0
\$\begingroup\$

I'm using timer0 and a gpio pin to generate a 300Hz PWM signal on an ATmega2560.

The details if you care: F_CPU=16M, timer clock prescaler /256, count_max=208

The frequency changes depending on if I use an "if(){} else{} if(){}" or an "if(){} else if(){}". The first generates 300Hz and the second generates 244Hz. Below is the code that generates the correct frequency. If I delete the curly brackets after the first two else's, I get 244Hz.

void pwm_on(uint8_t duty){
    if(TCNT0 <= duty){ //portion of counter active
        PORTB |= (1<<7); //turn on PB7
    } else {}
    if(TCNT0 >= duty){ //portion of counter inactive
        PORTB &= ~(1<<7);//turn off PB7
    } else {}
    if(TCNT0 >= COUNT_MAX){ //if counter full
        TCNT0=0;//reset counter
    } else {}
}//pwm_on

Why am I getting such a significant frequency change with code that doesn't change the functionality?

Also calling this function is the only thing I do in Main after initializing.

\$\endgroup\$

closed as off-topic by brhans, Finbarr, RoyC, Elliot Alderson, Voltage Spike Apr 3 at 16:40

  • This question does not appear to be about electronics design within the scope defined in the help center.
If this question can be reworded to fit the rules in the help center, please edit the question.

  • 5
    \$\begingroup\$ I'm voting to close this question as off-topic because it's a C syntax understanding problem and has nothing specifically to do with the hardware the code is running on. \$\endgroup\$ – brhans Mar 28 at 11:29
  • \$\begingroup\$ @brhans I edited the title of the question to specify it is for PWM control since that seems to be the largest point of contention for moving this question to stackoverflow. I am not a computer programmer and this is not a computer programming question. It is about how the syntax is affecting register level hardware timer0, NOT why doesn't this syntax work. \$\endgroup\$ – TammerTheHammer Mar 28 at 11:57
  • 1
    \$\begingroup\$ It doesn't really matter what the end application is - your problem is that you're not understanding standard C syntax. You would have this same problem if you were writing similarly structured code to run on a PC. The fact that you're writing firmware for a micro is only relevant in how this problem is presenting itself - it's just the symptom. \$\endgroup\$ – brhans Mar 28 at 13:13
  • 4
    \$\begingroup\$ Why aren't you using the perhipheral to generate the PWM for you? Which would allow you to dedicate zero CPU time to this task. \$\endgroup\$ – Attie Mar 28 at 13:20
  • \$\begingroup\$ @brhans I have only been asking about the symptom, that being the change in frequency. My understanding shouldn't matter since the question has always been about the symptom. \$\endgroup\$ – TammerTheHammer Mar 28 at 13:49
5
\$\begingroup\$

This is more a code syntax question. Could be offtopic, might be moved to stackoverflow.

if(){} else{} if(){} and if(){} else if(){} are different.
With correct indentation and brackets the issue is immediately visible:

if(){} else{} if(){}

if(condition){
  statement
}else{
  statement
}
if(condition){
  statement
}

And

if(){} else if(){}

if(condition){
  statement
}else{
  if(condition){
    statement
  }
}
\$\endgroup\$
  • \$\begingroup\$ i'm asking about the change in frequency (hardware) caused by the change in syntax (code) which is why I asked it here. \$\endgroup\$ – TammerTheHammer Mar 28 at 8:29
  • 7
    \$\begingroup\$ @TammerTheHammer Of course, but you should learn to use if-else-elseif statements to solve the puzzle. It's not the hardware, but your code. \$\endgroup\$ – Marko Buršič Mar 28 at 8:46
  • 1
    \$\begingroup\$ electronics.stackexchange.com/help/on-topic / @TammerTheHammer you asked in the right stack. this has been discussed many times before and some people try to redirect others to stackOverflow while actually you should ask here for softwares related to programming microcontrollers \$\endgroup\$ – Hasan alattar Mar 28 at 9:12
  • 2
    \$\begingroup\$ @TammerTheHammer Well, looking at the question tittle, people would think you are asking how to use if then statements. \$\endgroup\$ – Marko Buršič Mar 28 at 9:27
  • 1
    \$\begingroup\$ @TammerTheHammer You haven't asked how to use if-else, but that's only because you don't realize it's the root of your problem. code that doesn't change the functionality - here it is. \$\endgroup\$ – Agent_L Mar 28 at 11:01
2
\$\begingroup\$

The reason why my frequency goes down and the period goes up with the "if(){} else if(){}" is because TCNT0 >= duty is always true when TCNT0 >= COUNT_MAX so the last statement that resets the counter at COUNT_MAX never runs so my counter overflows instead of resetting at a count that gets me a frequency of 300Hz.

I figured it out pretty quickly after posting, sorry if I am misusing this site, I'm posting a question on meta to see what the collective thinks so I can learn and use this resource as best as possible.

\$\endgroup\$
  • 4
    \$\begingroup\$ Posting of your own solution to a problem is encouraged. \$\endgroup\$ – AndrejaKo Mar 28 at 8:51
  • \$\begingroup\$ if you compile without optimaziation i think you will get same frequency but much lower than 244Hz anyway. i find it wierd that you right else {}. check this out : cpp.sh/3rcrt \$\endgroup\$ – Hasan alattar Mar 28 at 9:21
  • \$\begingroup\$ @Hasanalattar I'll need to research more on compiling before I'll understand this, but why I use else{} is because i learned misra-c standard in school and it was drilled into my brain that all if(){} statements must have its accompanying else{}. \$\endgroup\$ – TammerTheHammer Mar 28 at 9:34
  • 2
    \$\begingroup\$ Sidenote: To my understanding misra requires else{} only at the end of if ... else if constructs, and not on plain if statements. \$\endgroup\$ – user694733 Mar 28 at 11:08
  • 1
    \$\begingroup\$ @TammerTheHammer I tried looking for that MISRA rule, and as far as I can tell there is no such rule. There is one for chained if .. else if ... else if ... which sort of make sense but for a simple if there is no need for an empty else \$\endgroup\$ – r_ahlskog Mar 28 at 11:09

Not the answer you're looking for? Browse other questions tagged or ask your own question.