Help add feedback to Sine Wave Generation using ATmega16

[AnwarB]

Hi all
On the subject of Tahmid
tahmidmc.blogspot.com/2012/10/generation-of-sine-wave-using-spwm-in_10.html
Talk about generating a sine wave with SPWM in the PIC16F684
Feedback added here
tahmidmc.blogspot.com/2012/11/feedback-in-sine-wave-inverter-pic16f.html
To be like this

---

Code C - [expand]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
const unsigned char sin_table[416]=
{
0,16,33,49,64,79,93,106,118,129,138,146,153,158,161,163,163,161,158,153,146,138,129,118,106,93,79,64,49,33,16,0, //65%
0,17,34,51,67,82,97,110,122,134,143,152,158,164,167,169,169,167,164,158,152,143,134,122,110,97,82,67,51,34,17,0, //67.5%
0,18,35,52,69,85,100,114,127,138,149,157,164,169,173,175,175,173,169,164,157,149,138,127,114,100,85,69,52,35,18,0, //70%
0,18,36,54,71,88,103,118,131,143,154,163,170,175,179,181,181,179,175,170,163,154,143,131,118,103,88,71,54,36,18,0, //72.5%
0,19,38,56,74,91,107,122,136,149,160,169,176,182,186,188,188,186,182,176,169,160,149,136,122,107,91,74,56,38,19,0, //75%
0,20,39,58,77,94,111,126,141,153,165,174,182,188,192,194,194,192,188,182,174,165,153,141,126,111,94,77,58,39,20,0, //77.5%
0,20,40,60,79,97,114,130,145,158,170,180,188,194,198,200,200,198,194,188,180,170,158,145,130,114,97,79,60,40,20,0, //80%
0,21,41,62,81,100,118,134,149,163,175,185,193,199,204,206,206,204,199,193,185,175,163,149,134,118,100,81,62,41,21,0, //82.5%
0,22,43,64,84,103,122,139,154,168,181,191,200,206,211,213,213,211,206,200,191,181,168,154,139,122,103,84,64,43,22,0, //85
0,22,44,66,86,106,125,143,159,173,186,197,205,212,216,219,219,216,212,205,197,186,173,159,143,125,106,86,66,44,22,0, //87.5%
0,23,45,67,89,109,129,147,163,178,191,202,211,218,222,225,225,222,218,211,202,191,178,163,147,129,109,89,67,45,23,0, //90%
0,23,46,69,91,112,132,150,167,183,196,207,217,224,228,231,231,228,224,217,207,196,183,167,150,132,112,91,69,46,23,0, //92.5%
0,24,48,71,94,116,136,155,173,188,202,214,223,230,235,238,238,235,230,223,214,202,188,173,155,136,116,94,71,48,24,0, //95%
};
 
unsigned int TBL_POINTER_NEW, TBL_POINTER_OLD,
 
TBL_POINTER_SHIFT, SET_FREQ;
unsigned int DUTY_CYCLE, FBV, FB_Step, adder;
 
void interrupt()
{
if (TMR2IF_bit == 1){TBL_POINTER_NEW =
 
TBL_POINTER_OLD + SET_FREQ;
if (TBL_POINTER_NEW < TBL_POINTER_OLD){P1M1_bit =
 
~P1M1_bit;}
TBL_POINTER_SHIFT = TBL_POINTER_NEW >> 11;
DUTY_CYCLE = TBL_POINTER_SHIFT + adder;
CCPR1L = sin_table[DUTY_CYCLE];
TBL_POINTER_OLD = TBL_POINTER_NEW;
TMR2IF_bit = 0;
}}
 
void main()
{
SET_FREQ = 410;
TBL_POINTER_SHIFT = 0;
TBL_POINTER_NEW = 0;
TBL_POINTER_OLD = 0;
DUTY_CYCLE = 0;
FB_Step = 0;
adder = 0;
T1CON = 0b00001001;//1:1 Prescale Value, Enables Timer1
ANSEL = 0b00000001;//RA0/AN0 as analogue input.
TRISA = 0b00000001;//PortA as output except RA0 as input
ADCON0 = 0b00000011;//Analog Channel (AN0)
ADCON1 = 0b01100000;//FOSC/64
CMCON0 = 7;
PR2 = 249;
TRISC = 0x3F;
CCP1CON = 0x4C;
TMR2IF_bit = 0;
T2CON = 4;
while (TMR2IF_bit == 0);
TMR2IF_bit = 0;
TRISC = 0;
TMR2IE_bit = 1;
GIE_bit = 1;
PEIE_bit = 1;
ADC_Init();
 
while(1)
{
FBV = ADC_Get_Sample(0);
if (FBV < 512){FB_Step++;
if (FB_Step > 12) FB_Step = 12;
}
else
{
if (FB_Step > 0){FB_Step--;}}
adder = FB_Step << 5;
TMR1L = 0;
TMR1H = 0;
T1IF_bit = 0;
}
}

---

And on the topic of Thamed

https://tahmidmc.blogspot.com/2013/02/sine-wave-generation-with-fast-pwm-mode_2525.html

Talk about Sine Wave Generation with "Fast PWM Mode" of AVR - using ATmega16
I tried adding a feedback to it and it stopped at ADC pins
Please help with that as my experience is not enough
This is what I modified, is it correct?
I am doing an experiment on the ATmega8
Thank

 

[danadakk]

Just an observation but un-commented code makes it doubly difficult to
understand what code is doing.

Also helps when posting code to use code tags so indentation preserved hence
logic block flow more visually apparent.


Regards, Dana.

 

[andre_luis]

At a glance, once you did not eable peripherals interrupt yet, seems weird this part of the code:

TMR2IF_bit = 0;
...
while (TMR2IF_bit == 0);

 

[AnwarB]

Is there a problem with this?

 

[betwixt]

while (TMR2IF_bit == 0);

Stops the program progressing until the TMR2 interrupt occurs. Usually, interrupts are designed to 'interrupt' the program flow rather than suspend it!
Brian.

 

[AnwarB]

Sorry, I'm inexperienced with this
This is from the topic of Tahamid
Please see here

https://tahmidmc.blogspot.com/2012/11/feedback-in-sine-wave-inverter-pic16f.html?m=1

Someone on this date October 2, 2013 at 1:28 AM
Put this

 

[betwixt]

I haven't gone back to see Tahamid's original postings but the software doesn't look correct to me and some of it seems to be missing. I assume the intention of having 13 sine tables is to give it voltage regulation capability but exactly how it does it isn't obvious. I'm not sure why TMR1 is used at all, it doesn't seem to do anything.

Brian.

 

[andre_luis]

In addition, the code could be optimized in memory usage, since the entire table is composed of values symmetrically distributed starting from the middle of each array. Frankly, considering that this code was not wrote by the author of the Blog, but rather from someone in the comment section, I would throw it away and take another from a reliable source, or read datasheet to do it yourself.

 

[AnwarB]

I mentioned that I have little experience in this area
Just want to add feedback to avr-atmega8

 

[betwixt]

We can't help with partial code, especially porting it from on device to another.
My guess is the way it is supposed to work is you step through ONE line of the table at a rate of (number of entries on one line multiplied by the frequency you want to produce) and use that value to feed the PWM generator. To regulate the output voltage you periodically read the ADC and use it's output to select which line in the table you use. As Andre pointed out, you can halve the size of the table by using each value twice, utilizing its symmetry. If I'm right about how the regulation works, personally I would use a different array for each set of values, it would make the code far simpler.

Brian.