TPS54233....design with WEBENCH or SwitcherPRO

I have exactly the same problem (24 to 4V, 1A).  When I plug in the WEBENCH compensation component values in SWITCHERPRO it says the crossover freq is 34kHz which is too high.  The SWITCHPRO design crossover freq is 14 kHz.. I built the WEBENCH circuit and it was unstable.   Is there a problem with WEBENCH?

I'm not sure I trust either of those tools.  Switcherpro has its strong points, but it has been out of production for a couple years now and there is no further development or bug fixes planned.  Webench does not have the features required for me to do any type of check into its operation.  The key characteristic is the power stage gain and phase, which webench does not support.  The only thing I could do would be to build that circuit and measure its performance in the lab.  Personally, I use pspice modeling for all my compensation designs, but there does not appear to be a pspice model available for TPS54233.  I'll see what I can find out, but it will take a few days.

I did a quick check in switcherpro.  For TPS54233, the power stage is modeled as ideal current mode.  There is no consideration for the effect of slope compensation.  For low duty cycle operation below 10% it is accurate, but the accuracy degrades as duty cycle increases above 10%.  So I think you should avoid switcherpro except to get a starting point for your design.  You would need to measure the loop response and adjust it based on your measurement.  As I stated previously, there is no easy way for me to examine the accuracy of the webench design.  I will have to pass that to the webench team.  They can get "under the hood" to see how the device is modeled.

Many thanks for Your fast reply.

I look forward for other Your information about Webench.

Do You have an application note or some info on how i can measure (in my laboratory) the fase margin and the gain of my project? Wich is the correct mode to test the stability?

Thanks in advance.

Best Regards.

Sandro

I found in Texas Instruments this:

AN-1889 How to Measure the Loop Transfer Function of Power Supplies (SNVA364A–October 2008–Revised April 2013)

For JohnTucker: Do You have news?

Thanks

Sandro

We are looking into it.  The software person thinks that slope compensation is in fact implemented in SP.  But It does not look like it from the designs I tested.  The only way to find out is to build the actual SP circuit and test measured vs calculated results.  If this were just the SP tool, we would probably let this alone as SP is no longer directly supported, but we need to verify WB as well.  It may take a week or so to get to the bottom of this.  I'll post the results when they are available.

Hi John Tucker,

Do You have news?

Thanks

Sandro

I'm going to have to run some hardware tests to check against the tool results.  I don't currently have any time scheduled for that investigation.  I'm going to work it in, as I am curious about it myself.  But I have to arrange it with other projects which have hard deadlines (and take priority).  I'll post something in a week or two.