Please see the simple SMPS buck circuit below (24V to 12V)

enter image description here

Notice the load on the 'first' secondary coil is about 100mA (marked 'fb' This coil is used as the feedback to the MC43063 to regulate the switching duty cycle.

The other secondaries have different loads. One has a 120 ohm resistor (100mA) and the others have 10K resistors applying less load.

Now please see this output voltage plot :

enter image description here

Notice that FB (green) rises to 12V and remains steady.

Notice that one (blue) rises to just above 12V and remains steady.

And notice that the other outputs with smaller loads, raise to above 12V and keep slowly rising.

I notice that if I use a smaller load on the feedback output, then the outputs still drift.

What is causing this drifting in the secondary output voltages, and how can I stop it?

I realise I can use a voltage regulator (78L12 etc.) but until I know what is causing this drift I'm reluctant to try and regulate it.

Also what will happen if I apply a variable load to the output coils, will the output voltage fluctuate?

Thank you in advance.

  • 1
    \$\begingroup\$ You are only regulating one output, so as long as the others have different loads, there will be different charging/discharging times for the capacitors. \$\endgroup\$ – a concerned citizen Dec 3 '20 at 15:46
  • 2
    \$\begingroup\$ Google flyback cross regulation. Include coupling factor in your transformer model and re-simulate if needed. \$\endgroup\$ – winny Dec 3 '20 at 15:54
  • \$\begingroup\$ Your expectation of perfection for a multiple output flyback converter is a tad naive (even when k = 1). Try making k = 0.98 and step into reality. You have a simulator - try it with a load as from a current sink in parallel with a regular load resistor. \$\endgroup\$ – Andy aka Dec 3 '20 at 16:13
  • 1
    \$\begingroup\$ @winny Thanks.. 'flyback cross regulation'... The magic google incantation. I think I should be able to find an answer here. \$\endgroup\$ – Richard Dec 3 '20 at 16:54
  • \$\begingroup\$ @winny is it ok to braid secondary windings together when doing multifilar winding? I realise this will couple them in some way, but apparently multifilar winding helps with cross regulation... I've never seen braided secondaries before though. \$\endgroup\$ – Richard Dec 3 '20 at 17:55

To add an example to the hints in the comments, here is a basic way to do it:


The top plot shows the current in the secondary, and the two branches vary differently, yet the bottom plot shows how both the outputs are affected. The outputs try to stay at 12 V and 5 V due to the common feedback, but the outputs cannot be isolated since they are physically on the same transformer. The slight variations in the "stable" portion are also slightly afected.

  • \$\begingroup\$ Thank you for the answer. The isolation is the important part though. While reading about cross regulation I did see some examples where several of the outputs are opto-isolated and used in joint feedback, but that's too complicated for my basic needs. I think my plan is to multifilar wind the secondaries and use a linear regulator on each. Also to apply 100mA load to the feedback coil. This is very inefficient, but it should work. \$\endgroup\$ – Richard Dec 3 '20 at 21:26
  • \$\begingroup\$ @Richard In that case, why not simplify even more by not buying an IC and MOS and ferrite trafo, and just use the ol' reliable 78xx or 79xx with a mains trafo? Might be bulkier, though. \$\endgroup\$ – a concerned citizen Dec 3 '20 at 21:47
  • \$\begingroup\$ It's part of a larger project. I could do this several other ways but this may end up being the cheapest if I can get it working. \$\endgroup\$ – Richard Dec 3 '20 at 23:06

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.