Jump to content
Premium Aquatics Aquarium Supplies

The Next Step in DIY LED Lighting


evilc66

Recommended Posts

I know that from a user point of view, programming the controller from a user friendly GUI program with check boxes, etc would be the preferred way. I'm all of that user friendliness. However, keep in mind that the controller cannot accept C or assembly code without converting it to hex codes. You'd need a unique compiler for that. Furthermore, having a GUI auto generate C/assembly code and work is a huge feat in itself. What I'm saying is it's very impractical to make a GUI that generates codes and compiles to be uploaded to the controller. Programming is unavoidable, I think. The good thing is, even if you're a beginner and requires help, people in this forum would be able to help you because the platform would be the same for everybody.

 

Awesome features:

individual color intensity adjustment from 0-100% (adjustable K value)

an adjustable dusk/dawn duration, say 2hrs sunrise, etc.

moon cycles would be just awesome

overall LED system intensity adjustable from 0-100% while keeping previous setting (K rating etc... good for acclimating purposes)

I like the LCD idea. Simplifies alot of programming work for the average user.

Link to comment
  • Replies 71
  • Created
  • Last Reply

I think something that would be cool would be to leave it plugged into the computer and be able to display current data such as the temperature of the unit, # of lights on, level of dimming or any other setting currently applied.

Link to comment

Hmm, jewbilee's comments shed new light, lol. If there's a computer program that can display numerous data, potentially, it can also set data settings. I went into too much detail w/ my other post ranting about a GUI to spit out code and compile it, blah blah blah. But if you look at it like jewbilee did, it's definitely possible. Just have to make sure that the settings like dimming level, intensity, etc is stored in a flash memory of something.

 

<3 moar ideas. It gives you different perspective of things.

Link to comment
I know that from a user point of view, programming the controller from a user friendly GUI program with check boxes, etc would be the preferred way. I'm all of that user friendliness. However, keep in mind that the controller cannot accept C or assembly code without converting it to hex codes. You'd need a unique compiler for that. Furthermore, having a GUI auto generate C/assembly code and work is a huge feat in itself. What I'm saying is it's very impractical to make a GUI that generates codes and compiles to be uploaded to the controller. Programming is unavoidable, I think. The good thing is, even if you're a beginner and requires help, people in this forum would be able to help you because the platform would be the same for everybody.

 

Awesome features:

individual color intensity adjustment from 0-100% (adjustable K value)

an adjustable dusk/dawn duration, say 2hrs sunrise, etc.

moon cycles would be just awesome

overall LED system intensity adjustable from 0-100% while keeping previous setting (K rating etc... good for acclimating purposes)

I like the LCD idea. Simplifies alot of programming work for the average user.

 

I think you are over complicating things here. There are many platforms aout there right now with simple to use programming languages with free compilers. They also have the power to take straight C code when the user gets to that point, or the advanced user wants to dive a little deeper. I'm not talking about reinventing the wheel here.

 

A check-box style GUI would be nice, but kinda falls out of the scope of the controller. With and LCD and a keypad, you should be able to adjust any parameters necessary. There are other means of hooking up to a computer that do not require additional programming support. Due to the open source nature of this project, a GUI might be something that we end up leaving to the community to develop.

 

I like the idea that many of you have for much more finite control of each LED, but keep in mind that this adds an enormous amount of hardware, and cost. I think a more practical application would be to simulate dawn/dusk and moonphases by groups. It would be up to the end user to define how the groups are oriented.

 

<3 moar ideas. It gives you different perspective of things.

 

No doubt! I'm sure there will be tons of ideas that we haven't even touched on yet.

Link to comment

I don't know, but I still think we should not revolving around the idea that people are going to learn and write codes themselves, especially when we are talking about writing code for IC level hardware, not the regular "computer programs" or "website programming", it's a total different ball game. For someone who have done it before or have experience might think it's not hard at all, but for someone who is a bit of computer challenged let alone programming challenged, this is really really a daunting task. Maybe you already forgot the time when you first learn a computer language, but for someone who have not touch anything like that before, it will not be a good feeling, it's like if I go to learn Russian or Farsi right now, I would probably kill myself before I can speak a complete sentence, or maybe I am just stoopid. But honestly, how many people out there are really programming nerds? and not trying to offend anyone, I have seen a lot of Comp Sci or Comp E graduates barely have clues about programming, or can't operate a computer without a mouse, sad.

 

anyways, carry on ;)

Link to comment

I never intended any of my comments to sound like you had to learn a programming language to use this system. For anyone who wants to just "use" the system, all the features will be accessable. What I was stating was that we should use a platform that is easy enough for those who want to try their hand at programming the system, not that they have to.

