I was wondering if anyone knows of a good UV sensor. I’ve been looking around for a UV sensor so that I could offer a 1-Wire UV kit to go with my solar kit but I have been unable to find a decent sensor.
Thanks,
Eric
www.hobby-boards.com
I will be watching this thread and for developments in this area very intently 
Eric get ready to ship one of those babies to Jax once you get the minor details worked out LOL
-Bob
i know nikoshepared has made alot of progress with one he made
True, I have made one, and it does work. Since I don’t want the Bluewave InTernet Conspiracy Hitsquad again accusing me of putting them out of business I’ll PM Eric with some information.
I’ve purchased one of nikoshepherd’s early kits, I must get round to putting it together :oops: Sounds like a good job for a Sunday afternoon…
[quote author=41South link=topic=9426.msg67561#msg67561 date=1118459880]
I’ve purchased one of nikoshepherd’s early kits, I must get round to putting it together
Nikosheperd has all the information - hopefully he will pop up here with a link.
Don’t get excited, “kit” overstates it. I did buy 4 sets of the critical parts (to minimize shipping) and supplied them to a couple of people on here. I finally :roll: managed to get one full day of sun/UV captured yesterday so I will post a write up later.
I’m interested in the UV sensor, well the front end of it If I could get some info that would be great
I’ll PM you some info. Due to the high cost of the sensor diode (>$ 30.00) I doubt this is a viable thing for anyone to offer as a kit. It does work, but as expected it doesn’t give very consistent coverage through the day due to the lack of optics. Here’s the output of my home built UV sensor and a solar sensor over a full day of hazy sky. The peak values have been adjusted to be the same to show the difference in response. This was logged with a LabJack and graphed in Excel.
Well I am still holding out for a possible 1-wire UV sensor solution BUT if push comes to shove I will give the solar sensor a shot…
-Bob
who has a graph with a davis solar and UV on it to see the relationship?
This was logged with a LabJack and graphed in Excel.
Any ideas on a 1-wire interface as yet, although I’m guessing the same DS2438 as the solar sensor should do it. Even then though doesn’t this sensor need some math to calculate the UV reading? Or am I losing the plot #-o (if you’ll excuse the pun).
24 hour and 48 hour view
The hitch on the left side (each morning) is a neighbors palm tree…
Well…
To this untrained eye they look almost identical…
If that is the case then I guess I will be getting that solar sensor after all and figuring out how to mount it…
-Bob
Thanks Kevin!
Unfortunately confirms what I thought, if the sensor is optically correct the UV curve will more or less track the solar.
Creating/obtaining a good UV transparent wide angle lens is likely to be tricky, so for those Heath Robinson type people out there I offer a few interesting potential solutions (remember me when you make your first million from the idea!) The problem is that the sensor is designed to measure UV that hits the window on top of the can. At low sun angles the can sides/top shield the sensor so the UV levels are reduced in the morning and afternoon.
-
Calibrate the sensor and do some maths on the incoming data to correct for the error (not very good for WD integration though).
-
Fix the sensor to the shaft of a stepper motor and rotate the sensor to point at the sun as the sun moves. Drive the motor using a signal from your PC telling it where the sun is. This solution would need tailoring depending on your latitude. On the equator the sensor would need to point vertically with the shaft rotating horizontally. Further north you’d need to mount the sensor and the motor at an angle.
-
Use a stepper motor as in (2), but drive it using a local servo circuit that seeks out the strongest sunlight. This avoids the need for any computer control.
-
Use a stepper motor as in (2), but drive it continuously between the two extremes of the sun’s range. If it rotates between the extremes in 2 seconds, then just pick the highest level of UV reported every 2 seconds as the current value.
In all but (1) make sure you use very flexible coupling wires!
I’ll let someone else try these whilst I’m busy with my other project.
Close to impossible I think. It’s not like the solar where there’s a lot of sensitivity to spare and almost any diffuser will do. Just about all plastics and most glass block a very large % of UV light.
1) Calibrate the sensor and do some maths on the incoming data to correct for the error (not very good for WD integration though).
That’s a great idea! Wouldn’t be a trivial calculation, need to use the Broadstairs sun angle stuff, but very doable
Got me thinking…I don’t like all them moving parts, calculations, etc. I was thinking along the lines of a prism, an optic that would bend the light from the sun to always hit the sensor type without all that moving around…