Solution Finder Get Tech Support

Vbias current in TOP2xx flyback power supply

Posted by: florin70il on 09/08/2022

Hi guys, I have a question, why PI expert is finding  2x#AVG25 in parallel for bias secondary? Diode is 4148, cap is 10uF, the only load is  optocoupler output. Why is that? I strongly believe is a PI bug, but this is affecting trafo size or fitting all the copper in a specific window. Thank you 


Attachment Size


Submitted by PI-Wrench on 09/09/2022

The bias winding is also being used as a shield - the bifilar winding with larger wire will help to maximize coverage to make a better shield.

Submitted by florin70il on 09/12/2022

As I said, this approach is not right, at least a text should appear to explain why this Bias is using 2 wires in parallel even is not usefull at all from the electrical point of view. Also, some designs is calculated for nonstandard gap ferrites. I am sure nobody is willing to play adjusting the gaps to a specified value. Solutions given by PI is just  theoretical, not physically viable. Ferrites should be chosen from existing part numbers not resulting from some algorithms. As size/format. and PI part numbers


Submitted by PI-Wrench on 09/12/2022

Using the bias winding as an additional shield is standard procedure in power supply design. Also, non-standard gaps are used routinely by high-volume supply manufacturers so that they can optimize their design. They have the ferrites custom ground by the core manufacturer or an outside contractor, and enjoy economy of scale. A standard ferrite gap/AL might not yield an optimized design. We at PI have in house a core grinder  that is used to custom gap ferrites to optimize our in-house designs.

If you wish to use a commercial core with standard gap/AL values, it is possible to manipulate the KI and KP values in the design spreadsheet to come up with a design that has a gap/AL close to a standard value. It may also be necessary to manipulate the number of secondary turns in the design.