Opamp capable of driving low z load.

[diy didi]

HI! I'm currently building a headphone amp using NE5532's. They drive my 32ohm phones quite ok....
I was wondering if anyone knows of an opamp that can deliver a bit more drive to my phones.
Any help will be appreciated.

[mctylr]

While there are op-amps (and buffers such as video buffers from TI, Analog, etc.) that have higher power handling, you could also go with a single push-pull transistor power amplifier after the op-amp's output. Some exaples would be Analog's BUF03/BUF04 which are obsolete, AD811, AD817, OPA633, LT1010.

See: <http://waltjung.org/PDFs/WTnT_Op_Amp_Audio_1.pdf>.

Transistors for power stage amplification include, but are not limited to, 2N5089/5088, 2N4124/4126, MJE170/MJE180, BC556B/546B, or BD139/140 for NPN/PNP pair.

Hope that is useful for you.

[veteran]

You can also use OPA552 or LME49610 buffer for low impedance headphones.

[AndrewT]

Originally Posted by mctylr

See: <http://waltjung.org/PDFs/WTnT_Op_Amp_Audio_1.pdf>.

particularly the series he compiled on composite opamps.

[5th element]

There's always the TPA6120

http://focus.ti.com/lit/ds/symlink/tpa6120a2.pdf

But I wouldn't want to use it without a good PCB.

[AndrewT]

Originally Posted by 5th element

There's always the TPA6120

http://focus.ti.com/lit/ds/symlink/tpa6120a2.pdf

But I wouldn't want to use it without a good PCB.

Impressive specification.
Does fig34 demand a ground plane? Or can a very short ground trace remove the risk of oscillation? If so then how does this need to engineered?

[jackinnj]

There's a Head-FI thread on the TPA6102. http://www.head-fi.org/forums/f6/my-...pa6120-274337/

I know this is anathema to the group -- but there's an evaluation board for $50! (They save you so much work, and you can see how the folks at TI got it to work without singing!). $50 is less than it would cost to get the parts together and have a PCB done by expresspcb.com.

[jcx]

I've built a multiloop prototype with LT1210 30 MHz 1A cfa output in TO-220 on a clearance hole gnd plane perfboard with apparent success ("tombstone" ceramic smt caps to the power leads)

high speed prototyping is possible but tedious - usually I don't find the energy to make a second channel

http://www.linear.com/pc/downloadDoc...28,P1219,D4138

a 'scope with a few times higher bandwidth than the fastest part is real helpful

[diy didi]

Thanks! this is surely a gr8 help.!!

[5th element]

Originally Posted by AndrewT

Impressive specification.
Does fig34 demand a ground plane? Or can a very short ground trace remove the risk of oscillation? If so then how does this need to engineered?

I see what you mean, it does imply that it could work without a ground plane - after all you do remove the copper beneath the input and output pins as it is. The only thing left is the powerpad.

Originally Posted by jackinnj

There's a Head-FI thread on the TPA6102. http://www.head-fi.org/forums/f6/my-...pa6120-274337/

I know this is anathema to the group -- but there's an evaluation board for $50! (They save you so much work, and you can see how the folks at TI got it to work without singing!). $50 is less than it would cost to get the parts together and have a PCB done by expresspcb.com.

$50? The price quoted by TI is $99, either way you can just look at the PCB design in the evaluation modules datasheet -

http://focus.ti.com/lit/ug/slou169/slou169.pdf

where you can indeed see that most of the ground plane around the chip has been removed.