TPA3255: Clip Indication

Mayur M Bharadwaj

Part Number: TPA3255

Hello team,

We are using TPA3255 for our application for 240W output.

We are currently evaluating the performance on the Evaluation Board with 4 ohms load, and it is observed that, as soon as the output voltage reaches 27-28V, clip warning starts with low duty cycle i.e., it will not be on continuously. Further, if the output voltage is increased to around 31V, the clip duty cycle increases i.e., 40%.

We want to understand why this clip is observed as soon as the voltage reaches the ~27V threshold.

As the fault is high, and clip is low, as per the datasheet, it can only be a overtemperature warning. And to check the same, we reduced the output load by using a 8 ohms load, and the power will be reduced. Even in this condition, we have observed that clip indication starts at 27V. Also, even after a fresh power on, clip starts immediately if output is 27V.

We want to understand how it being driven, as clip starts at a particular voltage irrespective of load in the Evaluation Board, and even after a fresh power on.

over 2 years ago

0 Jesse Ji over 2 years ago

TI__Mastermind 20541 points

hi Mayur

what 's the value of your pvdd power supply voltage.

clipping usually happen when power supply cannot provide enough output voltage.

tks

jesse

0 Mayur M Bharadwaj over 2 years ago in reply to Jesse Ji

Intellectual 410 points

Hello Jesse,

The PVDD voltage is 48V. We have tried with 51V as well, clip would come around ~29V of output.

This voltage is within the allowed range of the IC.

0 Luis Fernandes de Pina over 2 years ago in reply to Mayur M Bharadwaj

TI__Genius 16370 points

Hello Mayur,

You will see pulses when it's pre-clipping for a short period and if you inspect the waveform at those points there will likely be a short instance of clipping. The CLIP_OTW pulses start at the onset to output clipping, typically at a THD level around 0.01%, resulting in narrow CLIP_OTW pulses starting with a pulse width of ~500 ns.

best regards,
Luis

0 Mayur M Bharadwaj over 2 years ago in reply to Luis Fernandes de Pina

Intellectual 410 points

Hello Luis,

We have seen the case where the output is intact, but pulses are seen in CLIP_OTW.

0 Shadow He over 2 years ago in reply to Mayur M Bharadwaj

TI__Guru 63310 points

Hi Mayur

This CLIP_OTW is active not only with voltage clipping, also when large current situation, CB3C is working. According to your description, it's quite possible caused by current. Also in your waveform, the CLIP_OTW working area, the output has clear ripple, which is also proven the CB3C working.

What is the OC_ADJ resistor on your board? And what is the load resistor value you are using? Maybe have a test with current probe, check what is the waveform.

0 Mayur M Bharadwaj over 2 years ago in reply to Shadow He

Intellectual 410 points

Hello Shadow,

We are using 27k for OC_ADJ.

Regarding the load, it is designed to give max output at 4 ohms load. When testing at 4 ohms, we have observed CLIP_OTW indication will come at ~27V.

We have also checked at 8 ohms load, where the power and current is reduced. Even with 8 ohms load, we are observing CLIP_OTW at ~27V, which is not as expected.

0 Mayur M Bharadwaj over 2 years ago in reply to Mayur M Bharadwaj

Intellectual 410 points

Adding one more image, where CLIP_OTW is just started, and a clean O/P available.

0 Shadow He over 2 years ago in reply to Mayur M Bharadwaj

TI__Guru 63310 points

Hi Mayur

This device seems indeed behave like this. If you test PWM waveform, you'll find when CLIP_OTW react, the PWM duty cycle is changing, which is similar with clipping behavior. Although the PVDD margin is still very large, it looks like there's strict limitation for the max duty cycle.

But sadly I haven't found any material left about it. Let me check with more team members.

Audio

Audio forum

TPA3255: Clip Indication