OK, I understand. There are some techniques (like you mentioned from Roman's page) where you can pre-distort the lookup table to reduce certain harmonics. Also, changing your PWM frequency may allow you to de-emphasize some of the harmonics. However, I think that you will eventually have to rely on the low-pass characteristics of whatever is connected to your H-bridge. For example, if you are driving a transformer, then the leakage inductance of the transformer can be used to help smooth the waveform. You may even want to make a resonant circuit if your sinusoid is at a fixed frequency. Sean On Fri, Nov 16, 2012 at 12:36 PM, Manu Abraham wro= te: > Hi Sean, > > What I was asking was the signal generation with a controller. I need to > feed this signal to a H bridge (push-pull) stage and thereby use the > switching elements in either cut-off or the saturation regions, rather th= an > in the linear region. > > Regards, > Manu > > > On Fri, Nov 16, 2012 at 10:51 PM, Sean Breheny wrote: >> Normally you would do this with a low-pass filter. >> >> On Fri, Nov 16, 2012 at 12:19 PM, Manu Abraham = wrote: >>> Hi Scott, >>> >>> On Fri, Nov 16, 2012 at 10:27 PM, Scott Dattalo wro= te: >>>>> In a push pull H bridge configuration, aren't the even harmonics >>>>> supposed to get cancelled ? >>>> >>>> The even harmonics of a square wave are cancelled only if the duty cyc= le >>>> is 50%. In general, the harmonics have the component: sin(n*pi*d), whe= re n >>>> is the harmonic number and d is the duty cycle. For 50% duty cycle, on= ly >>>> the odd harmonics are present. For a 1/3 duty cycle, every third harmo= nic >>>> is cancelled. >>> >>> >>> Thanks for the detailed info. >>> >>> I have a sine table over here, for which it's entries are a sinusoidal >>> variation >>> of the duty cycle. >>> >>> viz: >>> >>> 0, 2409, 4794, 7134, 9404, 11585, 13653, 15590, >>> 17377, 18997, 20433, 21673, 22704, 23517, 24103, 24457, >>> 24575, 24457, 24103, 23517, 22704, 21673, 20433, 18997, >>> 17377, 15590, 13653, 11585, 9404, 7134, 4794, 2409, >>> >>> Looking at the output spectrum, I can see all the harmonics, including = the >>> fundamental, with amplitude gradually reducing for the higher order har= monics. >>> >>> Is there any way that I can eliminate the harmonics, leaving the fundam= ental >>> as is ? >>> >>> Thanks, >>> Manu >>> -- >>> http://www.piclist.com PIC/SX FAQ & list archive >>> View/change your membership options at >>> http://mailman.mit.edu/mailman/listinfo/piclist >> -- >> http://www.piclist.com PIC/SX FAQ & list archive >> View/change your membership options at >> http://mailman.mit.edu/mailman/listinfo/piclist > -- > http://www.piclist.com PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .