C library for Sun/Moon effects

[abhi_123]

My Light are totally off @ 4 p.m. local time.

[rufessor]

Ok.. this is starting to help... but I am still a little unsure whats happening. The difference between your sunrise and sunset values is pretty much exactly what your day length should be today (around 49,000 seconds or 13+ hours).

I *think* that what is happening is that the way I have this written is that it may actually be ending up calculating the sunrise and sunset values for the day before today... then because all the values are unsigned long, when I do the math to convert to time since midnight its going to be negative (i.e. prior to today) but those variables are NOT allowed to be negative, so "roll over" giving you wild elapsed times for sunrise and sunset, but they roll predictably so I can still tell your day length is correct. I changed the order in which things are calculated and actually changed it so the entire program should now be running in GMT rather than local time and I leave rise and set in GMT (it was the conversion to local that was rolling them).... this may work now... I hope.

I am posting this without having tested it on my system but that now seems not so useful as I was getting correct times on my system so I want you to test this...

If one of you beats the other to it and posts good results I am confident it will work for everyone, but please if you have time load and post your results as well so we can see it working in three different hemispheres of the globe. If it bombs for one person, its still worth doing it again as something about time and local time v.s. GMT is causing problems since I get it to work for me even for your lat lon pairs but only during certain times of the day.. then I think the GMT conversion rolls it to the next day (as my time changes) and I start getting values you posted... so were REALLY CLOSE. REMEMBER TO CHANGE THE LAT LON TO YOUR VALUES_ CHANGE GMTOFFSET TO YOUR POSITIVE VALUES (7200 South Africa, 19800 India) and to be sure the DST is OFF or false.

Please copy paste over the first half of CalcSun() as you did... with this code... you might be able to simply edit as there are only three changes (I think) but to be safe just copy paste.

Code: Select all

 newDay=((now()-(946684800+hours+minutes))-GMToffset);//GMT time at midnight in seconds from year 2K for todays date

        SecInput=newDay;//Here we want to convert our seconds to GMT so change sign of offset
        rise=SecInput;// 
        set=SecInput;//
        
       //convert degree minute sec to decimal latitude longitude into required units for rise/set calculations
       latitude=dmsToSeconds(40,44,15);//United States of America- Salt Lake City, local time is -7 hours GMT (DST enabled is OK) 
       longitude=dmsToSeconds(-111,49,25);
      
       //Calculate rise time and set time using Epherma Library   
       SunRise(&rise);//call to Epherma Library using reference- memory is modified by Epherma
       SunSet(&set);//Call to Epherma library using reference- memory position is modified by Epherma
       Serial.print("local now() value is=");
       Serial.println(now());
       Serial.print("raw rise value=");
       Serial.println(rise);
       Serial.print("raw set value is=");
       Serial.println(set);
       
       rise=(rise-newDay);//this should now be in seconds from midnight GMT for sunrise at our latitude
       set=(set-newDay);//as above but sunset

Then... please look at the bottom of the loop and find where I set elapsedTime= ((now()-etc....)

and change it to this including the two Serial.print lines below elapsedTime=

Code: Select all

elapsedTime=(((now()-946684800)-newDay)-GMToffset);//we are now in GMT time... or damn well should be
Serial.print ("Elapsed Time=");
Serial.println(elapsedTime);

The new code for cal sun I posted above also has active serial debugging. When you load, please start the serial monitor and copy and paste the results here.... it will keep putting elapsed time= xxxxx forever... just copy and paste so I can see the other data and a few lines of the elapsed time=xxxxx just to see the value change (it may require you post 20-30 repeats before the value actually changes.... don't worry about it just copy and paste.

At that point, I would just reload your working fix with GMT=0 and Greenwich lat lon. If your on the other hand seeing values for the sunrise and sunset times that look like about 20,000 (for the Elapsed seconds since midnight field) and something like 69000 ish for sunset then its working and fixed and you can just comment out the serial.print statements and you should be totally good to go. At this point I am not seeing any issues with the cloud/storms and my program is stable. So far as I know this is the last issue.

We can get into tuning the frequency of storms/lightning and length as individual changes if people want to... its easy.

[abhi_123]

Alright. I have already edited calsun portion once do i need tovedit once again? Also wat to do for Secinput?

[abhi_123]

One small thing if it can help us. When i am using my local lat/long only one channel is working that too on 21% where as when m using lat/long of Greenwich all my channels are working.

[rufessor]

I had deleted a post with a response to your prior questions... don't worry about what light is on... I am fairly sure the old version of code was overflowing variables and that you light was on at all was totally random...

Don't worry about SecInput... its just an intermediate variable and not causing any issues...

