I am am building a oscillator circuit for testing different transformer designs. I need the the freq to be idependent of the dutycycle adjustment. For testing at a certian freq say 50Khz, I would be able to select the best dutycycle for the given transformer design, at 50khz. I was thinking of using a PWM controller IC like the TL494. Instead of using feeback from the PS output I would use a pot with a V.D. Any ideas on how I could achieve my goals?

 

This could be used as your generating part. You can easily drive another switching device to enable you to switch a transformer winding. The frequency shown is about 50kHz, with the scope showing the min and max duty cycle (without any significant frequency variation) according to the position of the pot. The 2 fixed resistors limit the min and max settings.

Attached Images

duty_cycle.gif (14.7 KB, 550 views)

__________________
\"You can\'t make a circuit fool-proof, cause fools are so ingenious!\"

 

If you need signals in anti-phase for example push-pull drive, you could use the circuit modified as shown. You need to insert 2 more inverters, one in each line if you logic is active High (B). Also (remember to double the switching frequency F)

Attached Images

duty_cycle1_172.gif (16.2 KB, 542 views)

__________________
\"You can\'t make a circuit fool-proof, cause fools are so ingenious!\"

 

The following modification to B will also add some dead-time to prevent glithes on the output, because of the possible "race problem" when switching. Yet another way (without the diode-cap-res) is to create a delay with inserting a buffer (2 x 4093's in series) in one leg as shown.

If all this sounds too much for you :shock: , simply take a SG3525 and apply a DC control voltage directly to the compensation pin (or to pin 3 on a TL494 for that matter) to adjust the duty cycle manually. Sorry, I always like to show how things can be done discretely so that newbies can also learn in the process.

Attached Images

dead_time_162.gif (14.9 KB, 525 views)

__________________
\"You can\'t make a circuit fool-proof, cause fools are so ingenious!\"

 

Thank you so much! You more than answered my questions. I will get to work testing each circuit. The push-pull will deff come in handy!
Merry Christmas and Happy new year
William