PMP10215: Transformer Wuerth, 750315800 and Flyback-Transformer 7508110338

Hi Paul,
Thank you for your inquiry and I apologize for late response.
I'm working internally to get right person to answers to your question.

Thank you for your patience.
Regards,
Yuki

Yuki, Paul,750315800r02 T1.pdf7508110338r00 T2.pdf

Both magnetics are custom from Wuerth. I attached the datasheets accordingly.

Please contact your local Wuerth representative to get samples of them.

Best regards,

Roberto

Dear Roberto,

thank you for your answer. Please let me ask some more questions about reference design PMP10215:

1. I noticed the thermal analysis of that design in document tudubs3.pdf (page 18). Temperature especially of T1 (and L3) is very high. It even seems to exceeed temperature range given by Wuerth (+125°C) for their magnetics. Temperature of the seconday side's recitifier area is high as well.  It seems to me that some kind of forced cooling is necessary for this design. Otherwise it has been mentioned that full load is not a typical application for "audio power supplies". Are those very high temperature areas just because of the rather arifical "full-load-for-some-time-condition" or are the parts chosen a little too economical. As the magnetics are custom magnetics it would have been easy to chose little bigger ones?

2. Apart from the magnetic components secondary side rectifier seems to be responsible for a large amout of heat. TI always uses traditional Schottky diodes (here MBR20200CTG) for rectifying. Wouldn't it be advantageous to use mosfets as well or are there problem with their switching characteristics for audio application?

3. It has been mentioned repeatedly that it is not necessary for audio power applications that the power supply can provide full power continously. In this design it is said that it can provide more than 4x the power of its continous power for 10ms (200 W continous, 840 W for 10ms). What determines that large peak power the most? Slow reaction of current limitation, a large output condensor, no saturation within the magnetic components, heat design ...? And how does the output of the power supply look like when it provides power that is considerably higher than its contionous power. Do the output degrade considerably.

4. Compared to the SLOU293A audio reference design, PMP10215 does not use LLC resonant topology anymore. Is this just a coincidence or is there a reasen for this?

Thank you for your effort

Paul

Dear Paul,

Here are the answers to your questions:

1. This power supply can be used at maximum load (200W) for short time, I would say 10min. 30min maximum, then derating or air convection should be applied.

This is because in audio applications, we need peak power but the average is typically very small. If you need 200W or more continuously and keep the temperature low, then you need pretty bigger magnetics.

2. You are right: synchronous rectification can be used here; I didn't do that because my specification was focused on low cost solution.

3. Correct: normally in audio applications we have high peak current and low average power. The average power is limited here by thermal performance (there is a thermal protection, which reduces Vout @ high temperature) and peak power is limited to protect active components and to avoid saturation of DC/DC inductor.

4. I preferred to employ 2-Switch Forward compared to LLC to keep the frequency constant and avoid beat frequency that might interfere with class-D audio amplifier: there are some amplifiers that are more sensible than others.

Best regards,

Roberto