I made a few changes to what I had posted please ignore prior posts because I think this will now work for everyone..

in your loop near the end there is a line that sets elapsedTime=.....

please change this line to
elapsedTime=now()-(946684800+GMToffset);//we are now in GMT time... or damn well should be

Then... Please just replace the top part of CalSun()

copy and paste all of this over and including this as the start line for the paste (select all the stuff including these lines, delete it and paste in the code below.

void CalSun(){

up to and including this line
//*********************UNLESS YOUR TANK LIGHTING IS IDENTICAL etc et etc...


this should work... its working for me... and when I load your lat lon pairs with GMT offset my tank is dark (it should be) and I am getting correct day lengths out of the calculation. There are actually only deletions to this... so if you feel like it just look at yours and edit to make it look like this... I post the whole thing because its small and easier to get right. This supersedes any edits you previously made and will overwrite them when you copy and paste. I don't even have serial debugging in it live right now because its kinda making sense and should work. No guarantees... but its working on my setup for any lat lon combination

if your copy pasting here is the code

Code: Select all

void CalSun(){
   //Serial.println("CalSun Run Now");
   trigger=false;
    // Every day at midnight, we calculate rise, set, time of highest sun, moon phase, or we do this if system has reset
        if (ApplyDST==true){
            CalcDST(day(),month(),weekday());
        }

       unsigned long SecInput;//intermediate variable to set rise and set input times
       unsigned long hours;//avoid casting and use UL
       unsigned long minutes;//avoid casting and use UL
       time_t t=now();
       hours=hour(t);
       minutes=minute(t);
        
       //Using time library from Arduino.time_t now(); returns seconds since 1970 in local time with NO DST correction
       //Arduino Time.h Library states #seconds to subtract to bring to Jan 1 2000 as origin for time is 
       //#define SECS_YR_2000  (946684800) the time at the start of y2k
        hours=(hours*3600);//convert to seconds
        minutes=(minutes*60);//
        newDay=((now()-(946684800+hours+minutes))-GMToffset);//GMT time at midnight in seconds from year 2K for todays date

        SecInput=newDay;//Here we want to convert our seconds to GMT so change sign of offset
        rise=SecInput;// 
        set=SecInput;//
        
       //convert degree minute sec to decimal latitude longitude into required units for rise/set calculations
       latitude=dmsToSeconds(40,44,15);//United States of America- Salt Lake City, local time is -7 hours GMT (DST enabled is OK) 
       longitude=dmsToSeconds(-111,49,25);
       
       //Calculate rise time and set time using Epherma Library   
       SunRise(&rise);//call to Epherma Library using reference- memory is modified by Epherma
       SunSet(&set);//Call to Epherma library using reference- memory position is modified by Epherm
       /*Serial.print("Sunrise seconds=");
       Serial.println(rise);
       Serial.print("Sunset seconds is=");
       Serial.println(set);
       Serial.print("Day Length in seconds is=");
       Serial.println((set-rise));*/
      
       if (isDST==true){
         rise+=3600;//during DST we move our clocks ahead so rise is actually 1 hour later than strict GMT offset correction
         set+=3600;
       }
        //*********************UNLESS YOUR TANK LIGHTING IS IDENTICAL IN EVERY WAY TO MINE----YOU NEED TO CHANGE THESE VALUES*************

[rufessor]

Just FYI-

My tank lights set an hour late... because I had DST enabled.... but somehow with the new time system I am running I don't need to track DST anymore... which kinda makes sense as I have converted all times to GMT. So I can confirm a day works appropriately with the new code...

Once we get confirmation that its working in other hemispheres... were pretty much good to go... no changes planned really but for the comm to the main to get the ability to display current day length remaining, storm cloud etc...

Let me know when you can confirm. I believe its good- the way time is being tracked now is much less manipulated- its nearly just system clock time corrected to start at the year 2000- then corrected to be in GMT... thats it. Then I don't need to convert the rise and set times that the lat lon calks spit back. I just use them as is... since I know they are spitting out correct day lengths for your lat lon, I am willing to bet the whole thing is now working. I have been stable for a while now with no stuck storms etc... things seem to be on a firm footing.

[rufessor]

So it occurs to me... when I started this I thought I really badly needed to correct for Day Light Savings time to get the clock to sync and have the sun rise and set at the right times... well- that was not thought all the way through. There is ZERO need of a daylight savings time correction. The sun kinda ignores (you might say) our clocks... so to reduce code size and clean it up. I deleted day light savings time. If anyone wants a compact bit of code to determine it... Its available.

And- in finally (I think. still awaiting a second party to confirm this) fixing the rise and set, I was also able to delete a few global variables freeing a bit more memory and saving yet more space in the code (a few lines but hey...).

And then- I wanted to try to see what the strikes would look like if you could choose to use any or all of your channels for the strike. So I added in a Strike Channel array that lets you determine which channels will be used for the strike effect- its a bit more robust looking if they all strike... now its up to you to configure. Its just like WChannel- an array with 8 positions, put a 1 in the array positions of the channels you want to use to strike- positions 0-5 are the PWM channels, 6 and 7 are for the main controller when thats up and running through comm (working on that).