Link to comment

Sweet,

Great idea guys. I'm working on a similar controller for my diy led setup. I'll be using a arduino, RTC module and a LCD to prototype. Hope to move to something smaller for a controller for the final version. So far I have planned these functions for the first version:

set time/date

set white on/off time

set blue on/off time

set white level

set blue level

set ramp up/down time

 

Once I get that rolling I would like to add a few other things.

 

There has been some great stuff mentioned so far!

Link to comment

The progressive fade idea is a great one in theory, but realisticly the hardware and wiremesss would be a nightmare. I'm good at coming up with good theoretical ideas...lol.

 

Fade on/off in different groups would be much easier to accomplish. I see alot of great ideas here, but many of them seem to go beyond a basic entry level controller...then again, it seems we all like to think waaaaay outside the box.

Link to comment

But this is half the point. You are right that most of these features may not make it into the base unit, but it's not to say they can't be added later on down the road.

Link to comment
But this is half the point. You are right that most of these features may not make it into the base unit, but it's not to say they can't be added later on down the road.

 

Yeah that is what makes open source so awesome. People start using the basic version, then mod it, and then add these mods to the site. Then other people can pick and choose what they what to add to their model. Then they create new stuff and the cycle continues.

Link to comment

oh no no, evilc66, my comments wasn't directly at you, it's for other people who are asking for "what kind of language", "GUI interface for easier programming" and all that.

 

All I am trying to say is, don't worry about all the programming bit if you don't want to touch it, and also I wouldn't pursuade anybody to touch it also, coz it might not be as easy as you think it is. moreover, the LCD screen and button should take care 90-95% of all the functions you will ever need. so don't you worry about the open source and all that, coz you might never need to use it if we do this right. and that's is exactly what this thread is about, to keep end user programming need to a minimal.

Link to comment

Evil what is a good programming language to learn that is somewhat easy ?

I might try to ebay a book to learn in my spare time.

Something very usable,that I could eventually apply to this project.

Link to comment

Well, I wasn't listing it for the price :)

 

Any programming language will take a while to learn, but C is so versatile in many applications, and lends itself to other programming languages (like Java).

Link to comment
Psychographic

Seeing as you are asking the Average Joe, their intake on this. Here are my suggestions.

 

First and foremost, electrical, other than your basics, is very confusing. I'm a great DIY'er but once you get past the basics of electrical, I'm lost. So it has to be able to be usable by the average person who can read a manual.

 

My opinion of basic features,

 

PLUG and PLAY! or as close to it as possible

Built in timers for on- off

Battery back up for keeping the settings stored

The ability to expand and or upgrade.

Support, if you can't get help with it, it's useless. I think if this caught on, it could be overwhelming trying to help everyone. Remember, this comes to you guys somewhat naturally, to the rest of us, it's confusing. (just something to ponder)

 

 

Now if you want to go a step up.

 

Sunrise/Sunset fades

 

Lunar Cycle

 

Lightning Storm ( I think it would be cool, I'm not so sure my fish would like it)

Link to comment
So it has to be able to be usable by the average person who can read a manual.

 

sorry if you read the manual, you are a step above the average person already. :P can't help it :D

 

carry on...

Link to comment

All the settings will be made easily accessable. Thats first and foremost. A product that is hard to use is useful to no-one.

 

If we do decide to make the hardware DIY (buy a kit) we will try and make it as easy as possible to put together. Through hole components and the like. Easy to solder. This is a decision that is still to be made.

 

So far it seems that the bare bones basic feature that you want to see are:

 

Built in timers

Battery back up settings

Seperate adjustability of white and blue

Modularity

Ease of use

 

Would giving you the option of not having and LCD and keypad be something you would consider for a really cheap, super bare bones system? You would still be able to change settings by hooking it up to a computer. Or would you consider this to be an absolute mandatory feature?

Link to comment
sorry if you read the manual, you are a step above the average person already. :P can't help it :D

 

carry on...

 

As Chick Mcgee says on the Bob & Tom show. "Manuals are just some other guys opinion". :P

 

I am so excited to see what you guys can come up with. I would love to see what these bad boys can do. I have one question. Do the l.e.ds create the shimmer that halides do?

Link to comment
I like the idea that many of you have for much more finite control of each LED, but keep in mind that this adds an enormous amount of hardware, and cost. I think a more practical application would be to simulate dawn/dusk and moonphases by groups. It would be up to the end user to define how the groups are oriented.

I dont think you need finite control of each led to get the same effect. You would just need finite control of the leds in one group. For example 6 leds on is the same as 1 led on plus one bank of 5 leds. also I would think you need this for lunar lights etc. I would not want a whole bank on for lunar, just one probably not at full power.

 

Things I would also like:

Control of other devices, fans(temp control of tank, PWM?) and heater. Pumps is probably to complicated.

fuge compatibility/expansion.

I would prefer ethernet setup over lcd.

Link to comment

Archived

This topic is now archived and is closed to further replies.

  • Recommended Discussions


×
×
  • Create New...