DIY Headphone AMP, any ideas???

[cterawskyj02]

Hey guys, this is my first post and i am wanting to build a headphone amp, more because i can than any other reason, my PC does OK.
But i am looking for ideas, or if any one has made one or seen a good post on how to make one. Let me know.

[Psi]

My vote is this, because it's just so cool.

Guide showing exactly how it was made
https://www.instructables.com/id/Crystal-cMoy-Free-Form-Headphone-Amplifier/

[cterawskyj02]

Thanks, this is really cool! I will probably do something along the lines of this.
I don't think i'll use resin as i am not that skilled.

[nick_d]

I built something like this using an LM386. There are lots of circuits and tutorials available. This one looks really good (based on a quick skim read):

http://www.circuitbasics.com/build-a-great-sounding-audio-amplifier-with-bass-boost-from-the-lm386/

Mine was a bit special since I used two LM386s to create a bridged output and then trimmed the DC offset with a digital pot. Yours will be much more straightforward. Try building each circuit in the above tutorial, starting with the simplest then adding the extra parts once you get it working each time.

A word of warning: LM386 on Ebay or Aliexpress are fake. Always use Digikey or Mouser for development. Once you get all problems ironed out, you can safely evaluate the fakes in your circuit.

cheers, Nick

[magic]

Nice and cool but better use good electrolytics 'cause there will be no replacing them

LM386 isn't great fit for headphones. Too much gain and crap specs even if you get a genuine one.

The (in)famous O2 amp gets better performance from two paralleled halves of an NJM4556A. Other high current opamps often seen in HP outputs are RC4580/NJM4580.
Not sure how NE5532 or LM4562 would perform. Lower output current but technically better opamps, may possibly maintain reasonable distortion even close to their current limit. If they don't run out of output current, that is.
I think there are also some high current options in the LME49xxx and OPA16xx lines, but pricey.

IMO a CMoy or some variation thereof (like the O2) is the way to go.

[nick_d]

For the specific case of headphone amplification, a CMoy might be the way to go. I just read some articles about them and they seem interesting.

On the other hand, I don't see a problem with the LM386. The gain is settable with external components, the built in gain options are just to save parts in common cases from what I understand.

As to amp specs, I am not really sure what the fuss is about. I built a lot of amps with LM358 or LM324 back in the day and never had a problem with them. I am aware they have horrible crossover distortion, I myself cannot hear it.

Nowadays I would use LMC660 just to be on the safe side... now NE5532? That is hifi man! So fast, such low noise and distortion... well I innocently suggested this op amp to somebody in another thread and was told it is super noisy and whatnot. Ah there is no pleasing some people. It may be >20 years old and sure there are some amazing op amps released since then, but could you hear a difference in this application? For a re-mastering studio it might be important.

I recommend the LM386 as a beginner project because (1) low parts count (2) designed as a power amp (3) many tutorials available (4) more versatile as once built you can use for speakers too.

Nick

[jeffheath]

I made this with a 60 cent audio amp ic... I really just wanted to have a composite stereo (red & white) to headphone amp for as cheap and simple as possible. I was told it would be better to have the volume pots on the output side, but I don't have time to experiment.

[Zero999]

tech_builde, along with many others, designed a guitar headphone amplifier which worked quite well. It's mono only and not suited for your application, but with a few modifications, it should work well.
https://www.eevblog.com/forum/projects/guitar-headphone-amplifier/msg1248873/#msg1248873

[calexanian]

The first suggestion there looks pretty cool. There are lots of very good op amps out there these days. I still however am working of some tubes of regular old 4558's I got some years ago. +/- 5 volts and one of those (I have never liked using blocking caps and single sided supplies in the headphone amps I build for customers) and you are in business. Its amazing how much more headroom and cleanliness you get out of a simple circuit like that over some lame computer headphone output for just a few dollars of jellybean parts.

[cterawskyj02]

Thanks, I'll take a look at some.

