Cs10

Light convergence | Energy concentration | spectral analysis | Caustics

When using concave surface with some reflective material , the risk of light convergence need to be carefully watch out.

If the imaging of the caustics it creates on the ground, is now available on almost any renderer , the evaluation of the energy concentration is a bit more difficult to get.

Fig A: Photometric animation showing several outputs before and after noon Paris Sun approximation spring

Fig B | Reference case without any glass

Fig C | comparison | no glass | 100-2500 nm | 380 – 780 nm

Fig D | Low iron glass | 380 – 780 nm

Fig E | Low iron glass + solar low E coating | 380 – 780 nm

Fig F | Low iron glass + solar low E coating | 100 – 2500 nm

Very few simulation technology process full spectrum , which means that the simulation can be done for all wavelength between 100 to 2500 nanometer. In other words, from the uv range , to the visible light and beyond into the infrared up to 2500 nm.

In this case study we will use this technology , to some photometric analysis of the energy concentration due to a concave geometry. We work mostly in the visible range (380-780) and do some comparison with an extended range (from 100 to 2500 nm) .

We will use 3 probes , A B C to record the measures . The probe are fully Lambertian grey with 18% reflectance . There are placed in order to compare:

Probe A: an area fully outside the zone of light concentration,

Probe A: an area fully outside the zone of light concentration,

 Probe C: a area at the fold of the caustic with the higher concentration

The light source is an hosek wilkie sky , turbidity 3, set in Paris with an angle of 65° from the ground which correspond to a summer sun angle approximately.

We are not interested by the energy levels themselves as they are dependent of the spectral range that is considered. (Fig C) The effects of an amount of energy will be different if it is considered in the UV range, or in the visible range or in IR range. They will also vary according to the “material” which receive the energy, some are more sensible UV, like the skin or some plastics, some are more sensible to heart and therefore IR …ect

We are not interested by the energy levels themselves as they are dependent of the spectral range that is considered. (Fig C) The effects of an amount of energy will be different if it is considered in the UV range, or in the visible range or in IR range. They will also vary according to the “material” which receive the energy, some are more sensible UV, like the skin or some plastics, some are more sensible to heart and therefore IR …ect

It seems more relevant to get some order of magnitude to make first some ratio which compare an area with more (Probe C) or less (Probe B) high light concentration with one area without (Probe A)

The glass and coating brand are not disclosed but they are all based from measured spectral data from 100 to 2500 nm with 10 nm step.

For the reference case, we consider a double laminated 10/10/2 in low iron for the inner and outer pane of the IGU.

Fig D shows that the highest point of concentration is already between 2.6 time higher energy level (based on a visible range outtake) than in a area with only direct light.

The next two simulations consider a glass configuration based on low iron with a solar coating + low E, with the following performance : TL 50% g 0,27 RE 20% (rounded up numbers)

Fig E and F shows the same glass configuration of glass with two different spectral range , Fig E In the visible range and Fig F , 100 – 2500 nm including UV and Infrared wavelength

The wider the range the higher the amount of energy which is considered, but also we can see that the infrared contribute a lot as the ratio which has already grown from 2.5 to 3.6 with addition of the coating in the visible range is getting to 8.3 times if you consider it with the wider spectral range.

Fig G | Low iron glass + solar low E coating | 100 – 2500 nm | 380 – 780 nm

Fig H Low iron glass + solar low E coating | Light shelves | 380 – 780 nm

Fig G is showing the same result with the same color shart and range

It is not a surprise that something need to be done to reduce the impact of the concentration or to suppress it, regardless of the glass composition since even with just a generic glass with no coating it is already a subject.

One idea is to put a basin of water … with no fishes. Since we know that the maximum risk comes with a higher sun, we could devise some light shelves that are here to cut the reflection.

Fig H shows that with adding some light shelves we can to cut entirely the light concentration for the higher sun angle.

Fig I | Low iron glass + solar low E coating | + reflective coating

Fig J Low iron glass + solar low E coating | Light shelves | 380 – 780 nm

The Solar + low E coating tend to affect the appearance of the glass. Here we try to enhance the aspect of the glass by adding some reflectivity to the glass , about 10 to 15%. Fig I

This additional reflectivity will raise the amount of energy recorded on the probes, about 4.6 times instead of 4.1 time with a glass with no additional reflective coating, which is not that much comparatively with the actual raise due to the geometry itself. (it would start to be substantial if the additional reflectivity was higher) Fig J

Fig K | Low iron glass + solar /low E coating + reflective coating 10 % | light shelve| 380 – 780 nm

In this serie of simulations , we keep the previous configuration which tends to look better and add the light louvers with a anodized finish to cut off the light concentration