If your interested in trying this updated code- PM and I will send it to you. I don't want to post yet more code until we know its all working (need confirmation of rise and set in hemispheres other than mine). Else, I am playing around with strike durations to get it to look as good as possible.

Thanks to everyone for the help along the way. It was totally worth it! I love the effects. Its just subtle enough not to scream cloud or storm.. but it always catches my eye and I find myself watching the tank a bit more.

Working on continued troubleshooting with help from others RE sunrise set time outside of USA.

[abhi_123]

Hi i am getting this error. expected primary-expression before 'int'

error is on int Choffset[]={

[rufessor]

Opps... it looks like you are missing a bracket or possibly a semicolon at the end of a line... probably a copy paste error- easiest is to send you code... so I sent you code via PM to recopy and paste and I covered a bigger area so you should fix the error.

Based upon some serial debugging JNewh (sorry... forgot spelling!) has been conducting- this is looking good but for a possible transcription error. When I get back the next set of tests from him I will post and confirm that this is in fact working or that XX needs to be changed. He is getting accurate calculation of my day length as well as his (two different hemispheres) as well as for egypt (now three hemispheres) and his program is spitting out rise and set times in terms of the elapsed day length that are 1 hour ahead of what they should be for my house (but I think this is just an error in correcting time across the globe). I am awaiting a final run for his region with some new debugging to absolutely confirm its working for him... will post when I get this.

THANKS to you both for being the guinea pig(s). Too bad it would be prohibitively expensive to meet for a beer!

[JNieuwenhuizen]

Could be a Guiness World Record for most expensive beer ever

Plane ticket+Visa costs+transport+accomodation+beer

Anyways, seems like all is well on my side, code is working! I even get no-moon currently, and we are in a no-moon cycle for the next 7 days

I will confirm tommorow if sun rised, but it sure set correct today!

[rufessor]

You ROCK!

THANKS for sending me the serial debugging information. It confirmed what I had hoped and let us figure out what the issue was. Please PM me your private email address so I can send you the most recent complete .ino which has the updates for the strike generator to enable you to choose which channels are used to strike (and that is or can be different from which dim during a storm). It also has refined internal settings for strike duration and strike interval timing so its looking a bit more realistic. Read on if you not in a hurry.. that code will be posted but just nor immediately.

Just from a purely analytical perspective, the relative cost of the beer on such a trip is zero... so its actually DIRT cheap to drink that beer... from a certain perspective

Ahabi... your email contains the code that is working for SouthAfrica... would bet your now good (I changed it to your lat lon). Thanks for your help as well. I guess I probably should post this entire code .... but probably going to start a new thread just with instructions and code. Any mods will continue here as a development thread (such as comm etc). But its 20 pages to wade through of mostly garbage, bad code, and struggles... now that this period appears to be over- best to make it easy. Look here for continued development efforts but look for the new post to direct your friends for download and config of code.

[abhi_123]

hi the code you mailed me is not working. none of my channels are lighting. its 8:52 am local time here.

[rufessor]

At this point the only reasonable conclusion is that I made an error in trying to set it up for you. I will look over what I have and send you another.... But I really cannot be held responsible for knowing your latitude and longitude and GMT offset among other things.. so for this to work your going to have to look over the code, its your tank, your time zone etc... Please double and triple check my work.

I really want to wait to confirm sunrise in South Africa tank that had a successful sunset... then I will look over code, be sure my tank rises (I was just playing with things somewhere else in the code... but...) and then if my tank is running, and the South Africa tank rises with the new day... my code is probably beyond a doubt good. It would be nearly impossible to get it correct in three hemispheres but somehow get the wrong answer in a fourth, and I have tested your values and get the correct answer... so please check the arrays that are customized. Not to belabor... but its VERY easy to screw this up if you miss a - sign your cooked-

Wait for new code until tomorrow my time probably morning for me so later day/evening for you.

[rufessor]

oh... as far as the moon lights... I have none this evening- there is no moon. So all is looking well!

[abhi_123]

i am also eagerly waiting for the sun to rise in South Africa. will be cross checking the code to find any defects in it.

[abhi_123]

Bump

[rufessor]

Sorry.. not been paying attention much this weekend.

I was confused a bit by the result in SA... using the serial debugging things were looking good but we discovered a lat/lon error and then we got it working. What I am now unsure about, given that the new code I sent is not working... is if they were running the code with the serial debugging active, in which case they were using the last (I tried three) version of feeding the sun rise and set functions the current time (I tried current clock in GMT, Midnight of day in GMT, and Midnight of day in local time. It was looking like the first two were equivalent (this is more or less how it should work) but then when they reported back to me that the new code was no longer giving rise times that were accurate.. I had used the first calculation (midnight corrected GMT time for the day) where as the serial debugging version would have been using the last calculated value which was current time of day in GMT (which should only matter on reset of board as all other calculations would be at midnight. So. I need to know a few things from the SA set up.

