Foreword

This article is designed for novice solders or people with absolutely no experience, since the device in question is quite simple and does not require any training. People more or less advanced here, perhaps, nothing interesting will not open.

Despite the simplicity, the device sounds quite decent, but it reinforces everything perfectly, so if you want to do something with your own hands, without spending a week on it, you are welcome. If you just need an amplifier for headphones, then go and buy. It will be cheaper and easier.

What for

Like many developers, I like to work in headphones. This creates a necessary barrier between the brain and conversations / telephone / pr. office noise. Not so long ago I bought relatively good AKG 272 HD headphones, and I stuck them in SB Audigy 2. The sound picture as a whole pleases and gives all sorts of pleasures, but one moment came to light when listening to not very modern music, music not affected by volume wars , as well as compositions that do not suffer from painful compression and recession, the volume of the sound is not enough, even when the sliders are turned up to a maximum. This is probably due to the sensitivity of the speakers used.
')
To resolve this issue, I decided to spend a free evening to build a small headphone amplifier.

Such devices are not uncommon, and they are quite a lot at the price of "from $ 15", mostly made in China. Such a purchase personally seems to me a) unsporting and b) difficult to predict by the sound quality. What is inside the box for $ 20 can be found out only after receipt, and on which chip the marking of the pathetic audio amplifier is marked - it will always remain a mystery forever. Walking around the local shops and listening to different models is banal laziness and a pity of time.

Requirements

I did not present any special requirements for the device, the main points are as follows:

• stereo;
• external power to not bother with batteries;
• moderate overall dimensions;
• sound quality is above tolerable (given that listening in headphones does not make sense to build a hi-end);
• simplicity of the scheme.

Under these requirements, the CMoy Pocket Amplifier is thousands of times proven. There are quite a few English-language resources on the Internet with an analysis of this scheme and details of its work. Russian-language information for some reason is not very much.

Under some exceptional headphones and a sound card, this scheme, perhaps, does not fit, but for normal "home" equipment is quite suitable.

The original scheme uses a relatively high-quality (and quite expensive) Burr-Brown OPA132, which, among other things, is also quite rare in our area, but there are a number of alternatives (link to testing at the end of the article) so you can safely assemble. Other rare or expensive parts in the scheme is not used.

Training

The device is shown in Fig. 1. More precisely, there is a power supply circuit and half of the amplification circuit is for one channel.

Fig. 1. Power circuit and single channel

As you can see, the scheme is extremely simple. Without further ado, we dilute the fee as it will, according to the size of the parts available. I did it as shown in fig. 2

I note right away that I collected the device from what was at hand — resistors 0.5 W 5%, high-voltage film, large electrolytes. The small box was not on hand either, so I was not shy in size. If desired, all this can be done three times less.


Fig. 2. Sketch PCB

Now we collect the details - on the bottom of the barrel or in the store.

Useful tips purchased for the first time:

• do not forget to buy the socket for the chip in the dip 8 case (and, accordingly, buy the chip itself in this case). The socket is needed to easily change the chip and pick them up by ear, well, and overheat it with a soldering iron will not work;
• resistors will go 0.25W. The more accurate the better, but without fanaticism;
• electrolytic capacitors to take on the voltage of 35V and above;
• if you do the LED - do not forget the holder for it, the corresponding diameter.

Now you can proceed directly to the manufacture.

Board making

Using any convenient method you need to make a board. I used LUT , I already have experience and a great combat iron:

Fig. 3. Iron

To etch in this way will need the following devices and materials:

• Foil textolite;
• Hacksaw or drill with a cutting wheel;
• Laser printer;
• Glossy magazine (better without women, so as not to be distracted);
• Iron, which is not a pity;
• Fine sandpaper;
• Acetone;
• Rags or cotton pads;
• Ferric chloride;
• Pickling vessel;
• Sink, soap.

