can this headphone amp be such a ****? or is there something wrong in my eyes?

[facundonu]

hi guys, i was taking things apart from a "crown mustang" dvd player that was dead, to use in future reparations.
and i found that the headphone amp (it had a jack in the front of the equipment) was a separate board with a 5 pins connector marked as "Lin, Rin, GND, +9, -9".
BINGO!!! i needed a headphone amp (not aundiophile, not even very good sounding) and this one was perfect. just a power supply and it would be working.
i tried the amp, and it sounded ok but volume was kind of low.
i decided to take a look at the circuit... and i can't believe it can even produce some sound.
the design is awful! just a couple of opamp stages (jrc4558d) to drive the 32 ohms load.

An externally hosted image should be here but it was not working when we last tested it.

i put a 741 beacause i didn't have the 4558 pspice model.

am i losing something in the way this circuit works or is it really that bad?


edit...
those 2 5k variable resistors is the way i represented the 10k volume pot.
this is one channel, the other is identical.
simulating its behaviour with pspice i see that, with pot at max volume, the whole circuit is a 2 dB attenuator. and i think it's not considering the fact that a 32 ohms load will lower the output even more.
thanks a lot

 

[jcx]

most consumer equipment uses simple op amp headphone outputs

better op amps for low Z drive are available, I would have expected to see jrc/njm/rc4580 in low cost headphone amp

if you plan to replace the op amps you can do much better - depending on package - many of the better new op amps are only offered in smt

building your own with op amp + buffer is easy

http://headwize.com/projects/index.htm

for many options descrete, op amp, and tube

 

[rpapps]

Hi
Why not forget the simulator, power it up, feed it some good audio and let your ears decide?
Cheers
Rob

 

[facundonu]

i've done that and that's why i say i'm surprised this circuit can even put out some sound. as a matter of fact, it sounds well. but the volume is kind of low.

thanks for the 4580 recomendation! i didn't knew that opamp. it's a little more expensive than a regular one, but can certainly deliver significantly more power into the load.

i guess i'll change the opamp of the second stage for a 4580 and see what happens. maybe alter some resistor value in order to increase circuit's gain a little bit.

thanks a lot guys.

 

[Stocker]

changing the resitors to alter gain is a good idea to experiment with, just be sure to watch the output power stays in tolerance... I'd hate to have you report a plume of smoke from a tiny hole in one of the amps...

been there, done that, thrown away the chips...

 

[raromachine]

So there are 2 inverting stages?

The first one gives something like 2.5x inverted gain -(10k/3.9k), and the second .245x inverted gain.

I'm taking a leap past what I understand here and saying that 2.5 * .245 = .6x total gain?

Change that 22K resistor for something lower perhaps? or replicate the first 3.6K and 10K setup across to the second set for more gain. I've no idea how this would affect the noise.

 

[dnsey]

I think your mistake is in assuming that this is a headphone 'amp'. It's clearly a buffer with below unity gain - not intended to amplify.
As has been suggested, you can try increasing the gain, but you may well be better off starting from scratch.
I understand your delight in finding what appears to be what you want already made up, but unfortunately such things are rarely ideal - I have a large box of 'useful' working sub-assemblies from various equipment, but none of them have turned out to be quite right...

 

[facundonu]

i've replaced the 22k resistor for a 18k one, replaced the first 4558 for a 5532, replaced the second 4558 for a 4580, and the "amplifier" turned out quite ok. with 1vrms input i can get more than 20mV in the output, and without pushing the 4580 to the limit.
got what i needed.
thanks a lot all of you, been very helpful.

 

[sreten]

Hi,

32 ohm is now the standard load for headphones with i-pods and
the like. 32 ohm is needed due to the small voltage swings available.
To drive these no voltage gain is particularly needed, input depending.

Op-amps are more suited to directly driving older 600 ohm headphones.

I'd change the last op-amp to a small power amplifier e.g. :

http://www.national.com/pf/LM/LM386.html

/sreten.

 

[facundonu]

i thought about using a LM386, but that "minimun gain=20" was a turn-off.
with a jrc4580, assuming "I,out,max=50mA" on 32ohm, i could get 80mW. i think it's more than enough for me.
thanks!