please read it ( real time)

[paul borgmeier]

If you are interested in changing your approach, I posted some picbasic pro programs that can keep time at +/- 2 seconds per day (no RTC required). These are in the “Code Examples” Section at:

(Easy and Accurate Clocks without RTC IC)
http://www.picbasic.co.uk/forum/showthread.php?t=2129.

Both programs require a 4.000MHz Crystal, use TMRO, but do not require preloading counters. The programs are short as well. Good Luck.

Paul Borgmeier
Salt Lake City, Utah
USA

[moud_man]

hi all

i know that you arenot understanding me but all of your words are very usefull to me .

the missunderstanding is that i am building a clock project (you know, if i am building any other project i wouldnot need the time acc. ) that need a very high acc. that is it. ok!

i will upload to you my project code so you can really understand me, there are notes On it :

1-i am putting the Elapsed Timer,this code will calculate a very accurate 1 sec every a 100 ticks ***BUT THAT IS THE CASE IF THE ELAPSED TIMER CODE IS ALONE IN THE PROJECT****
2-the pic is driving a tow seven segment (if you want to build) and rotate the display around the tow seven segmant by porta.0 and porta.1 that is the main project code.
3-the dispaly is updated every one sec. of the ELAPSED TIMER CODE
4-the 2 seven segment is in the inverted mode that why i am putting 255-()
my observation is:

the display of the 2 seven segment is very well the problem is that there are some seconds that the project updates the display for more that 1 sec. and as the number increase the period of the sec. become shorter near to the actual one THe most long period of second is from 0 to 1 like (20 > 21)i donot know why?
No, i almost know, that instant is in the conditioning code of the program and it take more instruction cycles to excute is that correct?
if yes and i hope that it is the case what is the sol.

i had an idea i tryied to insert a condion "if " (like in the transition from 0 to 1 as i mentioned ) and will not do some thing like if d1= 100 then d1= d1 ,but it
doesnot work why again?
i want the soluion by editing the code and send it to me back or any thing ....

i am in the begining of the project in the issue of second i want to proceed in it to the hours but that problem is blocking me from that but it is the most important one, it is the base of the project


THE TIME

[Darrel Taylor]

moud_man,

Although your program is doing things it doesn't need to be doing. There's nothing in there that would cause the described problem. The instructions there will execute in only a few microseconds, and there won't be a problem of not having enough time to make it through the loop.

So let's look at the hardware instead.

Are you using transistors to drive the commons of the LED's? The PIC output's can only source 25ma. It takes more than that to drive a seven segment LED.

Are you using one of those Proto-Boards with a thousand holes that you just plug stuff into? If yes, where are the crystal and capacitors located? Are they very close to the chip? Long wires to the crystal will cause unstable frequencies. Especially if they are right next to high current outputs such as you are using on PORTA.0. The more segments that are lit, the more interferance you would see.

What kind of power supply are you using. Is it capable of providing at least 150ma while still maintaining +5V?

Now, back to the program. The Elapsed Timer keeps track of the seconds, minutes, hours, and days. There's no need for the Update: routine to try to count it again. Just use the variables that are already there. Hint: DIG.

***BUT THAT IS THE CASE IF THE ELAPSED TIMER CODE IS ALONE IN THE PROJECT****

That is incorrect. I currently have the Elapsed Timer running in 3 different commercial products with many other things going on at the same time.
<br>

[moud_man]

hi

sorry for delay that is because of my prescaler is asigned to 256 ('_')

well i will give you my answers and give you some ques (ques. ans.)


1- ques. what is the things that my program is doing and it doesn't need to do?
2- ans. i donot use a transistor to drive the 7-segmant i am using direct pluging from the pic , the 7.s is a common cathods that the pic sink current from it by a 1k ohm resistor.
3- ans. yes i am using one of those Proto-Boards with a thousand holes that i just plug stuff into,and the cristal is very close to the chip .

4- ans. i am using 7805 regulator
5 - ques. i dont understand what you mean with that no need for the update function? your variables canot be used because i am using tow seven segmant every one has a 1 variable controling it (d1, d2) if you know a method to extract those from your variable "Seconds" plz tell me

waiting to answers

thnx for repl.
bye for now

[Darrel Taylor]

(1,5) I guess the Hint didn't work. You can use the DIG function to extract the individual digits.
<pre> d1 = Seconds DIG 0<br> d2 = Seconds DIG 1</pre>Then you don't need to re-count everything.

(2) For common cathode 7 segment displays, you should have seven resistors, 1 to each Anode, around 270-330 ohms. The cathodes should have an NPN transistor (1 per digit) with a 10K resistor going from the PIC (PORTA.0 and PORTA.1) to the BASE of the transitors. Emitters connect to ground, and Collectors go to the cathodes.

The transistor will invert the signal, so instead of putting a LOW on the pin to turn on a digit, you would make it high.

(3) OK, the crystal is close to the chip. Good. What about the capacitors? No long wires?

(4) 7805 is OK, do you have good sized caps for filtering? And more important, do you have 0.1uF capacitor between +5V and GND? Close to the chip? For decoupling.
<br>
P.S. "my prescaler is asigned to 256 ('_')", good one

[moud_man]

now i understanded the HINT

what amazing (dig ) function do it is very very helpfull in my case

you know

with your code and your hint i can update my project before i completely build it !!!!!!!!! that is the truth

frist of all thank you for help

---------------------->
1- i donot have any capacitor in the power circuit.
2-i am using a 2 capacitor with the crystal.
3-the desister!!!

i cannot know really what type of 7 segmant i am using common(cathod or anode)

but the common line must kept high and the other terminal i will give it low to power the segmant assocciated that is it,,, what type is it? common cathod or
anode?


till now i didnot modify my code with the amazing (DIG) function but i will try it now


thnx a lot for help i will update the result sooooooooooooooooooooooooooon wait for me


bye for now