2
\$\begingroup\$

A modified version of the PCB that I discussed on this post was implemented. It is still 2-layer with all components on top and the ground on the bottom layer. The transformer windings were directly soldered to holes on the PCB, but the input and output used connecteors. Here is a screenshot of the output voltage waveform (from experiment), before adding any common-mode chokes.

Flyback Output Voltage

The switching frequency is around 140kHz. My initial questions are:

  1. It is expected that the transformer leakage inductance will resonate with the parasitic capacitance of doide (and MOSFET too?). Is this usually the cause for the two ringings per switching period? Does the EMI play a role here as well. One thing I noticed was that when I press with my hand on the big (variable) resistor load, the peak of the riniging dropped by almost half (which was the moment at which the screenshot was recorded) - so it seems the an alternate impedance path was provided for the noise. Another thing is that the load was connected through really long cables (which we can also fix).

  2. Why is the ringing in the output different at diode turn-on than at didoe turn-off? Mainly, how come the freuqencies seem to be different?

There is a lot of room for improvement here, but at this stage it is just trying to understand what causes what...

\$\endgroup\$
  • \$\begingroup\$ The first question is how you are measuring this ... specifically, how long is the scope earth lead? Ideally the scope probe should be into a coax socket whose shield is soldered to the ground plane; the closer you can get to that, the better. \$\endgroup\$ – Brian Drummond Dec 16 '12 at 12:19
  • \$\begingroup\$ @BrianDrummond The probe has its +ve/ground connected at around the same point. It is somewhere along the cable going from connector to load. That cable is of a first shorter (attached to connector) and a second longer one (connected to load). Will keep what you said in mind. \$\endgroup\$ – Ahmed Shehada Dec 16 '12 at 16:00
  • \$\begingroup\$ Just a thought. Have you tried optimizing your snubber circuit? Maybe it can help on the output. \$\endgroup\$ – JC203 Jul 28 '15 at 15:11
1
\$\begingroup\$

I have also seen the same waveforms on Multiple occassions and yes they are due to common mode noise, if you use differential probe you won't see it on Flyback outputs. Also if you use 2 single ended probes with one at the output and one at ground and use A-B function on DSO you won't see it. thanks for this important discussion.

Regards,

Chaitanya B

\$\endgroup\$
1
\$\begingroup\$

Seems like a very old post so probably obsolete, but for the record:

The converter is running in discontinuous mode, the small high frequency ring occurs at turn off when the main primary side transistor (usually MOSFET) switched off and the current is transferred to the secondary winding. The resonant frequency is determined by the leakage inductance of the transformer and the stray capacitance of the drain (collector) node mostly made up of the drain - source capacitance assuming a MOSFET is used. The leakage inductance is a relatively small value (typically a few percent of the magnetising inducatance) hence the high frequency.

The low frequency ring occurs after all the energy is transferred from the transformer (more accurately a coupled inductor) and the current in the output rectifier diode stops conducting. At this point the magnetising inductance and the Drain node capacitance begin to oscillate this time at a lower frequency due to the relaitively high value of the magnetising inductance. This ringing stops when the transistor is switched on and the current in the primary side is ramped up, there is no ringing at turn on as there is no current flowing at that point.

The change in amplitude when for example placing your hand on the load is perhaps a little more difficult to accurately describe. It will most likely be common mode noise and may simply be that placing your hand on the load presents a lower impedance load to the noise and hence the voltage level of the noise drops. There may well be more going on here but I hope that this clarifies matters a little.

\$\endgroup\$

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.