Did you run it overnight with the code we were getting serial debugging results and see a successful sunrise without having unplugged it, or plugged in the serial cable, or started the serial monitor from arduino (all would reset the program and prompt a slightly different calculation). If that is the case, and you saw a successful sunrise the next day, but when you loaded the code I sent you (without serial debugging but using midnight corrected GMT of the current day) and then did not get a successful rise... did you leave that code on overnight and see it fail the next day?

The reason I need this, is it may be that the time it is feeding the algorithm when the day rolls over is somehow wrong and its calculating sun rise and set for the prior day, whereas when we force a reset, its working in one of the versions. Because every version of this works for me over multiple days running and restarts etc etc... I don't know what to do other than to ask.

Can You post here the exact configuration that we thought worked (i.e. serial debugging code, set was fine, rise next am was good...) or... serial debugging set was fine, rise failed....

i can send everyone the code without the midnight correction (which is what I think was being used in SA because that was the serial debugging last line calculated) but I want to hear back first. This is getting ridiculous. It literally works no matter WHAT I feed it here... and i know its working for you too... because the day length is exactly correct but somehow its getting the elapsed time of day all goofed up... but its too simple to be math (its literally basically a single line of subtraction between two variables.... DUH... so its something else slightly more evil. I am going to PM someone who might be able to give a new eye... but please, post as much information as you can. I know the code you were running if you tell me its the serial debugging (I still have it).

[abhi_123]

Hi,

When i was testing the code on serial monitor with 57600 baud it is giving me no values either.

[rufessor]

I dont think you have any code with active serial debugging. This is normal. Serial should not be sending anything. I wiil
email you code if you want to reply with a copy and paste of the results from serial.

[abhi_123]

Hi.

I was just telling i dont want to debug it. All i want is to have sunrise & sunset.

[rufessor]

Ok... disregard email with .INO containing debugging...

If your on the other hand willing to help... just run it and copy and paste to an email the serial debugging output.

Hopefully this can be worked out, I am confused to say the least but the new serial debugging code contains EVERY single variable required in the calculation so perhaps I can see whats getting goofed up. I am betting on alien interference at this point.

If this next set is not informative, well... Roberto... you will be getting a PM

[JNieuwenhuizen]

LMAO!!! Alien interferance

[rufessor]

Small update.

I have to admit to having little insight into how we are getting the correct day lengths for any location on the globe- regardless of which set up is computing it... but somehow the conversion to real time rise and set is simply wrong. Its seems that no matter what I feed it, real GMT time, Local time, or any combination of processor clock and corrections to GMT... its getting correct day lengths but the rise and set times are simply not meshing with the input, they are often a little bit behind the input time, they are in fact mostly correct because if I push it and feed the function times that are behind or ahead of the current day the rise and set times change by 84600 or very close to that- which is exactly a day... so its advancing them correctly. I need to sit down and start looking at what the offset is and if its a fixed number or ratio or something to see if I can code in a "fix" I wish I could say I have a handle on this... but its really odd- It may be some time before I can offer help but I am looking at it.

[chort55]

I have been following for quite a while now, and often confused (coding isn't an area I have any knowledge in really, bought the controller from another guy who lied about setting it up for me and screwed me over royally....)

Just wondering though, are you setting LED's to follow like a sun path (I'm 99.99% sure you are, but want to make sure I have been following correctly lol)? As in, rise at say 6:00 am, and go down at 9:30pm like they are roughly here..... thats a LOOONG time to be running high powered LED's and would more then likely cook even the most light craving SPS, granted they would only be at 100% for a few hours I'd assume, but still seems like it may be a bit much. If I am misunderstanding this I am sorry.

Also, if they are "a little bit behind" could this possibly be just a timing issue with the internal clock on the module/ controller... or the timing is set from the day the code was created/ 1st uploaded so it sets it behind slightly compared to the current testing day since you made changes it "resets" the date/ time possibly??

I wish you all luck in figuring it out so I can give it a try. I'd love to help "get it right" but I would probably just screw things up more unfortunately.

[rufessor]

Hi and thanks for following-

Yes, I am programming the LED to follow a "sun like path" in that they start (you can modify this) at sunset and finish at sunrise. However, I am also calculating what the intensity of solar radiation would be every 3 seconds between sunrise and sunset with the MAX solar irradiation intensity being defined as being midway between sunrise and sunset- the LED are then programmed to run between 0-100% of their intensity set point (you set what you want their "noon" max intensity to be) throughout the day, this function is a cosine so it starts at basically zero at sunrise and then follows the cosine to 100% intensity at noon and then back to zero at sunrise.... so although it seems like a long time to run your LEDs, they are only on at full intensity for a few hours around noon. The cosine function modulates their output throughout the day, as close to the actual variation in sun intensity as I could get it.... thus its not really all that intense overall. For instance, I have a 3 ft long 2ft deep tank 18 inches wide, with 56 LEDs (48 Cree XP-G and XP-E and 8 True VIolet) and I have them very near to 100% output (but I am constraining them to about 80% max current on the drivers) and my corals actually colored up MORE when I started this program (its been about a month now) because there is actually LESS light than the standard 10-12 hour cycle of steady intensity. I was bleaching things before and was going to lower my intensity- then I started this program... and its all good.

[rufessor]

its 100% fixed. The error was very deep in my understanding of just exactly WHAT the system clock was referencing as time. I simply did not understand the relationship between the clock on our hand and the clock in the arduino processor. They both show the same time but the one in the processor actually thinks its on GMT time, i.e. it thinks that its in Greenwhich England (if you want to think of it that way) and that the prime meridian comes directly though your house... so There is absolutely NO NEED to correct for your actual GMT offset, the processor is in GMT its just that every processor from different parts of the world thinks that GMT midnight=local time midnight. So... this makes all my work to fix this, completely barking up the wrong tree. It also makes making it work very simple indeed.

New code will be posted by tomorrow. GMTOffset is now removed, you no longer have to care about time period, at all... all you need to do is plug in your Lat and Lon coords and it will work. I am 100% certain this is fixed, everything that was not working now makes 100% perfect sense as to why it was not working and will never work. In the end, this was a great error to have, I learned a lot- and Roberto was trying to tell me how this worked from the beginning, but I just could not conceptualize it as I was so hung up on time zones, DST, lat on, and GMToffsets I never really totally 100% checked and confirmed i knew what the system clock was doing..... as they say.. I ASSuMEd... and it made an Ass out of ME! Thats for sure!

[rufessor]

Lets give this a try.....

You will want to change the latitude and longitude lines to your region (REMEMBER TO USE THE CORRECT SIGN...).

This is the entire top part of the code up to this line in the CalSun function.. you will need to copy and paste in your settings. There IS NO GMToffset... it was not required and an error on my part to even include it in trying to fix this... Just copy over the set up section. The top part with all the ChMax etc etc and then the latitude and longitude which a grouped with ChOffset in the CalSun function... so anything you need to deal with is grouped into two segments. Remember if you copy and paste you old stuff... DELETE GMToffset its no longer in the code and you will get a compile error if you leave that line in.


these last lines are in the copy and paste below, but should be the same in your copy so copy up to them or whatever just dont go past it or duplicate them....

so you end here but look at the code segment... it contains these lines as a marker for you to see the end as continuous with your code. Copy everything from this to the top, delete and insert code below- then edit as appropriate and noted above. DELETE GMToffset from anything you copy from old code with that in it. Good luck

//Calculate rise and set times for all channels in equivlants to elapsed seconds from midnight today
//populate array for chRise and Set as well as chSlope for 0.5pi/ half day lenght for each channel from midday (asymmetric days are allowed)
float deltaY=1.570796327;//1/2 * pi as integer by scaling* 10^9 to fill UL
midDay=(((set-rise)/2)+rise);
long HalfDayLength=((set-rise)/2);
//Serial.print("MidDay");
//Serial.println(midDay);[/code]

Code: Select all

//By Matthew Hockin 2012.  
//SWFLTEK library functions, #includes, and #define were copied directly from the Epherma library
//Epherma (SWFLTEK.com) was written by Michael Rice- thanks Michael it works great. 
//If you copy from this (its all open source) please 
//acknowledge Michael for SWFLTEK library code (obviously labeled) or Matthew Hockin (for the rest).

#include <Time.h>
#include <Wire.h>
#include <OneWire.h>
#include <Time.h>
#include <DS1307RTC.h>
#include <avr/wdt.h>
//includes for SWFLTEK functions
#include <stdlib.h>
#include <math.h>


//***********************************ARRAYS YOU MUST MODIFY TO MAKE YOUR TANK SET UP WORK*****************************
//YOU MUST READ EVERY WORD IN THIS SECTION IN ORDER TO APPROPRIATELY CONFIGURE THIS PROGERAM- READ EVERY LINE BELOW///
byte ChMax[]={200,220,200,220,225,0,0,0};//Incremental value (Max-flicker) above flicker you want as max intensity (!!!!!!! Light Set Point is ChMax PLUS Flicker !!!!!!) 
byte flicker[]={31,30,31,30,28,0,0,0};//need to input actual values here for flicker point on all channels in PWM expansion box
boolean Wchannel[]={1,0,1,0,0,0,0,0}; //use 1 to designate white channel (i.e. off during storm and used for lightning).  Array corresponds to PWM channel 0-5 in order
//Array to give direction to dimming.  e.g. DimOrder[]={0,0,1,1,0,0} (cloud chase effects, just group channels you want to dim together during a cloud or storm 
//as either a 0 or a 1, i.e. all left side channels are 0 all right are 1 or all front are 0 all back are 1 or whatever
//(which is zero or 1 will change who dims first).  set them all to 0 if your tank has no left/right or front/back lights.
byte DimOrder[]={0,0,1,1,1,0,0,0};
//set all channel positions that you would like to use for the lightning strike effect to 1 (0-5 are PWM channels 6,7 are Main PWM outs)- and channels with a 0 are not used in strike
byte StrikeChannel[]={1,1,1,1,0,0,0,0};
//**********************************DONE CHANGING THINGS HERE BUT YOU MUST CHANGE ChOffset array IN CalcSUN function******


//defines for SWFLTEK functions
// arc-seconds per radian
#define _sec_rad 206264.806247096370813

// axial tilt of earth at epoch, in radians
#define _tilt 0.409092804222329

// tropical year in seconds... rounding error accumulates to 26 seconds by the year 2136
#define _tropical_year 31556925

// 'zenith' of rising (setting) sun in radians (360 - 2 * 90.833 degrees)
#define _zenith 3.11250383272322

//*******************GLOBAL VARIABLE DECLERATIONS FOR MHOCKIN Weather package*************************************
//Unless your planning on editing the program DO NOT CHANGE ANYTHING HERE
long latitude, longitude;
byte TrueIntensity[8];//array used to place hold final write values for PWM intensity setting
long elapsedTime;//used multiple places as elapsed since midnight
long newDay;
unsigned long rise;//time in seconds from the year 2000 (GMT) for sunrise
unsigned long set;//time in seconds from the year 2000 (GMT) for sunrise
long ChRiseSet[16];//times of rise and set for all 8 channels based upon offsets from calc rise and set values
float ChSlope[16];//slopes for 1/2 day calculations based upon time from offset to midday for channel 1-8
long CloudMaster[20];// Set up array to hold start and end times for clouds for the day-
long midDay;// exactly 1/2 way between rise and set, i.e. solar noon for latitudes <60 close enough for us... 
byte PWMports[] ={
    3,5,6,9,10,11};
byte ChannelValue[8];// Array to store output of insolaiton which may be modified and stored in TrueIntensity which is used to write to the PWM channels
unsigned long StrikeStart;//timer to keep track of strike sequence
int StrikeMaster[20];//Array to hold random strike pattern generated by weather array is sized to MAX needed given strike patter generator (8 strikes=16 positions)
byte StrikeNumber;//place to hold total number of strikes this sequence
boolean StrikeNow;//starts lightning strike sequence in loop state change made in weather/storm loop
byte StrikeCount;//Used to properly sequence strike sequence for delay between strikes
byte cmdnum=255;
byte datanum=255;
byte dow=0;//day of week
boolean trigger; //used a few places as a switch to ensure things only run 1x when needed trigger 2 is for daily reset of sunrise sunset
byte strikePattern, strikeTime;//used in Lightning() for timing of particular instance of strike 
boolean Cloud;// are we in a cloud interval on days that have clouds
boolean CloudToday;//set in CalcSun if randomization yields a day with clouds.
boolean IsStorm;// are we in a storm
byte CloudsTotal;// how many clouds today
long lastmillis;// variable to track millis to enable cloud and insolation loop restriction by time
boolean StormAdvance;//storm timer for light effect advance
boolean InsolationAdvance;//when true we recalculate light intensity during clear sky (every 3 seconds seems more than often enough)
byte counter;//used to track millis advance for insolation,cloud trigger
//****************************************
//END HEADER/Global Variable declaration//
//****************************************

//Setup
void setup(){
    Serial.begin(57600);
    Wire.begin(8);
    //Wire.onReceive(receiveEvent);
    //Wire.onRequest(requestEvent);

    pinMode(3,OUTPUT);
    pinMode(5,OUTPUT);
    pinMode(6,OUTPUT);
    pinMode(9,OUTPUT);
    pinMode(10,OUTPUT);
    pinMode(11,OUTPUT);
    wdt_enable(WDTO_1S);
    unsigned long seed=0, count=32;
    while (--count){
      seed = (seed<<1) | (analogRead(5)&1);
    }
      randomSeed(seed);//start random generator at a different point each time (not perfect but whatever its gonna be pretty damn random)
    setSyncProvider(RTC.get);   // the function to get the time from the RTC
    setSyncInterval(SECS_PER_HOUR);  // Changed to sync every hour.
    
    dow=0;//set Day Of Week (dow) to a 0 value which is impossible (day()=1-7)... so we trigger calcSun on restart 
    StrikeNow=false;//no lightning strike yet
    CloudToday=false;//set to no clouds so CalcSun can set correctly if should be true
    Cloud=false;//set cloud to false
    IsStorm=false;//set storm to false
    lastmillis=millis();//start our millis timer now
    counter=0;//used in weather for triggering a storm, triggering lightning in a storm.
    StrikeCount=0;//Number of lightning strikes in the sequence.. set to zero until initialized in sequence
}
//End Setup
//*********************************************************************************************************************************

