TPA3251d2

I thought it was out???
if "out" means "general availibitlity", nope.

Of course there are some pre-production samples around. This thread even has a picture ;)

TPA3255 datasheet revision history (page 2) :

9W3iqfv.png


update date :
J6PQ4Pb.png


As of today, TPA3255DDV sample request or purchase order can be placed on TI website, but no shipment date yet.
Mass production not yet reached the shelf.
 
TI published few days ago an appnote to design ClassD LC output filter.
The TPA3251 EVM is used to measure about 20 different inductors.
Worth reading
http://www.ti.com/lit/an/slaa701/slaa701.pdf

Good find. The original link no longer works, here's a related app note: http://www.ti.com/lit/an/slaa702/slaa702.pdf

It has the following blurb:

By increasing the gain of the EVM op-amp input stage, the op-amp output noise will also increase. This causes higher overall system noise even though the self-noise of the TPA32xx amplifier in PFFB has been
significantly reduced. To achieve lower overall system noise in PFFB with higher front end gain, considerations beyond the scope of the EVM must be taken to design a lower-noise op-amp front end with the desired system gain.

Any ideas how the op-amp front end can be addressed?
 
Guys, after a year of hibernation I am completing my TPA3251D2 design for a 2+1 FAST system for desktop listening. I've been studying SLAA719 (analog grounding for TPA32xx) and ensuring I keep the analog grounds for the preamp section (I have a high/low pass filter + Linkwitz transform) well separate from the power section joined at a star point for the main power line filter capacitor.

In the app note I see mention of separation of digital and analog grounds on the TPA chip itself (pins 12/13 being analog ground) with no mention of this in the datasheet and the evaluation board also has these pins connected to the GND track underneath the IC so being shared with the power pins & these pins would 'see' the switching power currents from the output stage. This creates extra noise especially for single ended analog stages before the TPA IC, which is what I'm doing.

The app note has some nice graphs showing the improvement in the noise with star grounding when both the preamp and TPA analog ground are star connected but it doesn't show the impact of the TPA IC analog ground sharing common ground path with the power output stage when the preamp opamps use a star ground. Reason why I ask is that having almost finished routing the PCB it is going to be a pain to try to separate the ground traces for pins 11/12.

I'm thinking of just ignoring it an using star ground for just the preamp stage. Any thoughts or even results of your experiments with grounding to share? Thanks.

I'll post the PCB when completed.
 
Good read,

For TPA3251 differential input, ...

Just for clarity.

TPA3116 two differential amplifiers. If only one side of the differential input is driven then both outputs will produce signals 180deg apart.

TPA3251 four independent amplifiers. Both sides of BTL have to be driven 180deg apart to get output signals 180deg apart.
 

Attachments

  • TPA3116-block.JPG
    TPA3116-block.JPG
    95.7 KB · Views: 598
  • TPA3251-block.JPG
    TPA3251-block.JPG
    98.1 KB · Views: 593
Sorry - but that can't be the whole story regarding the high prices. The TPA3255 EVM is 259 €

You can get european made class d amplifiers, for less that amount, and with proprietary solutions. You can get two Audiopower MDA400S for that price, and this includes PSU, and the Hypex UcD 180 standard is also cheaper than a TPA3255 EVM module.

I don't think TI will support the DIY community, and then there is a little extra charge for that. Also, if you are a developer with and TI account, you can often get samples for free or cheaper.
 
Last edited: