/diy/ - Do It Yourself

[Anonymous No.710894]

how do i make my own headphone amplifier?
how should i modify the circuit based on the headphone i will use?

why might i choose, say, class a over b or vice versa? which would be simpler on my brain?

[Anonymous No.710898]

there exists at least one forum devoted entirely to headphone design. your job is to find it, given that I dont see it in my bookmarks.

alternately, you can purchase one already made. my recommendation is the Burson Audio Conductor, which is a steal at only $1700.
http://www.headphone.com/collections/headphone-amps/products/burson-audio-conductor

[Anonymous No.710899]

>>710898
i don't care about quality, i just need to have made it

[Anonymous No.710906]

can someone tell about how much voltage would i need to amplify to what levels for, say, a 32ohm headphone?

[Anonymous No.710910]

why is there a transformer here, can someone explain?

[Anonymous No.710913]

> how much voltage would i need to amplify to what levels

do it yourself using ohm's law
E=IR (voltage = current * resistance)
P=EI (power = voltage * current) or
P=I*I*R or
P=E*E/R

so, if you put 1V across 32 ohms, you get 31mA of current, and power dissipated is 31mW.

if you're starting with an input voltage of 0.1Vrms, which is a typical AUX or LINE level, then to get 1V you need to amplify by 10, obviously.

> why is there a transformer here

it's a silly design. the transformer acts like a power resistor in the class A amp. if you want something that's well documented, google CMOY.

[Anonymous No.710914]

>>710910
To transform transformations. It's actually used as a big inductor so that the audio frequency shit won't go through it, but the transistor still gets its DC bias.

Since you can apparently google schematics, what is the point of this thread? Do you need help with obtaining parts or soldering shit or what?

[Anonymous No.710918]

http://www.instructables.com/tag/type-id/?sort=none&q=pocket+amp

Here you go.

[Anonymous No.710920]

>>710913
>if you're starting with an input voltage of 0.1Vrms, which is a typical AUX or LINE level, then to get 1V you need to amplify by 10, obviously
so you are telling me i'd need a gain of about 10? where would i put a, say log pot resistor, to adjust the volume?

[Anonymous No.710923]

>>710920

>Tries to get into amps.

>Not a single mention to OP amps.

Nigga what?

[Anonymous No.710924]

>>710923
i have to use transistors.

[Anonymous No.710925]

>>710924
I'm sure your course book and lecturer explained how to design such amplifier.

[Anonymous No.710928]

>>710925
i know how to design an amplifier. i just do not know how to make something that will actually work. i do not even know what the input will be. in theory i can analyze the circuit, find values, optimize for maximum voltage swing and all that good shit but as far as making something that will work for a real life headphone goes, i feel completely lost

[Anonymous No.710929]

> where would i put a, say log pot resistor, to adjust the volume?

there are two good places to put it. first is right at the input, like the 50K pot in >>710910
this reduces the input signal. the second, and better, place to put it is where it alters the amplification. this gives you a better signal-to-noise ratio coz you wont be amplifying the noise as much when the pot is set below 100%.

the first method is not as good, but simpler.

[Anonymous No.710930]

> i feel completely lost

if you can design transistor circuits, then the only thing different about a headphone amp is that the output impedance needs to be low, like 10-100 ohms, instead of the usual 5-50K.

[Anonymous No.710932]

>>710930
should the gain be around 10, then?
as >>710913 has mentioned?

[Anonymous No.710935]

I didnt say the gain should be 10, dude, that was just a sample calculation. it depends on your input signal and how loud you want it. assuming you do have a 0.1Vrms input, a gain of 10 would give you 31mW of power in to 32 ohms. is that enough? you would probably prefer 100-300mW so you could hear it outside, for example. so this changes the gain calculations.

[Anonymous No.710945]

The headphone impedance being such a low value is causing me problems. Is it even possible to do this with a single transistor?
There are designs with tens of resistors and 4-5 transistors, these are for better sound quality i am assuming, they are not necessary for something that just needs to work at all?

[Anonymous No.710955]

>>710945
See your OP pic.

[Anonymous No.710970]

>>710928
Google Nwavguy.

This guy built and documented an "ideal" headphone amp while dispelling a ton of audiophile nonsense. Hes got dozens of good articles on design parameters and whatnot for making an amp. Lots of good learning in there, but its very long and might be more than you want to know.

[Anonymous No.710971]

>>710928
So start by deciding on your input? The output from a lot of devices is high enough already, so the amp just acts as a buffer, best used from crappy sources that have high output impedance.

With that settled you can think about the headphones you want to drive. High impedance headphones won't need as low of an output impedance to sound decent. Earbuds and IEMs are usually very low, but fullsize headphones can be all over the map.

Are you using batteries or do you have a power supply?

This should give you a bunch of design parameters to stick on paper and work from there.

>>710945

It isn't always sound quality. It could be just multiple resistors used to get a desired value that isn't standard. Multiple transistors could be used to turn off the circuit when nothing is plugged in, short circuit protection, surge protection, etc.

[Anonymous No.710973]

>>710945
You could add an inductor in series with the headphone.

[Anonymous No.710989]

>>710973
why do i want to do that?

>>710970
thanks

>>710971
>Are you using batteries or do you have a power supply?
I will be using a power supply.
Would a fullsize headphone with a high impedance be easier to design for and sound better with a lower quality amplifier?

[Anonymous No.711052]

> fullsize headphone with a high impedance be easier to design for

high impedance is always good, but a fullsize headphone is more accurate so a shitty amp will sound shitty. earbuds mask some of the shittiness, thus are a better choice.

[Anonymous No.711072]

http://www.amb.org/audio/mmm/

For the win!

Unpopulated boards can be purchased from the site.

Enjoy the build its well worth the money and time.

[Anonymous No.711426]

>>710894
>why might i choose, say, class a over b or vice versa?
class b is not appropriate for audio, or much of anything but digital really. Your choices are A and AB, and the difference is A is simpler, wastes more power, needs bigger transistors and resistors, and is believed by hipster faggots to yield better sound quality.

[Anonymous No.711441]

>>711426
Maybe OP wants nice sixties transistor sound or something. And he can then apply ridiculous amount of feedback as a sad attempt to camouflage the resulting distortion.

With one transistor designs he's limited to class a.

[Anonymous No.711724]

>>710989
>why do i want to do that?

Impedance matching. You'll lose some power to the inductor but at least you would be able to hear sound from your low impedance headphones.

[Anonymous No.711807]

>>710929
Putting the pot into the feedback/gain path is tricky. We have an amp from the 90s (musical fidelity A1) which uses a volume control like that and the crappy pots on the first few batches would break after like 2 months. I'd just put a 10K RK27 on the input.