No-gain diamond buffer headphone amplifier advice

[BuckarooBanzai]

Hello all,

I came across a nice, simple circuit for a low distortion buffer that I'm thinking of using as a unity-gain headphone amp. I already have most of the parts in my bin, and it's straightforward enough for me to tackle.

My DAC output is pretty juicy, so I really don't need gain. I've attached the schematic below. What do you guys think of it?

The only thing I'm unsure of is what D1 and D2 represent. Are they constant current diodes? If so, what is the "2.0" likely to mean? Surely not amps . Is the 47uF input coupling cap a bit much? From what I've read it doesn't need to be anywhere near that big.

 

[hta95]

I think d1 d2 are LEDs. 2.0 means 2V forward voltage, it's a green led i guess.
In addition, 2 pairs of 8A current rating output transistor might be overkill

 

[Mooly]

They will be constant current diodes, probably 2ma. LED's would just turn on J3 and J6 destructively. They could be replaced by resistors but you would have to calculate and simulate it. It would work though.

These are basic circuits and you will find big differences in simulated values and real world components. The DC offset could be disappointing without individual tweaking or adding a servo.

 

[AndreasM]

Hmmm..., parallel output transistors and 24V supply voltages, I would estimate output power about 30W in 8 ohms. IMHO plenty oversized for a headphone buffer.

They will be constant current diodes, probably 2ma. LED's would just turn on J3 and J6 destructively. They could be replaced by resistors but you would have to calculate and simulate it. It would work though.

O.K., but why zener diodes in forward direction are used?

47µ for electrolytic coupling cap isn´t that unusual, but you will find a lot of different and controversial opinions and discussions about this topic.

 

[Mooly]

Its a zener symbol although the polarity is correct for a current diode though. Maybe the software doesn't have the correct symbol for a current diode.

 

[jam]

Something else to try if you don't have current diodes.

 

[mlloyd1]

hi jam;

just for grins, why not a little bit of cascoding, tying Q4 collector to Q20 (or Q1?) emitter and Q3 collector to Q19 (or Q2?) emitter.

mlloyd1

 

[BuckarooBanzai]

30W you say? Hmmm... since I lack constant-current diodes, and don't yet have plans to strap a pair of full-size speakers to my head I'll give jam's circuit a try . What supply voltage would be best?

Are there any other headphone buffer recommendations? I'll also do some searching.

 

[jam]

Michael,

That should work if you want to try it.

Hi Buckaroo,

15 to 18V split rails should work fine.

Jam

 

[jam]

Michael,

Since you like complexity how about this...............

or if you are really into pain the second one.

Jam

 

[mlloyd1]

i'm a chicken,
i'd be too worried about #1 being a headphone oscillator.
#2 is interesting; i'll pass on the filter (pardon the pun ), personally, i'd like a little more beef for the outputs.
i feel like i've seen something very similar before ... time to visit the archives ...

mlloyd1

 

[jam]

Michael,
They all came from this site i believe.

By the way I finally built a version of the pre-amp we were discussing a few years back. E-mail me and I will send you some photos.

Jam

 

[Gopher]

jam - what's the -3dB point of that 3rd order HP filter? Is it intended to feed a subwoofer or is it a rumble filter?

 

[jam]

That was just an example of a buffer. I would leave the filter out.

Jam

 

[mlackey]

A classic:

http://waltjung.org/PDFs/Op_Amp_Audio_Buffers_2_ED_100198.pdf