Oscillation in magnetising current of half bridge SMPS..how damp this?

Status
Not open for further replies.
T

treez

Guest
Hello,
The primary magnetising current in this half bridge converter has an envelope of oscillation at 1.39KHz.
The oscillation is obviously between the primary series DC blocking capacitor, (5uF) and the one half of the split secondary (the half that’s not conducting at the time), being referred to the primary [1.62mH * (NS/NP)^2 = 2.62mH ]

Since this oscillation is at such low frequency, how can we damp it out without dissipating too much in the process?

The Half bridge converter is provided here as a PDF schematic, and as an LTspice simulation, and the 1.39KHz oscillation in the magnetising current is also shown.
 

Attachments

  • Half Bridge _Voltage Mode _Full wave voltage doubler input.txt
    13.6 KB · Views: 63
  • Half Bridge _Voltage Mode.pdf
    27.3 KB · Views: 122
  • Oscillation in magnetising current.jpg
    78.5 KB · Views: 165

Make C4, C5 = 2.2uF,

If the oscillation doesn't affect the o/p or the pri side - then possibly you can live with it?


Regards, Orson
 
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
c4 and c5 are each in parallel with c11 and c18, so it wouldn't be worth making them 2u2's
 

The oscillation is essentially an initial transient response to charging of the DC bus. It doesn't occur in normal operation.
 

if you run the simulation you can see that this oscillation goes on and on.
 

The oscillation has high Q and is stoked up by the bus voltage ripple pulses. The magnitude is too low to cause any problems, I guess about 50 mA (*) rms. Of course it can be reduced by placing a RC snubber parallel to C7.

*Actually 25 mA rms.
 
Last edited:
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
Are you sure this is not just numerical error in the transient
solution? Got a lot of discrepant activity in the higher level
assembly, and some high gain points that can amplify any
residue to visible levels. I'd try one run with Euler method
just to see whether it's circuit or settings / software.
 
Reactions: treez

    T

    Points: 2
    Helpful Answer Positive Rating
Status
Not open for further replies.

Similar threads

Cookies are required to use this site. You must accept them to continue using the site. Learn more…