TDA7231A 1.6W audio amplifier

[timo]

Hi guys,

I had a bad cramp while surfing at Ebay and found myself having bought a bunch of these:

http://www.st.com/stonline/products/literature/ds/1466/tda7231a.pdf

TDA7231A 1.6W audio amplifier

For a total newbie it doesn't look like a bad chip for a Ruby-like mini-amp. It has fixed gain of 38dB but twice the input impedance of the 386 chip. The reference design has only eight parts in it, including the chip. It's in the standard DIP8 packaging I believe. And to top it all it says "soft clipping" (toob-like, eh?)

So, am I onto something here? And any ideas on improving the reference design would also be nice.

[Mark Hammer]

1) Nice little chip.  Of particular interest is the quiescent current spec (how much current it's drawing when you aren't feeding it a signal and asking it to do something), which is stated as 3.6ma typical.  That's pretty damn good.  The LM386 draws 4ma, which isn't much more, but when you get into power chips of comparable output (e.g., the LM380), you're looking at 7ma and more.  What does this mean for you or other builders?  Longer battery life, since much of the amp's "on" time is actually silence and spaces between notes.

2) One robust little sucker.  You could effectively operate it off a 3v lithium coin cell if you felt like it, but could also feed it up to an ampere at 12v.  If temperature is effectively managed, that little bugger can pump out as much as 4W!  Let's see a 380 do that!

While the datasheet does provide the convenience of a layout, consider perfing or vectoring it with a heatsink of some sort and a 12V supply (8 C or D cells).  Feed that into a 10" or 12" speaker and you'd have a seriously loud practice or busking amp.  Certainly potentially loud enough to run off batteries and take on camping trips or to the beach or wherever.

[timo]

Wow, thanks Mark! 

The prospects are looking good I'd say. When the chips get here I am definetly going to experiment with them on the breadboard. While the 4 watts sound intriguing I think I'm going to try them first as a headphone amp for my jazz bass. I reckon it will have no difficulty driving a pair of cheapo 32 ohm Philips headphones. Would 3 volts worth of C or D cells be sufficient for this?

The bad thing about these chips is that there seems to be no documentation on them aside from the datasheet.

[Mark Hammer]

Strikes me they could be implemented into a stereo headphone amp quite nicely, with fairly low distortion.  A pair of C cells might actually provide less than what you want.  If running them into 32-ohm headphones, I think I might want to go with at least 6v or maybe even a sextet of AAs to give a heftier 9v.  The datasheet shows 70mw into 8ohms at 3V which doesn't bode well for 32ohm phones.

There IS precious little in the way of appnotes or projects, but when you consider just how many small power-amp chips are out there for "personal listening devices" it's no surprise.

[timo]

Well, I've now breadboarded one for the first time. The setup was Ruby input buffer (0.1u coupling cap) with 10k linear (I know) pot into the reference design and then into 32 ohm   headphones with both channels wired in parallel (64 ohm load), attenuated with a 47 ohm resistor between headphone and circuit grounds. I also wired up the Ruby power amp for reference, using a 470uf output cap. PS was a 9 volt battery. tried it with a Tokai Les Paul copy and a Fender standard jazz bass.

It scared me. the sound was really soft and smooth, hi-fi like when compared to the Ruby. It was also really, really loud. A 3 inch half watt speaker could not handle it at all, and with the headphones the 47 ohm attenuation (okay for the Ruby) was definetly too little. Even turned down it kinda had an intimidating tone. I'm still feeling uneasy 

[JimRayden]

How 'bout that promised "soft clipping"?

---------
Jimbo

[Timebutt]

Interesting, a head phone amp intended to be used for bass!
Exactly what I need for my room when I go to university next year and don't want angry neighbours

What schematic did you use to get this done? What has given you the best results so far, etc.?
So far it looks very promising

[timo]

How 'bout that promised "soft clipping"?

---------
Jimbo

I can't verify that yet - if I drove the chip hard enough to clip It would probably destroy my headphones AND my head. I'm thinking about lowering the supply voltage for better controllability. Or if I can find usable speakers from fleamarkets.. no luck so far.

Timebutt, check out the ROG Ruby amp. I think it will make a great bass headphone amp with right cap values. (see my last post for ideas). No need for the gain pot on a headphone amp.  I'm gonna build one for myself!

T

[Timebutt]

Will you let me know something once you finished the Ruby as a headphone amp for bass?
I've been doing a lot of reading these past few days about headphone amps, and I would like to be able to plug in my bass in one channel and music (probably from a computer line-out) into another channel so that I can 'play along' with a song I choose on the computer.

The last requires a different kind of amplifier if I'm correct (such as this one, the CMoy: http://tangentsoft.net/audio/cmoy-tutorial/) and I would use a Ruby to amplify my bass signal, but I was wondering on how I have to create some kind of a 'blend' for these two channels. Would a buff 'n blend circuit do what I want or not?

[timo]

Timebutt, your project seems to be a bit more ambitious than mine  -  I'm thinking more in terms of a simple mono amp that I can attach to the strap and rock out around the apartment 

You might be well served with a cmoy amp and a small mixer in front of it.