Binary Seconds of Sunlight Counter - F-J's Physics - Video 195

[@AnthonyFrancisJones]

Calibration update: had a bit of a play with the potentiometer/variable resistor and it is now recording the hours of daylight to the nearest 10 minutes! Bit of a fluke I think!

[@markgreco1962]

This is a fun one Anthony Thanks

[@AnthonyFrancisJones]

Thanks Mark! Useless at the moment as our summer weather is so poor!!

[@david_pilling]

Old school, I could have made it in 1975. I've seen a thing that records sunlight by having a glass sphere that focuses the sun on a piece of paper and leaves a record by burning (The Campbell-Stokes sphere). Exercises for students, replace the binary counters with decimal. Use the LEDs to detect the sunlight. Replace it all with 20 lines of Arduino code. Use an ir thermometer chip to monitor the temperature of the sky. Good fun, thanks for the videos.

[@AnthonyFrancisJones]

Yes, I think the 555 timer at least comes out of a box of ones I had as a teenager! I did think of putting a decoder on it to show the values in decimal too on seven segment displays but I felt I wanted to keep this binary. Using an LED as a light sensor is a good application and one that not everyone is aware of. I have been meaning to do a video on the C-S solar recorder for a while. You can use a piece of film and a pinhole camera shoebox or a simple convex lens, toilet roll tube and paper. Obviously I would want students to be aware of the risks of working with bright sunlight. Do you remember that lady's house that burnt down, was it last year, with a glass item causing the curtains to go on fire! Always good to hear from you and hope you enjoy my next ones!

[@frogandspanner]

Where do you get the rule marked in mm? Most steel rules are in numbered in cm, although graduated in mm . I use my CNCd milling machine to do my drilling for me! I am awaiting boards from JLCPCB (currently at Heathrow - should be with me in a day or two) for CNCing my lathe. Perspex is an excellent material for mounting simple circuits. I have one of those printers too - all my components are logged in a database, and located in a raaco cabinet (or Gratnell for larger stuff) labelled with bar codes. I use Sainsbury's sauce pots - 40 ml with screw-on lids, two will fit in a component drawer. An app on my 'phone will scan barcodes and look up in my database - very useful as memory fades with age. Now, get your device to measure oktas.

[@AnthonyFrancisJones]

Wow! That sounds like you have a fantastic and well sorted set up! I have a few Raaco type cabinets mainly for electrical connectors (most of which I will never use such as MUSA etc!) and bolts of all sizes. As a teacher I love Gratnalls' trays. Brilliant in every way and I still use single ones for keeping project parts together. Currently I just don't have the space for a good pillar drill, lathe etc. Wish I did but I kind of like my less technical way of doing things! What are we to do with all this stuff when our time is up. My partner would have no idea. Just hope my son takes and interest! As far as the steel rule goes. I have had it for over 20 years. It does not have a name on it but I think we had sets of them in the school's DT department if I remember rightly. I had lots of wooden metre rules in the lab too and some of them were ridiculously badly marked. The kids were always struggling to read them correctly. Shame comments do not allow the posting of photos as it would be interesting to share ideas and workshop layouts. BTW seems to be far from CAVOK here and permanently 8 Oktas! Good to hear from you as ever and do keep the comments coming!

[@fburton8]

What an interesting project! I suspect fewer LEDs would be needed here in Scotland. A raindrop counter on the other hand...

[@AnthonyFrancisJones]

Yes! Good point! You have got me thinking... should I build one that measures the depth of water in my rain gauge in binary. Might get more use!

