Hello
I'm wondering if someone can explain this to me.
The UC3846 on this development is set up as a normal isolated push pull config. The primary peak current sense is via a 2 parallel 50mR
power resistors. The design works very well, but yesterday I noticed while probeing around. The current feedback into the IC after the RC filter did not seem right.
I have taken a screenshot to show. The yellow trace is the voltage developed across the sense resistors. This is fine as expected, drain currents are very well matched as woudl be expected with a current mode IC. But notice the blue trace, this is after the RC filter (1K & 4.7nF), for some reason the peak amplitude of every other ramp is lower (by 80mV in this shot)
The pulse widths of each switching cycle are the same. I have noticed the start of the current ramps are diffrent. Which I think is why the peak value is diffrent as the di/dt of the ramps are the same. I think this behavior is basicaly due to the RC time constant of the filter network, its much higher than the switching cycle time (clock frequency of 46Khz)
The PSU works well and this is not affecting operation but I would like clarification on this please as to the reason it happens?
Thank You
Alex