//*********************************************************************************************************************************
//Loop
void loop(){ 
    elapsedTime=(now()-newDay);//Elapsed time is seconds from midnight of today- local processor time.
    wdt_reset();
    if (cmdnum!=255){
        ProcessCMD(cmdnum,datanum);    
        cmdnum=255;
        datanum=255;
    }
    
    if (dow!=day()){ //used to see that were in a new day and need to recalculate sunrise and sunset
      trigger=true;//on reset this will also be true as numbering is 1-31 max
      //Serial.println("We set trigger to true");
      dow=day();//now set dow to actual day numbering so that at midnight/new day we again calc sunrise sunset
    }
  
    //Use millis to enable tracking of time interval
    if ((millis()-lastmillis)>=100){
        lastmillis=millis();
        counter+=1;
        StormAdvance=true;
        //now a bunch of stuff that may or may not be true at the same time but that all needs to happen when its true
        if (counter==0){
          InsolationAdvance=true;//so that it runs on start up to provide light immediately 
        }  
        if (counter%30==0){
          InsolationAdvance=true;
          /*long dayleft=set-elapsedTime;
          Serial.print("Day remaining=");
          Serial.println(dayleft);
          Serial.print("elapsed time in seconds from midnight is =");
          Serial.println(elapsedTime);
          Serial.print("cloud=");
          Serial.println(Cloud);
          Serial.print("IsStorm=");
          Serial.println(IsStorm);*/
        }
        if (counter==210) counter=0; 
    }     

   if (trigger==true) CalSun();// calculate rise set, clouds.
   if (InsolationAdvance==true) Insolation();//calculate clear sky solar intensity as the day advances
   Weather();//run the weather overlay (cloud, storm)
   //check to see if were need to have a lightning strike
    if (StrikeNow==true){
       if ((millis()-StrikeStart)>=StrikeMaster[(StrikeCount*2)]){//check if time has passed the delay (position 0,2,4,6,8 etc in StrikeMaster)-StrikeCount is indexed up by 1 after each strike so we see positions 0,2,4,6,etc in sequence
          byte intensity;
          intensity=random(180,256);// this little bit should generate a randomly bright flash variation between the series of flashes in StrikeMaster
              for (byte b=0; b<6; b++){
                  if (StrikeChannel[b]==1) analogWrite(PWMports[b],intensity);// set all strike channels to intensity of strike
              }
          delay(StrikeMaster[((StrikeCount*2)+1)]);//index to +1 position in array from 0,2,4, etc to 1,3,5 etc
          StrikeCount++;//so that the next time we look at elapsed time were looking at the right array position
          if (StrikeCount==(StrikeNumber-1)){
            StrikeNow=false;
            StrikeCount=0;
          }
       }
    }
    for (byte a=0;a<6;a++){//using all prior mods to light intensity (Insolation-->Cloud-->Storm) lets make some light
      analogWrite(PWMports[a],TrueIntensity[a]);//dont change this to 8 to refelct array for channels.. we only have 6 here!
    }
}
//End Loop
//*********************************************************************************************************************************

//*********************************************************************************************************************************
//Standard PWM Functions Receive/Process
void receiveEvent(int howMany) {
    wdt_reset();
    if (howMany==5){
        byte cmd1, cmd2, cmd3, cmd4, cmd5;
        cmd1=Wire.read();
        cmd2=Wire.read();
        cmd3=Wire.read();
        cmd4=Wire.read();
        cmd5=Wire.read();
        if (cmd1=='$' && cmd2=='$' && cmd3=='$'){
            cmdnum=cmd4;
            datanum=cmd5;
            //Serial.println(cmd4,DEC);
            //Serial.println(cmd5,DEC);
        }
    }
    else{
        for (int a=0;a<howMany;a++){
            Wire.read();
        }
    }  
}

void ProcessCMD(byte cmd, byte data){
    wdt_reset(); 
}

//End Standard Functions
//*********************************************************************************************************************************
//Start of sunrise, sunset and cloud calculations- runs on reset and once a day thereafter.
void CalSun(){
   //Serial.println("CalSun Run Now");

