This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

LM5180: LM5180 output abnormal if you touch the board

Part Number: LM5180

I used LM5180 to convert DC-24V to DC-24V.

output current under normal state is around 5mA.

Heavy load is limited to 50mA.

basic design requirement:

input range:18~28VDC

transformer is 30uH, np:ns =1:2

under normal/ heavy load/ light load condition, output OK.

under no load condition, if i remove the output Zener, output is about 70V, .

two layer of PCB.

my problem is:

1. if I use my finger to touch the board especially close the RSET pin, or if I use probe to measure RSET pin,  the output abnormal, transformer noise in the audio range which can be heard very clear. the FB point curve is above below. how does this phenomenon appeared? 

CH1 is FB pin, CH2 is output transformer/ VF anode.

2. i can hear a very light audio sound of the transformer even under normal condition. the transformer is varnished. why?

3. i see the switching frequency is from 12K to 350K, when the IC would've be working under audio frequency range?

4. under normal condition the RSET pin curve is as below, is that correct?

  • Tian,

    I wonder, with 24VIN to 24VOUT, why not use a 1:1 transformer?

    1. The RSET pin sets the GND reference for FB for VOUT regulation so introducing capacitance or resistance to this pin will change VOUT.

    2. The transformer windings may be vibrating, causing the noise.

    3. The IC operates at 12kHz at no load and very light loads. The device can only sense the output voltage during a switch cycle. To ensure we know if the secondary output falls, we switch at a minimum of 12kHz. But every switch cycle pushes some power to the output which is why we need the zener clamp. Once the load overcomes this power, the frequency will increase to maintain regulation.

    4. Let me check on this.

    -Sam

  • Tian,

    4. The Rset waveform looks reasonable.

    -Sam