Another headphone amplifier... Advices pls.

[Dxvideo]

Hi all,

I found a scheme in somewhere (I dont remember where). It looks simple and good working one!
However in the simulation (5spice) it doesnt work! Or I made a mistake...
May you say something about it?

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

Thanks in advance...

 

[Jan Dupont]

Don't know this circuit, but can't see why it shouldn't work.....
Did you try replacing the LED's with Constant Voltage Sources in your spice model ?? Many simulation programs have problems handling LED forward voltages......

 

[Dxvideo]

Youre right...
I've changed the led with GAAS Red Led (formerly GAS Led) and everything is ok now...
Thanks a lot.
Can you you say something about the circuit?
The CVS s look like very fine. What about the BD139/140? They sounds good? I will use that instead of a TPA6120 amplifier project (because I cannot manage SMD things ... Do you think it sounds like TPA?

 

[Jan Dupont]

As I wrote, I don't know this circuit. And I don't know the TPA you mention...... Maybe someone else can help you there?

However I must say that I only use BD139/140 to pull relays and other non chritical places.

 

[gmilitano]

That schematic is from Silicon Chip Magazine.

 

[AndrewT]

Hi,
does the circuit resemble one of the buffered opamp examples that Walt Jung discusses in the opamp book and in one of his papers on audio?

 

[Conrad Hoffman]

Armchair guess/opinion, worth every cent you've paid for it, but it looks like a very straightforward opamp buffer circuit. It should work with almost any parts, but watch out for high speed/slew opamps- you may need to fool with the compensation to make it stable. I'd also investigate the bias current by watching the voltage across the emitter resistors- is it what you want, and is it stable with temperature? I'd try to use a film cap on the input. IMO, it's sufficient effort to build anything and get it in a box- the circuit itself is almost trivial by comparison, so I'd tend to go with a bit more complexity, or with a class A design.

 

[Dxvideo]

In fact I have a class A SE design with opa134+IRF510 hexfet. It sounds really good. However its a very hot guy. And I dont like that!
If we come to your analyses;
About the frequency compensation.. Do you mean the 1n2 on NFB line + 47pF on input not enough for that role?
And I am worried about the thermal runaways also... So if I make a thermal connection between BDs+1N4148s+4.7Rs on the same heatsink. The thermal compensation will be done?
Thx..

 

[KSTR]

@Dxv:
I'd say a TPA6120 will outperfom that circuit easily.

@AndrewT:
It looks like a somewhat strange implementation of a regular current dumper output stage (and one is somewhere in W.Jungs PDFs, can't find it at the moment)... normally we would have a resistor instead of the back-to-back diodes, or at least parallel to them. Biasing/linearity of the buffer stage look a bit sketchy to me, at a first glance.

- Klaus

 

[peranders]

Youre right...
I've changed the led with GAAS Red Led (formerly GAS Led) and everything is ok now...
Thanks a lot.
Can you you say something about the circuit?
The CVS s look like very fine. What about the BD139/140? They sounds good? I will use that instead of a TPA6120 amplifier project (because I cannot manage SMD things ... Do you think it sounds like TPA?

I have made both an amp similar to the one you mention and also a TPA6120 based one.
Here and here

Personally I prefer the TPA6120 based amp but is made of mainly SMD parts which may cause some people trouble.

 

[Dxvideo]

I have made both an amp similar to the one you mention and also a TPA6120 based one.
Here and here

Personally I prefer the TPA6120 based amp but is made of mainly SMD parts which may cause some people trouble.

Hmmm..
Your (semi)discrete amplifier is better than mine.. Its more complex and I think it has better sound..
Anyway, I cannot read the schematics of your TPA6120 amplifier, and also cannot download the pdf file (because of non membership I think).
Youre too experienced diyer on TPA6120 (at least better than me!) So may be you can help me a bit;
I will use that headamp, after my PGA2311 project (I am working on). Which is;

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

I will split the output of PGA to a gainclone and this headphone amp. As you can see, I have a buffer (LM4562) before PGA and PGA also has an opamp inside.
Under that conditions; should I put a buffer before TPA like yours? Isnt it too much for a preamp? One input buffer, one PGAs opamp, one TPAs buffer and one TPA makes four opamp on signal line...
And a commercial question; I saw some price informations in you web site.. What rate you use? May you put prices in Euro? Because I cannot use SMD devices, and may be I can purchase one of yours...
Thx..

 

[peranders]

You'll need a buffer in front of the TPA6120 only if the driving impedance is rather high or unknown (not in your case).