        //*********************YOU NEED TO CHANGE THESE VALUES Read instructions in their ENTIRETY and CAREFULLY change to values for your tank and geographical region***************************
        //channels 0-5 are PWM expansion board lights 6,7 are ReefAngel Controller PWM outputs
        //offsets for rise/set all values in seconds offset from calculated rise or set value (-) am offset=longer day****** (-)pm offset=shorter day)
        //array order is ch0 am offset, pm offset, ch1 am offset, pm offset etc..
        //THESE values are the number of seconds that a particular channel will be offset from the rise/set time, i.e. negative to rise earlier/set earlier
        int Choffset[]={
            -600,0,-1800,6000,0,600,-1200,6600,-2200,6600,0,0,0,0,0,0};
        // NOW SET YOUR LATIDTUDE AND LONGITUDE COORDINATES as Degrees, Minutes, Seconds of Lat and Lon
        //If Your NORTH of the equator your LONGITUDE must START with a NEGATIVE number (the rest are positive) e.g. All of North America, Europe, Russia etc are negative
        //If Your EAST of the Prime Meridian your LATITUDE must START with a NEGATIVE number (the rest are positive), e.g. Most of Europe, All of China, India, Austraila, Russia etc are negative
        latitude=dmsToSeconds(-25,87,13); //Set to Witbank - South Africa
        longitude=dmsToSeconds(29,23,32); //Set to Witbank - South Africa
        //latitude=dmsToSeconds(40,44,00);//United States of America- Salt Lake City, local time is -7 hours GMT 
        //longitude=dmsToSeconds(-111,47,00);
        //**********************ok now were done changing things IF YOU CHANGED the Top part of the GLOBAL variable decleration AND this... your FULLY configured and ready to load******************************************** 
        
   trigger=false;//we only run once a day... or on a restart
    long hours, minutes;//store current elapsed local hours as total seconds from midnight
   
    time_t t=now();//store current clock time then parse to hour and minutes of current day using time.h functions
    hours=hour(t);
    minutes=minute(t);
    hours=(hours*3600);//convert to seconds
    minutes=(minutes*60);//convert to seconds
    
    newDay=now()-(hours+minutes);//Convert current local unix epoch time to local unix epoch time of midnight 
    //#define SECS_YR_2000 (946684800) the time at the start of y2k (need to subtract from unix epoch time to bring to Y2K origin
    newDay-=946684800;//convert GMT unix Epoch to seconds elasped since 2000 for GMT midnight of today
    
    rise=newDay;//set value to send to SunRise as midnight GMT in seconds from Y2K
    set=newDay;//
    //Calculate rise time and set time using Epherma Library functions (see end of code) 
    SunRise(&rise);//call to Epherma function
    SunSet(&set);//Call to Epherma functionunsigned long newDay;
    //set rise to elapsed time of day... its currently in seconds since 2000 for today rise time 
  /* Serial.print("rise and set=  ");
   Serial.println(rise);
   Serial.println(set);
   Serial.print("newDay=  ");
   Serial.println(newDay);*/
   
    rise=(rise-newDay);
    set=(set-newDay);
   /*Serial.print("rise and set as elapsed seconds of day=  ");
   Serial.println(rise);
   Serial.println(set);*/
   
    newDay+=946684800;//Convert current local to midnight 
   /*Serial.print("newDAy=  ");
   Serial.print(newDay);
   Serial.print("elapsed is");
   long elapsed=now()-newDay;
   Serial.println(elapsed);*/
   
   
   
   
    
        //Calculate rise and set times for all channels in equivlants to elapsed seconds from midnight today
        //populate array for chRise and Set as well as chSlope for 0.5pi/ half day lenght for each channel from midday (asymmetric days are allowed)
        float deltaY=1.570796327;//1/2 * pi as integer by scaling* 10^9 to fill UL
        midDay=(((set-rise)/2)+rise);
        long HalfDayLength=((set-rise)/2);
        //Serial.print("MidDay");
        //Serial.println(midDay);

[rufessor]

Ok-

I have confirmation that the newest code is fixed and we are getting rise and set times for LED PWM lights over a tank in South Africa that correspond exactly with the rise and set times of the sun out the window.

In order to simplify the process of individual users installing this code on their PWM modules and using the lighting effects it offers-

I am starting two new threads. The .INO FILE you NEED to download, modify and install on your PWM module is going to live in a thread titled

"Weather Simulation for PWM expansion module"
http://forum.reefangel.com/viewtopic.php?f=11&t=1450

Please go here if you want to download a working complete version of this code. This thread will stay active as I add new features (COMM is being worked on as of now). So if you want to help test/develop this is where you need to look. If you just want the latest complete working version of the code..... link is above.

[JoelIreland]

Hey all
I have just got a retro kit from Steve's LEDs for my Red Sea Max 650.
I have installed the kit and now want to add the dimmer expansion module to control sunrise sunset ( in Australia )
I already have the module and have been following this thread. The issue I am having is when i plug the module in to the PC to upload the code, it comes up with either faulty or unknown USB Device.
I run an IT company so this sort of thing is second nature but still having issues with it.

My question is, what is the best way for me to do this? and what code should I be using from GIT to control 4 chanels.

Thanks all
Joel

[rimai]

This would only apply to a very early module.
The new generation is not programmable anymore.
The code needs to be ported to run in the head unit for this to work.