Diana needs help about NCS2001 amp

Hello,
Can anyone help me on my little project herein. I'm using an rail-to-rail operational sub-1V amplifier
NCS2001SQ2T1 (ON Semiconductors), for ampifying very low (mono) audio signal from an old reciever. Ampifyer is powered by single 1,4 button cell. Earphones are on output.
Since I'm still a beginer in electronics (or complete retard , I don't know which pin on the
NCS2001SQ2T1 should I connect with bat-, bat+, ear1, ear2, input1 and input2 contacts. Because of very small gabarits needed (therefore I use button cell), it is desirable the whole circuit has as minimum aditional components (SMD resistors and capacitors) as possible, regardeles the sound quality.
No volume potentiometers also; I need maximum output power that NCS2001 can provide. Simple as that. But not for me !
Btw:Could I use 2 or 3 IC-s for aditional amplification ?
I would very apriciate a simple scetch if possible. Thx.

NCS2001 links:
https://www.onsemi.com/site/products/summary/0,4450,NCS2001,00.html
https://www.electro-tech-online.com/custompdfs/2004/06/NCS2001-DPDF.pdf (127KB)

What is the impedance of your headphones?

Your amplifier has 86 dB gain with 0.9 volts VCC and 2K load, so it will not need any more stages. The gain of this circuit is: headphone Z/RIN. You will have to compute the value of RIN based on the headphone impedance and input signal level. The input capacitor, CIN, should have low impedance relative to RIN.

First of all, thanks for quicq reply
Impendance of the earphones I'm planing to use is 32 ohm.
Btw: what about bat cappacity - how long would it last ?

Current drain from the button cell will be about 1.7 mA worst case using 1000 ohm headphones. The 32 ohm headphone is too low impedance, the amp will not drive it any where near rail to rail. A 1000 ohm headphone is not ideal, but the highest I found, available from Digikey, part number: 25DE139. You can look up the ma-hours of your button cell to compute the battery life.

Russlk said:

Current drain from the button cell will be about 1.7 mA worst case using 1000 ohm headphones. The 32 ohm headphone is too low impedance, the amp will not drive it any where near rail to rail. A 1000 ohm headphone is not ideal, but the highest I found, available from Digikey, part number: 25DE139. You can look up the ma-hours of your button cell to compute the battery life.

Click to expand...

Russ, you're right about the amplifier not being able to drive 32 ohms. My concern with your circuit is that the source may not be able to drive the load, even if it's 1k ohm. All the headphone current has to flow through the source. Diana wants lots of gain, and the gain of your amp actually is

Av=-(Zphone/(Rin+Rs)), where Rs is the source impedance. If Rs is more than a few ohms, she won't get much gain.

Maxim has MAX9721, which is a stereo amplifier, but it has fixed gain, so additional gain would be needed on the front end.

Russlk,
That's a good idea to use the headphone as the feedback resistor and avoid using an output cap. As an inverting amp, I'm trying to imagine what overdrive would sound like where the zero-crossings would be loud, but the clipping would have much less gain since the input would directly drive the headphone through the input resistor on the peaks. That will probably sound weird, I'll try it.
Your circuit will have a reasonable input impedance if you ground your input and make a new non-inverting input that is cap-coupled to the bias divider.
But how little volume will be produced by the 1K headphone with only 0.5VRMS max? That's only 250 microwatts on the peaks.

I have trouble foloving you...

Ron makes a valid point, the source impedance may limit the gain, so I added a unity gain follower (it could have gain also, I suppose). The sensitivity of the headphone is not given in the catalog, but higher impedance units should be more sensitive than low impedance units. I used to use 2K headphones with a crystal set which was microwatts at least.

Thanks Russlk,
If I understand well, you suggest me a circuit with two NCS2001 ICs.
I hope it will do the job I'll let you know.
Thanks again.

audioguru said:

<snip>Your circuit will have a reasonable input impedance if you ground your input and make a new non-inverting input that is cap-coupled to the bias divider.
<snip>

Click to expand...

Here is a schematic with that modification. It does require a large value cap, but it can be low voltage. The low end cutoff frequency could be improved by making the value larger, like 220uF.
I've set the gain at about 100, but if it is too high the 10 ohm resistor can be make larger, and the 100uF cap could be proportionally smaller. Higher gain will make the bandwidth suffer, and the polarized cap may get unreasonably large.
In Russ's previous schematic, gain could be added to the voltage follower.
As was pointed out previously, this amplifier is not suitable for 32 ohm headphones, as it can only drive a milliamp or so.

Thanks Ron,
I think a large value cap is not quite suitable for my application, because small dimensions of whole device is important.
A circuit Russlk suggested seems more appropriate, concerning dimensions. Is that circuit represents a bridge amplifier (with two NCS2001)? Will it do the yob ?