Solution Finder Get Tech Support

LNK306 bad regulation and noise problem

Posted by: dspro on

My power supply has a very audible noise. Noise is coming from the output inductor so maybe this is OK.

However, regulation is also inaccurate - slowly climbing from 12V to 13V with no load (3.3kOhm resistor minimum load present).
The regulated voltage looks clean and has no significant noise, only the strange change over time (approx 2 minutes from start to arrive at 13V).
There is no significant heating of any component and apart from the voltage issue everything is working well (I can successfully pull 240mA, my system maximum)

I hope you can help with suggestions on what may be wrong (or not) - I am including some relevant data and measurements.
The measurements look quite noisy, but perhaps you can verify that these are correct?

L1 is LPS6235-155MR (1.5mH 130mA at 20% drop)
L2 is MSS7348-105ME (1.0mH 270mA 20% drop)

Best regards,
DSpro

Comments

Submitted by PI-Chloe on 10/18/2019

Hi DSpro

Nice to hear that you are using Power Integration products.

I would like to know the following so that I can help you.
What is the IC series that you are using?
Vin range = ?
Vo =?
Io max =?

Thanks

Regards

PI-Chloe

Submitted by dspro on 10/20/2019

The IC is LNK306
Vin is 325VDC from rectified 230VAC.
Vo is 12V
Io max is 240mA

Submitted by PI-Chloe on 10/28/2019

Hi Dspro

For the TP1 waveform, I assume this is the yellow waveform.
The TP1 is between the Source pin U4 and the inductor L2 and the Freewheeling diode D2.
The waveform is 20V/div. Approximating the max of the yellow waveform, Max is approximately 70V. Which I think is wrong.
If the input is 230Vac (325Vrectified), We are expecting Vmax to be near to 325V in the TP1 voltage waveform. Please check.

May I know the regulation of the circuit @ No Load and Full Load?

For audible noise,
a “drum” or “dog bone” “I” core inductor is recommended with a single ferrite element due to its low cost and very low audible noise properties. However, the inductor should be selected as varnished type in order to get low audible noise.

Thanks

Regards

PI-CHLOE

Submitted by dspro on 11/12/2019

Sorry, I was temporarily using an isolated DC power supply at ~70V to capture the scope images.

The prototype has been running well for weeks on 325VDC(230VAC) now and switching to another inductor solved the audible noise problems.

The regulation is good from about 10-20mA up to with full load, but at the minimum specified load (3.3kOhm/3.6mA) the output voltage slowly drifts from 12V to ~13V after start-up. In our design, overvoltage up to ~16V is OK and will not cause problems - but it would be good to know for future designs if this is normal behavior.

Does the TP1 waveform looks correct to you (considering 70V DC input voltage)? If yes, I believe the design is ready for production.

Submitted by PI-Chloe on 11/18/2019

Hi dspro

Yes at minimum load 3.6mA the output can slowly drifts. This is a normal behavior. If you want to lower down the drift during minimum load, you can increase the load by changing R47.
Fyi! also that this will decrease efficiency.
If there will be now harm that during minimum load Vo is at 13V, No need to change R47.

Waveform is correct if you are probing on TP1 to GND.
May I know the output loading condition of the TP1 waveform.

Thanks

Regards

PI-CHLOE