Another DIY Sky (Quality?) Meter

[kbrown]

Hi,

So I have had already quite some time a TSL237 light to frequency sensor and also an MLX90614 IR thermometer in the view to make sort of a DIY SQM meter and cloud detector. Finally I've actually hooked them up to an Arduino Nano in a 3D printed enclosure for testing. I'm not really trying to copy one to one what Unihedron has done nor trying to match their readings. I just want useful relative data to compare the conditions from night to night. I'm thinking of making a device that sits on top of the scope looking at the same direction and focusing on what the sky conditions are there. So with this in mind I'm thinking of omitting the hemispherical lens in front of the TSL237 and just simply rely on the lens that it already has. However I'm not quite sure should I still add an UV/IR cut filter in front of the TSL237?

Here's the normalized responsivity at wavelengths from 300-1100nm from the datasheet. It seems to be peaking at the red end of the visible spectrum and is somewhat responsive to NIR region but not so much at UV region.

 

Here's the vertical and horizontal angular responsivity. Still pretty large FOV so I suppose it'll do just fine without an additional lens?

 

In regards detecting clouds, what is the general consensus? Is there some kind of a formula that tells what sky temperature vs ambient temperature corresponds to a certain level of cloudiness?

 

[skybadger]

I find that it depends....

Typically it's always clear below -18C 

It's always cloudy above 5C

It's in between where it can be tricky. I've had nights where the sky temp has been above zero and they were still clearer than other nights where it was below -5. 

I believed it was due to the direction of the prevailing wind at the time, bringing different bulk masses of air at different temperatures and humidities. 

If the air temperature is warmer than usual due to say coming up from the Sahara, it can still be clear and quite warm. If it's from the north or north east (for the UK) the air tends to be quite cold when it's clear. Which is when the -18 figure might be the useful figure. 

I have encoded this into my safety driver but what I really want to finish is my star detecror/cloud coverage estimator on the all sky camera.

[kbrown]

9 hours ago, skybadger said:

I find that it depends....

Typically it's always clear below -18C 

It's always cloudy above 5C

It's in between where it can be tricky. I've had nights where the sky temp has been above zero and they were still clearer than other nights where it was below -5. 

I believed it was due to the direction of the prevailing wind at the time, bringing different bulk masses of air at different temperatures and humidities. 

If the air temperature is warmer than usual due to say coming up from the Sahara, it can still be clear and quite warm. If it's from the north or north east (for the UK) the air tends to be quite cold when it's clear. Which is when the -18 figure might be the useful figure. 

I have encoded this into my safety driver but what I really want to finish is my star detecror/cloud coverage estimator on the all sky camera.

That's very useful information. Thanks for sharing.

I suppose sky will always be brighter than usual when there's clouds around at least in light polluted areas like where I am (Bortle 8).