Texas Instruments TAS5261

Status
Not open for further replies.
Yes .... so how do we ...

BRWX: " ... I was thinking of leaving the options open so that it would be possible to use a front end that is discrete, comparator based, or chip based. ..."

Discrete input stages for Class A/B and Current Dump power MOSFET amps like http://aussieamplifiers.com are quite common. The quality & specs. can not be denied.

So the question (as BWRX points out) might be: How do we apply a quality discrete input stage to a Class-D power amp output stage ? ... (or should we?)

....
I have just finished a pair of mono-block MOSFET amps with discrete FET inputs. My test bed can easily be adapted to alternate output types like mono-block Class-D amps, etc. ... that was my purpose here on this thread = build the TI '5261 as output stage for comparisons. ... Got alternate input suggestions? I would like to build several alternatives.

:smash:

( A Pirate Ship run up on the rocks and gutted for its cargo: http://3dotaudio.com/ampics.html )
 
Whatever we come up with, it would ideally work off a 3.3V to 5V supply and should be capable of supplying 0-3.3V inverted and noninverted PWM signals to the TAS5261.

My last suggestion to use some front end chip that can generate those PWM signals from an analog input would be the simplest solution in my eyes. The TI chip I mentioned would meet the above requirements and be easy to implement. It's specs aren't amazing but it would get the job done. Anyone know of a chip better suited to the task? Or a good discrete design that would work with a single rail bridged output stage?
 
pinkmouse said:


Now where's the fun in that? 🙂

I agree, I don't believe that TAS5261, alone, would be any better than UCD or TRIPATH.

Why then TAS5261?

Because together with a TI Equibit PWM controller one can completely avoid ANY digital to analogue conversion. The bet is only there: if without a DAC you get better sound than UCD or TRIPATH. Bruno Putzeys says the bet is lost.

Input for this digital amplifier is very easy: it accept I2S. Most CDP have it, or we can use USB->I2S or SPDIF->I2S converters.

I've not seen any digital amp project finished, here around, except POWERDAC, but it is not diy.
 
thomaseliot said:
Because together with a TI Equibit PWM controller one can completely avoid ANY digital to analogue conversion.

You're not avoiding digital to analog signal conversion, you are merely going from PCM to PWM to analog instead of analog to PWM to analog. The question is whether or not the PCM to PWM conversion can be sonically better than the analog to PWM conversion.
 
Brian, please excuse the ignorance, but I am just learning this digital stuff. I'm assuming that your idea just uses the output of the TPA2000 to drive the PWM inputs. This output would normally be filtered to get the speaker output, but we can use it without the filtering to drive the TAS5261 directly. Correct?
 
I want something like:

PCM4202 ---> TAS5504 ---> TAS5261.

This is for very High Quality, working with 192Khz Sample rates.

TPA2000 should do the job well done to direct analog to pwm conversion (read bellow).


pinkmouse said:
Brian, please excuse the ignorance, but I am just learning this digital stuff. I'm assuming that your idea just uses the output of the TPA2000 to drive the PWM inputs. This output would normally be filtered to get the speaker output, but we can use it without the filtering to drive the TAS5261 directly. Correct?


Great question ! So, should we ignore the inductors and etc, and just connect the speaker outputs to the 5261's pwm inputs? 😕




Cheers,


Fernando
 
The switching frequency would be set by the TPA2000 and can be changed by an external resistor and capacitor.. The TPA2000 datasheet recommends a switching frequency between 200 and 300kHz, which is below the maximum switching frequency of the TAS5261.
 
Al - which footprint do you need (this is for Eagle I assume)? I can't send it to you until I go home (in about 2.5 hours).

-_nando-_ said:
But wouldn't TPA2000 degrade the sound quality? Not in terms of distortion, I'm talking about imaging, life, dinamics and those "audiophile" sound things 😉

It very well could (and it would actually be in terms of distortion), but it's one of the simplest ways to make an analog input stage that outputs the proper PWM control signals.
 
BWRX said:
Al - which footprint do you need (this is for Eagle I assume)? I can't send it to you until I go home (in about 2.5 hours).


Yup, it was your TAS5261 footprint for Eagle. I was culling my mailbox and went a little too far!

No worries about the wait, I was just had an hour or so to play around and I thought I'd have a play. I might well search the standard libraries to see if there is a matching footprint I can adapt. Or I might go make my dinner...
 
Yeah, I agree with your mom !

Since I never made any class D amp, is a good idea to use the TPA2000, but the big problem is that I don't have a store near my house that sell these parts (farnell don't have too), so these parts took a long time to arrive in my house, and here we are obligated to pay 78% of the whole value to the government when something comes from other countries... So I'd like to build something simple that can have a good quality to avoid the government robbery and the time that the parts take to arrive :cannotbe: Anyway... I don't like much the idea of use TPA2000 since it is projected to drive speakers and have it's own power output stage that can real degrade the sound.


If we don't come up with a nice design soon, I'll try TPA 😉


Cheers !
 
Status
Not open for further replies.