[John B]

As to amp specs, I am not really sure what the fuss is about. I built a lot of amps with LM358 or LM324 back in the day and never had a problem with them. I am aware they have horrible crossover distortion, I myself cannot hear it.

You must have loaded the op amp in such a way to keep it in one conduction mode. Crossover distortion has a lot of high frequency content and sticks out like a sore thumb. Always a good idea to feed in the maximum amplitude signal you expect in order to make sure that even the very crests of a signal aren't starting to distort from crossover distortion.

Anyways, on the topic you might get some ideas from this place:
https://www.jlmaudio.com/shop/amp-head-ii-phone-kit.html?display_tax_prices=1

Looks like they are using ne5532 op amps, maybe in parallel.

Been meaning to pick up a kit myself.


Just don't parallel them like this:

http://www.barberelectronics.com/parallelopamps.htm

 

[DaJMasta]

I dunno, a "muscular" sound color may be just what the OP ordered in which case I think soldering opamps in parallel is probably the only thing I've heard of that will make that claim 

[Conrad Hoffman]

A parallel design with some LME opamps would be good. Or, do a simple class A design. There are a ton of ideas on the 'net, with fans and detractors of all of them.

[SiliconWizard]

You really need to define your requirements in order to select an adequate design, particularly in terms of max. output power and typical headphones impedance range.

Using a single opamp will not always be adequate. Unless you specifically use an high output current one, most will not be able to deliver more than +/-20mA or so (if that) without an enormous amount of distortion. With typical 16-ohm headphones, that's about 6mW! So, that could be enough for small earbuds, but with standard headphones (especially the lower impedance ones), the output level will be very faint.

Paralleling several opamps can work (with some precautions), but frankly unless you're targetting something very specific, I wouldn't bother, and I'd use an off-the-shelf headphones amp such as the one jeffheath suggested (there are many on the market, either class AB or class D): they can put out a lot more power and usually have integrated protections and such.

For a discrete design, class A is one of my favorites for that. Not efficient at all energy-wise, but will get you very low distortion with very few parts if done right.

[Bassman59]

Hey guys, this is my first post and i am wanting to build a headphone amp, more because i can than any other reason, my PC does OK.
But i am looking for ideas, or if any one has made one or seen a good post on how to make one. Let me know.

Use the OPA551. It is available in DIY-friendly 8-pin DIP. You need two, as it's a single amplifier chip. It has enough oomph clean the wax outta your ears.

[rdl]

There is some useful info here:

https://tangentsoft.net/audio/cmoy/

[wilfred]

There is this one available as a kit from a Silicon Chip magazine (Apr 2011) project. https://tronixlabs.com.au/kits/audio-kits/amplifier-kits/silicon-chip-headphone-amplifier-kit-australia/

http://archive.siliconchip.com.au/cms/A_112409/article.html
http://archive.siliconchip.com.au/cms/gallery/article.html?a=112409&i=2       (schematic)

[RES]

http://sound.whsites.net/project113.htm

Impedance   Power (Direct)   120 Ohm Feed 8 Ohms   1.5 W   35 mW 32 Ohms   2.2 W   99 mW 65 Ohms   1.1 W   136 mW 120 Ohms   595 mW   149 mW 300 Ohms   238 mW   121 mW 600 Ohms   119 mW   82 mW

[Doctorandus_P]

LME49600 Headphone Amplifier Evaluation Board User's Guide

AN-1768

[David Hess]

There are a couple of ways to use individual operational amplifiers in parallel to raise the output current.  Putting them directly in parallel would not be my first choice.

I have used a variation of the circuit shown below for small audio amplifiers.  The common emitter output allows close to rail-to-rail output voltages.  The current feedback output structure allows the use of slower audio amplifiers in audio applications because the fixed voltage gain of the output stage multiplies the operational amplifier's slew rate and full power bandwidth.  The circuit does require some understanding of the theory to get best results however.