Eliminate TPS54821 hum

Bryan Capdeville

Other Parts Discussed in Thread: TPS54821

I am using a TPS54821, converting 12vdc to 5.6vdc at a maximum of 5 amps. With a 1 amp load, I notice a tone or hum coming from the switching power supply. Attaching an o-scope to the output reveals an 8 Khz signal riding on the 5.6vdc. The load on the power supply is a Lithium Ion battery pack with an internal charger (not our product). As the charger in the battery back starts to reduce the charge current, this 8 Khz signal shows up. So I have a varying load at 8 Khz that creates a hum in my switching power supply.

Increasing output capacitance has no affect.

Components on the comp pin to gnd are 1.27K + 0.1uF in parallel with 390 pf.

Is it possible to change the R/C values on the Comp pin to eliminate the 8 Khz hum and not mess up the existing pole and zero locations?

over 11 years ago

0 JohnTucker over 11 years ago

TI__Guru**** 191445 points

Can you tell if the noise is coming from the inductor or output capacitors? What type of output capacitors are you using?

0 Bryan Capdeville over 11 years ago in reply to JohnTucker

Prodigy 80 points

Difficult to identify exactly who is "singing". Output capacitors are two 22uF ceramics in parallel. I have replaced them with a 150 uF tantulum low ESR with no improvement.

0 JohnTucker over 11 years ago in reply to Bryan Capdeville

TI__Guru**** 191445 points

if removing the ceramic output caps does not fix it then it is most likely the inductor. What is the pn for your inductor?

0 Bryan Capdeville over 11 years ago in reply to JohnTucker

Prodigy 80 points

Coilcraft XAL6060-822ME. 8.2uH shielded inductor.

0 JohnTucker over 11 years ago in reply to Bryan Capdeville

TI__Guru**** 191445 points

It sounds like the inductor is likely the source of the noise. The L and C values seem well chosen as the LC corner is below 8 kHz. It may be possible that different types of inductor construction will have improved audible noise, but I am not aware of any specific recommendation. You could possibly de-tune the loop response ny changing the compensation values, but once again I do not have a specific recommendation. I'll ask around to see if anyone has a suggestion.

0 JohnTucker over 11 years ago in reply to JohnTucker

TI__Guru**** 191445 points

Just to be thorough, can you post your output voltage waveform?

0 JohnTucker over 11 years ago in reply to JohnTucker

TI__Guru**** 191445 points

You might consider this inductor from Wurth:

Flat wire inductor has varnish between the layers. It helps to prevent acoustic noise.

7443550820 (8.2uH, Irated = 10A)

0 Bryan Capdeville over 11 years ago in reply to JohnTucker

Prodigy 80 points

I do not have the capability to copy the voltage waveform on my o-oscope. Thanks for your input.

0 JohnTucker over 11 years ago in reply to Bryan Capdeville

TI__Guru**** 191445 points

Are you interested in trying the inductor I suggested?

0 Bryan Capdeville over 11 years ago in reply to JohnTucker

Prodigy 80 points

I am not able to use this inductor. The general PCBA layout is complete. I can make small changes but the footprint of the suggested inductor is 4x the current one (6.5 x 6.5 mm available).

0 JohnTucker over 11 years ago in reply to Bryan Capdeville

TI__Guru**** 191445 points

From Coilcraft:

The XAL construction should be one of the best for these kind of noises.

Per Coilcraft Engineering – “The idea of lacquer on the windings won’t help in this case because it’s already molded solid.”

Is there any chance you can alter your load profile?

0 Bryan Capdeville over 11 years ago in reply to JohnTucker

Prodigy 80 points

Yes. For other reasons, I have had to add a separate LDO to drive the load that charges the external battery. Although my switching PS provides the input voltage to the new LDO, I am hoping the LDO will smooth out the oscillations prior to the load getting back up to the switcher. Thanks.

Power management

Power management forum

Eliminate TPS54821 hum