SMPS transformer design software

Thread Starter


Joined Feb 3, 2009
I am trying to make a Small fly back SMPS.
I have two software for doing this. Viper soft for Viper22 and Power Int Software.

When I give the same out put voltages and currents to both software, the calculation result seems to be much different. I can understand that both software uses different switching devices for SMPS.

Is it possible that by changing the active device, it can cause huge change in transformer winding? Or something wrong with my procedure. (Vipersoft gives 160 turns primary whereas Powerint gives 60T the same core). Can there be that much difference in primary inductance in two circuits?



Joined Oct 23, 2010
Software is good. Software is bad.
And you can create a list of whys under both of those statements.

A problem with using a software package to do design work is that many times it is used blindly without knowing what it is doing. Software is someone's implementation of what he thought was the correct design process. It often makes assumptions and/or uses 'rules of thumb' but unless the software documentation clearly states those the user hasn't a clue and simply takes the results as a valid design.

In the case you mention I would ask:
1. What design formulas does each SW package use?
2. What assumption about the design are made by each SW package?
3. Are any 'rules of thumb' used by the SW? If so, what are they? Do they apply to what you are doing?
4. Do you know how to design the circuit without using the software or you blindly hoping it will make all the right decisions for you?

Software design packages can be great aids for design especially where a lengthy complicated design process can lead to mistakes. However, unless they are well documented and understood by the user and verified for accuracy they can lead you down a primrose path.

You might build up both SW design implementations and see which one works then at least you will have a better idea which is right or better.


Joined Sep 7, 2009
As I always say, software design tools can only be safely used when the operator is smarter than the machine...... that way you know when it's lying or when it is making choices based on assumptions you don't want to use.

Designing a transformer requires a whole truckload of underlying assumptions like max core flux density, max ripple current, max op temp, etc.

Might be better to see if a transformer bendor has something "off the shelf" they probably do. Try coilcraft or pulse engineering.