Topswitch-jx transfer functions

Posted by: ashok.gak@gmail.com on 12/20/2011

Hi,

I’m working on TOP264 Flyback loop compensation. My problem is that Bode plots from transfer functions I derived and what PI expert shows is different (Poles and zero frequencies are approximately same in both, but I think the difference is mainly because of DC gains). Is there a place in PI expert where I can see transfer functions for the design? Can anyone please let me know what exact transfer functions (duty to output and feedback) are used in PI expert? I have confusion whether PWM gain is added to Power stage transfer function or to feedback.

Any help is appreciated.

Thanks
Ashok

Comments

Ashok -

Your question is a little above/beyond the scope of my experience and knowledge...especially with the inner workings of PI Expert. I have however, forwarded your question over to our resident PI Expert expert! He was a bit busy t get to your question yesterday but it's on his list of things to work on today.

-The Traveler

As far as the pure transfer function of the TOPSwitch is concerned its basically a control pin current input (Ic) to duty cycle output (D). This function is described is basically the slope of the curve described in the datasheet figure 6 (top curve)
The absolute number for the PWM gain is listed on page 24.
The PWM gain basically is part of the controller not the power train. The power train transfer function is duty cycle to output voltage. The complete transfer function is not listed anywhere in the software although the basic transfer function for a voltage mode Flyback is available in literature.

Thanks both of you for replying to my request. I just wanted to clarify few more things from my side so that you can correct me if I'm going wrong somewhere.

I'm working on a multiple output flyback, using TLV431-opto feedback without phase boost circuit. For duty-to-voltage output transfer function, I'm using the formula given in the attached image. Please let me know if I'm missing something here.

For feedback section, I'm going as per Figure 16 of AN-47 appnote. The changes from this circuit to my circuit are: No Lpf, Rbias connected between o/p voltage and cathode of TLV431.

My questions are:
1) The feedback transfer function plot includes the Poles and zeros given by Fig 16 on AN-47 and the ones due to 47uF and 6.8 ohms connected to control pin of Topswitch. Am I right?
2) What is the DC gain equation for this feedback circuit? (DC gain I calculated is not matching with that of PI expert's, due to which corner frequency is shifted and phase margin is not adequate)

Hope I explained this well. Waiting for your reply.

Thanks
Ashok

There is an additional pole internal to the TOPSwitch and it is placed at 7 kHz. You need to model this as it plays a vital role in the loop performance.
The DC gain as I explained is the slope of the Ic vs duty cycle curve. and is also laid out in the datasheet. The software uses the smallest value of gain to make its computations.

7kHz internal pole is already considered as given in the appnote. So, for DC gain, it is only Duty-to-IC ratio that matters? CTR of opto or gain due to any other components in the feedback ckt has no role in DC gain of feedback transfer function?

Thanks
Ashok