I will not give detailed photos of the process (it is quite difficult to say and show something new about LUT, there is everything on the Internet). In short, the process is:

1. Cut a suitable piece of PCB. Ideally - to process the resulting edges with a file, it is easier to evenly iron later;
2. We clean the copper layer with fine sandpaper, wash it with soap and water, and try not to touch it with our fingers anymore;
3. We print the layout of a board on a page from a glossy magazine with a laser printer;
4. Cut, apply textolite, cover with a sheet of white paper, and iron with a hot iron until the paper turns yellow (this is an indicator that the textolite has warmed up sufficiently). Iron diligently, evenly, without forgetting the corners. Overheating of the textolite should not be - the printing will blur, and the textolite may deform;
5. We put the textolite (carefully, it is hot) with the stuck picture in a thick book (unnecessary!), And we become all the weight on top for a couple of minutes;
6. We take out, throw in a container with hot water (I use a deep plate) for 5-10 minutes;
7. Under the tap with hot water, carefully peel off the paper, "pellets". In the end, there will be only black tracks;
8. We study. If it turned out really bad - we wash everything off with acetone and go to step 3. Minor defects can be straightened - remove patches with a knife tip, and fix bad tracks that are badly stuck - for example, take a nail polish from a friend, conveniently and firmly;
9. Dilute the solution of ferric chloride in a suitable vessel. Important! Observe safety precautions! First water, then iron, and little by little. You can not pour iron with water - if ferric chloride is good, a very large amount of heat will be emitted. We breed to color of strong black tea;
10. Shivering, waiting until the copper comes down. It is better to do this in a ventilated room, in heat (the higher the temperature, the faster the reaction rate);
11. Take out the board, rinse thoroughly and rinse the powder with a cotton swab / disc with acetone. That's all, the board is ready!

After that, you can begin to drill holes. As a result, I got such a board (try not to overdo it, as it happened with me. This is uncritical in this particular case, there are no large currents here, and charged particles will find where to move, but still):

Fig. 4. Bored out.

Precautionary measures:

• Do not use a good iron - scratch the entire sole and get a scolding from its driver;
• When working with ferric chloride, observe the precautions. Avoid skin contact, inhalation of vapor, chloride of ferric chloride on metal objects.

Then we charge the fee and solder the parts:

Fig. 5. Tinning


Figure 6. Fee Assembly

When assembling, I immediately sealed resistors R5 (designations according to the scheme) to reduce noise, but after testing I refused them, soldering jumpers. The rest of the original scheme did not depart.

I didn’t do the power switch and the LED, because battery power and energy saving are irrelevant for me, and the device will work from an external power supply.

As seen in fig. 6, I used the ALPHA dual sweat and a nest insulated from the case for power (this is important).

I also used 1/4 "connectors, simply because they are in my turn, like the corresponding plugs, although in this project it is more logical to use 3.5 mm connectors. My copies are not isolated from the case, and I even got an earthen loop in this way, however this did not affect the noise level in a noticeable way.

Connect, twist the knob, enjoy the result. Settings are not required, except to try different types of OU and supply voltage. Now you can take up the case.

Housing

At hand was such a box:

Fig. 7. The housing Gainta G0124.

Buy this can be, for example, here . The material is silumin, it is processed extremely easily and pleasantly. Ten minutes later we get the result:

Fig. 8. Bore body

Actually, that's all. We stuff the connectors into the case, cut off and solder the excess wires, sculpt something non-conductive on the bottom cover (on the inside) (so as not to put anything on the board), and we get the finished device:

Fig. 9. Ready device

In total, it took about 3-4 hours to produce from the search for the circuit to the inclusion of the finished device into the socket. For money (with the OPA2227PA chip) it comes out about 500 rubles. Given the resulting result - quite good. The device copes well with its tasks, the profit goal is achieved.

As a result, a link to a good English language resource:
tangentsoft.net/audio/cmoy-tutorial
there is also testing various OU:
tangentsoft.net/audio/opamps.html

UPD. Yes, yes, I know that the board turned out badly (a stale solution + forgot to pull out on time), control the process and try not to pervert, the copper should be smooth, shiny and pleasing the eye.