For Lux Sake - Again?

The Lux experiment has been an abject failure, well, certainly in terms of the placement of the Lux Sensor in the upstairs bedroom window.

I compiled a chart of Lux vs Solar PV power generation, as follows:

As you can see, the results were all over the place.

I have now placed the sensor in the kitchen window, still facing north and fixed to some tiles, but in a position more exposed to daylight. 

I applied the RayBans with a bit of Sellotape after it reached 1,000 Lux at 8am.

As you can see from the new chart (click to enlarge), it's infinitely more stable and there's an almost perfect, direct relationship between Lux and kW of solar power.

But......., the red anomalies in the top section of the cart are from the afternoon, when the Lux sensor went into partial shade from the house, so for the morning it's very accurate, but after midday it becomes accurate, but on a totally different scale of Lux. The perverse thing is that a lower Lux produced a high wattage. Now it's possible to make a separate programme for the afternoon, but what a pain in the arse. I shall persist, however.

One problem yesterday was the snow albedo. Snow has a high albedo, which means that it reflects a significant amount of the solar radiation that hits it. This is due to the snow's reflective surface, which is made up of small, irregular ice crystals. When light hits the snow, it bounces off the crystals in many directions, causing much of it to be reflected back into the atmosphere and into a Lux sensor.

That said, snow on the solar panels results in no solar PV anyway. The grey dots bumbling along the bottom of the chart show the Watts for Lux with the solar panels covered and the slight rise toward the right of the chart is where the snow started to slide off them.

I saw on the NEA website that there are smart devices that monitor solar PV generation, using a CT clamp around the wires from the transformer that measures the magnetic field, which is a neat thing, but no examples of cheap products were shown. The cost, apparently, is between £75 and £100. What I need is one that can be integrated into the Tuya App. The expensive ones divert excess solar generation to an immersion heater, but there's no reason why it can't be diverted to an ASHP.