High frequency noise in smps transformer

[shyamramnani]

I have assembeled one smps circuit for 25volts, 40 amps. Circuit gives unwanted high feaquency noise at part load. It is silent at no load. What could be the problem. I am not electronic engineer, Electronics is my hobby. Any body can help me.

 

[yungman]

You have a scope? IF so, look at the pulse waveform driving the MOSFET, if the width change back and fore when the transformer making noise, you have a stability problem. Normally if you have a constant load, the pulse width is constant, you see a square type wave with duty cycle depending on the load, and if the load don't change, the pulse width is constant.

If you see the pulse width vary, you have oscillation. This is closed loop feedback stability problem. You need to know either Laplace or Bode Plot to tame it. If you are a hobbyist and don't know what I am talking, post the schematic in detail, give the value of the filter caps. People here can take a look.

 

[Mike_In_Plano]

I take it your describing a noise you can hear?
If it tends to be hissy, you likely have what's called sub-harmonic instability. It's amazing how many supplies have actually made it to market with this problem, but still work.
The leading cause of this problem, in flyback style supplies, is that the transformer ratio is off such that the pulse width exceeds 50%. If your stuck with a given turns ratio, a technique called slope compensation can help.
Extra advice:
- As a novice, don't attempt voltage mode controlled supplies. Instead, use a current mode controller such as a UCC384x.
- Another problem area is the output rectifiers staying on and causing a current spike in the MOSFET when it tries to switch on. This makes a lot of supplies load sensitive because the current sensing line gets a big pulse at the beginning of each cycle. This in turn throws the current trip point off... If you try to filter the current sense line, this problem remains.
The easiest fix I've found is to operate the supply in discontinuous mode. What this means is that whatever energy the transformer stored during the on time is totally depleted during the off time. At low load conditions, you naturally operate in this mode. To get it for higher loads, resize the transformer. You want it's inductance low enough to ensure that it discharges fully during the off time. Just remember E=1/2 I^2 L and dI=V/L and you'll likely be able to find a transformer inductance.

Best of Luck,

Mike

 

[yungman]

Without more info, it can be anything. Don't think we can help until he gives more info.

 

[alphysicist]

There could be several reasons for the high frequency noise in your SMPS transformer. One possibility is that the transformer is not designed to handle the high current of 40 amps, which can cause saturation and increase the noise level. Another possibility is that the transformer is not properly shielded, allowing electromagnetic interference to affect the circuit. It is also possible that the components used in the circuit are not suitable for high current applications, leading to increased noise. I would recommend consulting with an experienced electronic engineer to diagnose and solve the issue. It is important to ensure that the components used in your circuit are suitable for the desired voltage and current levels to avoid any potential safety hazards. Additionally, using proper shielding techniques and selecting appropriate components can help reduce the high frequency noise in your circuit.