the LED thus goes faulty, only that has to be changed.

In the main thus these lights can easily be used for 10 years and more.

The following should be noted in terms of power factor for the new LED installation:

• The power factor of the new LED with its power supply is better than 0.9.
• The power factor of a some of the HID lights are usually not better than 0.9.
• Many of them are however as bad as 0.6
• When these become faulty, the power factor could reduce to typically less than 0.6.

Because the power consumption of the new LED’s is so small compared with the existing lights, the total varh consumption would always be less as illustrated below.

What this shows is that only if the power factor on the new LED power supply drops below 0.83, would the amount of vars generated from the new LED power supply reach the level of the current lighting supply.

installation of the new LED lights will have on the electricity bills of customers.  The following should be noted in this respect:

• The prices for customers differ all over South Africa.
• Eskom’s prices for large and industrial customers are generally the lowest in RSA.
• Different tariff structures are offered namely:
  • Maximum demand tariffs with demand / access charges. These will charge for:
    • the maximum demand taken in each month / year
    • and will also penalise for a bad power factor
  • Single or 2 seasonal energy charges.
  • Or Time of use energy charges.
  • Reactive energy charges.

The impact for every customer will be different and therefore a study has to be done for each customer individually.  The impact that has to be determined is the following:

• To what extent is the energy reduced.
  • This has to be done for kWh per the relevant period in the case of TOU and seasonal tariffs.
  • The extent that the reactive energy charge (c/kvarh) is reduced.
• The extent that the maximum demand is reduced.

For the examples the following has been assumed:

• For the 24 hour operation the price would reflect all the charges as for a 100% load factor customer. This was calculated at a typical municipal large customer tariff at 400V.
• For the 10 hour per day 22 days per month operation it was assumed that the full demand would impact the saving in maximum demand because the times that the lights operate would co-inside with the operation of the business. (This would not necessarily be the same for security lights burning at night only.)
• The impact on the reactive power is neglected because it is small.

The net result of the above assumptions yield the following average price impact for the different cases:

• 8 hours a day weekdays only.
• 10 hours a day for all days of the month.
• 24 hour operation all year.
• These will be escalating by at least 8% per year for the next few years.

The key assumptions relating to each option is shown below.

The annual NPVs for the 8 hour per day for weekdays all year is shown in the table below.

The typical breakeven situations are shown in the figure below.

It is thus clear that the all options break even within 21 12 months.

In all three cases the expected NPV over just 5 years exceeds R4 000.  This is excellent considering all the environmental advantages and the small investment of only R2 000 required.  In cases where there is ample sunlight, the savings will increase further with the daylight harvesting feature.