Skip to main content

Timeline for PWM generation using Registers

Current License: CC BY-SA 3.0

13 events
when toggle format what by license comment
Oct 10, 2014 at 14:04 comment added mike Thanks. What you suggest is in Atmels Documentation called fast PWM, where the counters just count until overflow. I however use phase-corrected PWM, where it counts up and down. This gives a more symmetrical output (and according to them) is prefered over fast PWM.
Oct 10, 2014 at 14:02 vote accept mike
Oct 10, 2014 at 14:02 answer added mike timeline score: 0
Oct 10, 2014 at 12:19 comment added Ian M I'm not 100% familiar with Arduino timers, but here is how I did PWM on an MSP430 with timers: imgur.com/2aOnW2g Essentially, you count up, and reset. When the timer hits the set value, it does the rising edge pulse, and when it hits the max count, it resets both the pulse and the timer. The alternative of this is to do the opposite: set the pulse high until the timer hits the threshold value, etc.
Oct 10, 2014 at 11:42 history edited mike CC BY-SA 3.0
added 787 characters in body
Oct 10, 2014 at 7:35 history edited mike CC BY-SA 3.0
New info, Changed the desired update frequencies.
Oct 8, 2014 at 2:07 history tweeted twitter.com/#!/StackArduino/status/519670406051950592
Oct 7, 2014 at 15:31 history edited mike CC BY-SA 3.0
added 219 characters in body
Oct 7, 2014 at 14:36 comment added Gerben That doesn't really make it any clearer. Seems I can't help you then.
Oct 7, 2014 at 12:39 comment added mike 2048 to be exact for this case. Why is that contradictory, if I want to have pulses between 1 and 2 ms (something around 48% and 97% duty cycle)?
Oct 7, 2014 at 12:26 comment added Gerben I don't understand what you want to achieve. Pulse width of 1000ms at a sample rate of 488.28Hz that seems contradictory, as the maximum pulse width can only be 2ms at 488Hz.
Oct 7, 2014 at 10:15 history edited mike CC BY-SA 3.0
edited title
Oct 7, 2014 at 8:30 history asked mike CC BY-SA 3.0