Desktop clock implementation, which indicator to choose? How to breed a board and where to order it from?

Question

Do it yourself

Good day. I want to make a desk clock instead of the usual ticking desk clock.
The implementation of the clock scares me. It is necessary to solve a number of problems:
1) Indication with brightness control for lighting (at night in order to save charge, well, so that it does not make your eyes callous). There are options with a light sensor (the easiest option is a photoresistor, I think so + special digital microcircuits so that the ADC is not turned on). Question How else to make it (indication) more pleasant? light filters? Or sold separately?
2) Autonomy. Without this point, the meaning of the clock is lost. I need a fully autonomous watch that will last at least a few months without recharging / changing the source. food
3) Corps. Here is the main tug for me. I have little idea how to make a case so that there is a pleasant indication, and not "bare" seven-segmenters, convenient connectors and buttons. I saw a minimal case. Just behind the LED indicators placed a board. Very minimalistic, but then there are strict requirements for dimensions.
4) Empirically, I found that poisoning boards at home is not mine. LUT has never turned out acceptable for me, or rather it turned out terribly (as soon as I didn’t try it). Hence the question of how much it will cost to manufacture a simple board for this project and where to order it from. Well, in what environment to breed ( sprintlayout exclude)
By priority, I'm interested in:
1) Autonomy. Those. the clock should work (indication, not the real time clock module itself) without being connected to the DC network. For a while, they should work completely on their own. I think at least a week / month
2) Interface - where to put the clock (body), and how they will look from the user's side (body / display dimensions). How to calculate it, fixings, etc. Where and how to make a case with minimal waste (time / money)
3) Ordering the board and its wiring
Shl. It's easier to buy, of course, but that's not the point. At this stage, I am considering the feasibility of the project. Because so far the idea is lame. On the Internet, only the lazy did not make watches, even designers sell them. But then again, I don't want to. Moreover, in many of them the problem of autonomy has not been solved (constructors).

Answer 1

1. Of course, it depends on what you design there or have already designed. Iron, indication method and wiring itself. Consumption you can roughly calculate by what you will use, in fact.
If nothing is used, then let's look at the statistics on ready-made controllers and let's not go far, let's take a pop, but workable Arduino and see that the Mini Pro and Nano have a consumption somewhere within 100mW ( currents of the order of 20mA ) .. well, that is. e. almost nothing and it is worth striving for the same result ( looking through the Freeduino schematics and specifications, for example). So, your seven-segment indicator will probably eat even more than the controller itself. There is an idea to replace it with an LCD with optional backlighting in the dark. Moreover, the backlight can be tied to the clock in advance. The photoresistor can also be tied to the clock so that it does not “stare” at the light during the day just like that. Then at night, if the light is on around, the backlight will not turn on, but turn on in the dark. Here again, you can stick a trick and put a sound sensor that will turn on the backlight only if it is dark and you ask for it, creating noise ( there is nothing for it to burn in the dark if no one needs it ). The inclusion of the sensor itself should also be tied up for a while - let it listen at night, it is not necessary during the day.
And of course there is such a thing: https://openhomeautomation.net/arduino-battery/
But in general, I recommend running from seven-segment and just LED indicators. Just in terms of logic - how many times do you look at your watch a day? Well 5..well 7 times. And they burn constantly and guzzle energy, in fact, into nothing. So .. you need to figure out how it will be convenient for you to highlight them as needed. The same e-ink, which will change the numbers only on demand, is quite good.
2. Another different thought. Down with the body - sew into a wall, table or door frame. Any external housings are closely related to your cutting capabilities. If you do not have access to turn-mill processing for a penny ( not to mention the casting of thin-walled cases), then you need to suffer with what is at hand. Check the shelves for small boxes. And then get attached to her.
As an option for a good approach to all the same remote housing - a piece of aluminum pipe D30-35, thin-walled. Inside there is a plastic lodgement for the board ( which, of course, must be immediately designed to fit the size ), made from improvised trash such as pieces from the printer, or even simply put on mounting plastic. Display the indicator on a cylindrical surface. The window can be made with a drill and a file. Well .. if you sketch very quickly, then this:
Bring the connectors and buttons either back or to the end. Bottom rubber blotch on the glue, so as not to ride and hi-tech in all its glory. If not laziness, then you can use a stainless steel pipe - it shines better .. but you won’t like drilling it.
There is nothing special to count here. There is no heat dissipation here .. only the dimensions of the board .. which are taken during design and wiring. And the mounting holes on the board are also placed at the same time, where there is space left. And it is possible without them, in fact.
Therefore, it is necessary to breed the board and think about the case in parallel and at the same time. They divorced the board - it doesn’t fit: we are thinking how to shove it in and how it will be easier - to separate it differently or change the case. These are all questions for the designer in the process of creating a device, i.e. to you in the near future. Work out the composition on the components on the breadboard, work out the design of the body on paper .. and then connect it together by filing here and there.
3.And here there are difficulties. Well, i.e. you can develop and distribute the board wherever it is convenient and where you know how. I've been comfortable with OrCAD since my college days. I don't do it, but if I need to design something by circuits and breed, then I will take it. Just because I know him. But ordering is difficult. A single and small one will cost a penny, even if they do it somewhere in your city. Those. here you have a choice - take a ready-made controller like Pro Mini and not be smart with the boards, but do software, or try to create yourself in order to master the "magic" , or look for a private trader. In the latter case, you can rush to the electronics forums and ask who can in the appropriate branches. You should have a sufficient idea of ​​what you need with you (dimension, track width, installation method), and ideally a finished model.

Answer 2

autonomy

the ionistor for 100 farads, in theory, should not have time to discharge during the night, and during the day it can be charged from solar panels.
If you're sick of building your own bike, you can add something out of the ordinary, like a fuel cell as a power source school-collection.edu.ru/catalog/res/7538961d-d857...
or a Stirling engine running from a candle
Take a dry block of more or less hard wood, give it a beautiful shape, cut it in half and remove the insides. It is not necessary to glue, the main thing is to polish it beautifully in the end. If you need to minimize the time, then it is better to use some kind of retro device as a case, for example, a gramophone or a video cassette.
make a hanging installation. Any interference for your task is of little interest, but everything will be compact. In addition, you can choose the desired shape of the insides.

Answer 3

advice. Before thinking about the case and the board, make a prototype on a solderless assembly. Solve (try to find a solution) the issue of autonomy.
you can not make a fee - there are large installations. Want to be really cool
here and there . But get ready, one board, and even not small sizes, can cost cheerfully. It may well turn out that ordering a LUT for someone is cheaper.