TPA3255: TPA3255ampBoard-question

Hi Alexander

thank you for your question.

let me clear some thing about your question.

before: 4700 mkF 75V electrolytic capacitors + 48V 7A Switching Power Supply unit turned up to 50 volts -> work normally

after: 70V 10000 mkF capactiors + 48V 7A Switching Power Supply unit turned up to 50 volts  -> chip burn out. 

your question: you want to take a try a more larger capacitor 20000mkF capacitor and want to know is it ok or not?

can you provide your 3 types capacitor's datasheet, then we can check what might be the reason.

thanks

jesse

Dear Jesse,
Thank you for yor message.
Let me answer your question.
Before: 2*4700 mkF 75V Samwha electrolytic capacitors (as can be seen on the photo) + 48V 7A Switching Power Supply unit turned up to 50 volts -> worked normally.
After: 2*10000 mkF 63V JB JNC Series electrolytic capacitors + 48V 7A Switching Power Supply unit turned up to 50 volts -> 1) chip instantly burned out; 2) the Power Supply unit is OK.
I would like to know the possible reason of the chip’s burn – is it because in this configuration it can’t sustain 20k mkF bulk capacitance or is it something else?
Datasheet for the JB's JNC series of capacitors is attached.
Please don't hesitate to contact me should you have futher questions.

Regards,
Alexander

Hi Alexander

it seems like no attachment.

can you share me again.

BTW, is that possible share use the schematic of your circuit.

thanks.

jesse

Hi Jesse,

I am here to attach the mentioned datasheet for Alexander. Please check and help further. 

 JNC-2000H-at-85-Snap-in-Aluminum-Electrolytic-Capacitor.pdf
If it is not available, there is a web one as well:
https://www.jbcapacitors.com/pdf/JNC-2000H-at-85-Snap-in-Aluminum-Electrolytic-Capacitor.pdf

Regards.

Hi Chong

thank you for you information.

i haven't seen any problem with your capacitor. it might be caused by some other issue. it is hard to say.

can you take a try soldering another chip on the board and take a try that capacitor again?

have you measured the output of power supply when start up the board? 

if you don't have the chip on hand, TI.sample should be able to apply free sample.

thanks.

jesse

Hi Alexander,

Just chiming in here I’m not a TI expert on these but have used this board and built my own versions.  These boards have capacitors that do not have a lot of voltage margin, and the heatsink they use is much smaller than the one on the TI EVM.  I personally think it’s a bad idea to run that board over 48V.  Most places that sell that board usually indicate 48V.     

Regards,

Jerry G

Hi Jesse,

I am here to provide the repsonse from Alexander. Please check and help further. 

Dear Jesse,

Thank you for your attention. In response to your questions:

1 – ‘Is that possible share use the schematic of your circuit’

Unfortunately, I don't have the Ayima board schematic. Let's assume that it is the same as the TI's TPA3255EVM Evaluation Board (at least, there’s high probability to that): 

https://www.ti.com/tool/TPA3255EVM 

In exactly this schematic, is it generally possible (is it safe, to be more precise) to replace two C31 and C46 PVDD 4700 mkF 80V capacitors with the two 10000 mkF 63V JB JNC ones? Under condition that the board is powered from a 48V 7A Switching Power Supply turned up to 50V… 

2 – ‘Can you take a try soldering another chip on the board and take a try that capacitor again?’ 

Yes, theoretically I can solder another TPA3255 chip, but I wouldn’t like to risk burning the fresh chip that without knowing the reason of the initial failure (and that’s why I’m asking your esteemed company:)). 

3 – ‘Have you measured the output of power supply when start up the board?’ 

No, exactly in the moment of starting up the board with the resoldered capacitors I, of course, wasn’t wise enough to measure the power supply’s output power.

However, we have experimented with that power supply and a single 22000 mkF 63V capacitor to see will there be any parasitic short-term power surge in the output during the start-up as a result of such a large capacitance paralleled to the output. Our oscilloscope didn’t show any parasitic impulses – the 50V in the output seem to appear quite clean, without any initial unwanted voltage peaks/surges.

Kind regards,
Alexander

If the picture is correct it does not look like that board is using the high voltage version of the 2576-12 which means your exceeding ABS Max input since the HV input runs directly to the 2576-12.  This also hurts the SNR as the part puts out a lot of noise with that high of voltage in.  If it is indeed not the HV version I would personally swap that part.  Sorry for the intrusion just trying to help.

Hi Jerry,

I am here to provide the repsonse from Alexander:

Dear Jerry,

Thank you for your notice.

Yes, earlier, before the capacitors' replacement in question I had changed this voltage regulator to the' -HV' counterpart. That enabled me to feed 50 volts safely to the amp board from the switching power supply.

Kind regards,
Alexander'

Hi Alexander

   It's a little hard to speculate the real reason by discussing here. Technically, it should be no problem to put large capacitor at PVDD side. Most likely the damage is caused by ESD or some shortage on the PCB. Previously, if we really need to find out the possible reason, we would give this failure device to TI's local CQE, they will help to do some failure analysis like X-ray, or decap and have a close look into the device. 

  I'm sorry maybe not enough help could be provided here.