[@kallewirsch2263]

Your 13 hour versus 15 hour deviation may also come from the way you generate the clock signal. It is pretty hard to calibrate a RC unit to a specific frequency. The capacitor will have deviations from the printed value and worse the whole thing is temperature dependend. Even with a quarz crystal you will not get the exact frequency printed on the crystal, which will eg. lead to deviations in the range of 20 to 30 seconds in a week. You may use dedicated 32767 Hz crystals which are cut in a way to keep that frequency as close as possible but even they need to be trimmed by using capacitors at the crystal to give long time stabililty. Watchmakers will adjust those to the typical temperature also, as there is a temperature difference in between a wrist watch and a watch hanging on the wall. One very easy way to get a pretty good, long time stable clock is to use the power grid as a clock source. The frequency (in Europe 50Hz) is actually tracked precisely and regulated to give a stable clock signal. Over the course of some month it may run a little slow or fast but then it is regulated to counteract that and bring the difference to eg. an atomic clock back to zero. In former times this signal was used to drive the clocks at railway stations but this precision is also needed to synchronize the generators in power plants. In summary: creating a stable, precise clock signal is not an easy task, since even small deviations will add up over time pretty fast. If your 13 to 15 hour deviation bothers you (I understand that this is more of a fun project then some precise measuring equipment), the first thing I would do is check the clock source by letting it count for a few days uninterrupted and compare the number of ticks with the offical time.

[@AnthonyFrancisJones]

Thanks for the excellent comment (more than just a comment but that's what KZbin call them!) Yes, agreed, this is just a bit of a fun project as you say. As you say the 15% error could come from so many places, from temp changes to component tolerances (though I would be surprised if these change that much - rather being fixed in value). I think the main thing is down to my calibration of the LDR part of the circuit. It only points in one direction and through the window at that! With a clear sky I notice our solar panels are never as effective as when there is scattered light! The 555 calibration was only done over a short time period too. I am tempted to adjust the trim pot to see if I can get a better setting but I guess as soon as I knock the LDR board out of place it will all change again! Just a bit of fun but I enjoyed the design and build! Now, I really should buy a caesium clock... Well, I'm not Curious Marc!

[@leonhardtkristensen4093]

Yes it may be an interesting thing to make when learning about electronics. A number of other comments have criticised your work but I don't think you clamed it to be perfect. It is probably impossible to make anything perfect so don't worry about it. If you want to make it better then yes it is possible to make the timing circuit more accurate by using a crystal setup and keep it temperature controlled. The supply voltage also has to be controlled. The sunlight control could of cause also be made more accurate say by using an operational amplifier to make a sharp counting cut off but in all occasions any little bit of cloud in the sky will change the amount of sunlight/daylight that you get to your sensor and there fore make it less accurate. You can really only count how much time the light has a certain intensity. In stead of digital time counting you could have made it to indicate in hours, minutes and seconds. It would be slightly more complicated but there are components that do most of it. I have an education as an electronic design engineer so of cause it isn't very complicated for me. Unfortunately I never really worked as one but I have once designed a circuit that measured the output voltage of a transformer and then adjusted the input to the transformer by switching input setting on the transformer to achieve a constant 240V output. I did it by having it counting up and down. If the output was a little high it counted a step down and low counted up. It was designed for compensating for varying input voltages. It was never put into production as noise from the output could blow up the triac's that where used for the switching. Triac's where the only thing available that could handle the current 35 years ago. The factory I designed it for decided to copy some body else's design in stead. They just measured input voltage and then adjusted so that the output voltage would be close to correct but that doesn't compensate for varying loads on the output. My design did.

[@AnthonyFrancisJones]

Leonardt, thanks very much for this. Yes you are totally right. I am just sharing some of my ideas at builds to encourage those that are beginners to go that little bit further and think about what they could design and build as a device without worrying too much about its final functionality. As well to build something without resorting to the likes of Arduinos and code to solve problems. As a kid I knew what most of the chips did but did not really visualise how to string them together to do things. Building like I show now helps to get this modular idea across. It's just a bit of fun really! You understand that fully! I do, however, value all the wonderful comments I have had including constructive criticism and how one could improve the designs. They are really interesting too. Thanks for sharing your experiences. Always great to hear other people's stories and experiences. That's what makes it all worthwhile!