TPS5450&TPS5420 EMI/EMC problem?

Ambalavanan Shanmugasundaram

Other Parts Discussed in Thread: TPS5420, TPS5450

We are using TPS5450 and TPS5420 for our project. The output at pin no:8 is distorted output. Schematic has done as per datasheet given by texas instruments. Kindly suggest to arrest the harmonics at 500KHz frequency.

Technical Clarification

1)The below circuit consists of EMI filter board. The ferrite beads and choke are suggested by wurth electronics.

Fig.1. EMI Filter Board

2)The below circuit consists of TPS5450 buck converter from Texas instruments. As per datasheet we are designed the circuit.

Fig.2. Voltage regulator Board

The output of EMI filter is given to input of the voltage regulator Board. As we are getting EMI/EMC issue on conducted emission test.

Graphs for CE-03 is given below:

The waveforms for CE-03 is given below,

over 9 years ago

0 JohnTucker over 9 years ago

TI__Guru**** 191445 points

None of your images are visible. Can you try re-posting them?

0 Ambalavanan Shanmugasundaram over 9 years ago

Prodigy 30 points

0 Ambalavanan Shanmugasundaram over 9 years ago in reply to Ambalavanan Shanmugasundaram

Prodigy 30 points

Hi john,

We are trying to reduce the harmonics at 500KHz at input filter stage and also i have tried the inductor in buck converter above 200MH. So the 8th pin of the buck converter seems to be constant square wave. Whether it will helps to resolve the problem in CE-03 test at Lab.(mil std-461c). Kindly suggest your feedback. can you call regarding this issue. The harmonics graph is given below at 500KHz from buck converter ic.

0 JohnTucker over 9 years ago in reply to Ambalavanan Shanmugasundaram

TI__Guru**** 191445 points

None of your images are available. But yes pin 8 is the PH pin (switching node). It will always be a rectangular pulse waveform at a fixed 500 kHz frequency. You should maintain the suggested L and C values on the output filter to ensure stable operation. You can consider additional LC stages on both the input and output of the device. If you use an input inductor, be sure to add some electrolytic capacitance between the inductor and the input ceramic capacitors at the VIN pin of the IC. Carefull routing and layout can also minimize EMI issues.

0 LeonardEllis over 9 years ago in reply to JohnTucker

TI__Guru 71526 points

There was a snubber design App Note for reducing EMI referenced in this post: https://e2e.ti.com/support/power_management/non-isolated_dcdc/int-non-isolated_dcdc/f/198/p/83940/289431#289431 but I cannot seem to find any such App Note. Please send me the link, and any other design notes for reducing EMI.

0 JohnTucker over 9 years ago in reply to LeonardEllis

TI__Guru**** 191445 points

I have a snubber design guide that I use:

0652.RC Snubber.pdf

0 LeonardEllis over 9 years ago in reply to JohnTucker

TI__Guru 71526 points

Thanks John,

I'm getting some Spectrum Analyzer plots from this customer (TPS5450 design), and just wondering if we did any EMI testing of the TPS5450EVM, if we have any test results or documents of such?

0 JohnTucker over 9 years ago in reply to LeonardEllis

TI__Guru**** 191445 points

We do not generally do any EMI testing. EMI testing is generally applicable to end equipment, not individual sub circuits. End equipment may have enclosures to reduce radiated emissions and line filters to reduce conducted emissions. To include those would make the EVM overly complex. I know for sure that we did not do any for TPS5450.

0 LeonardEllis over 9 years ago in reply to JohnTucker

TI__Guru 71526 points

Here are scans from the TPS5450, plus one from a different manufacturer. Other than the snubber circuit, what other recommendations for reducing this EMI can we recommend?

base line:

RE30-200MHzVerticalSCF_UTX2SBaseLine.pdf

vertical mount:

RE30-200MHzVerticalSCF_BOB12VOFFMonExt#2.pdf

other IC vendor:

RE30-200MHzVerticalSCF12V_xxxy13.pdf

~Leonard

0 JohnTucker over 9 years ago in reply to LeonardEllis

TI__Guru**** 191445 points

It looks like the primary difference is in the 80 MHz range. Check the PH node waveform to look for any ringing in that frequency range. If it is present a snubber may help. You can also check the layout to ensure the circulation current paths are small and well designed, especially the SW node. Additional high frequency bypass capacitors may help as well.

Power management

Power management forum

TPS5450&TPS5420 EMI/EMC problem?