High-tech casing made from electrical tape. Initially, I made the board under a heat-shrinkable tube - but literally a millimeter was not enough, it didn’t fit. Well, nevertheless, I like it.

Price issue

A piece of one-sided PCB: 2 rubles
MAX9724 - 7.78 rubles
4 resistors - 0.07*4 = 0.28 rubles
Capacitors - 0 (even if you buy, ~30 rubles max.)
Connectors - 0 (if you buy, ~20-30 rubles)
Insulating tape for high-tech housing - 1 ruble

Total - this is exactly 11.06 rubles for me, and about 61.06 rubles if you buy everything :-)

Results

Of course, I immediately came across a known problem: when working with audio, you cannot connect to the same ground in two places (USB ground and audio jack ground). In this case, interference creeps across the ground, which cannot be filtered out, and no power stabilizer will help here. (the problem is that USB has its own ground level, sound has its own, and our board has its own. Depending on the current consumed, the ground rises differently everywhere and this causes irremovable interference).

You can solve this problem either by getting rid of the audio connection (USB DAC) or the power supply (battery or other power supply). I was completely satisfied with using a power supply with a USB output due to the fact that they are available everywhere and are standard.

The end result is beyond any expectations. No complaints about quality, absolutely 0 noise, comfortable volume level - from 22 to 40%, and a reserve for “pulling out” quiet recordings. The sound is richer (the main thing to remember is that the bass here starts from 0Hz) and all that, and in general - audio devices made by yourself always sound especially good :-)

It differs from ready-made Chinese devices (like the FiiO E3) by more low price(sic!), assembly with spare components, absence of capacitors in the audio path, greater power when working with high-impedance headphones (300 Ohms) due to more high voltage power supply and the sound quality in theory promises to be higher (in practice I probably wouldn’t hear the difference).

PS. As I mentioned above, an amplifier is needed not to spoil your hearing with ultra-high volume (not to mention torn headphones), but to drive “heavy” headphones with low sensitivity if the output of the sound card is too dead. Well, extract quiet recordings/movies without software...

PS2. The difference between pluses and “added to favorites” is 4 times, a record :-)

Introduction

I’ll frankly say right away that this was my first audio amplifier, and this is my first such article, and if any of the more experienced and knowledgeable datagorians see the weak points of this project, please let me know about them, I will be very grateful!
It all started with the fact that New Year I decided to give myself a small gift, namely headphones from a well-known German company. Since I’ve been listening to music all my life either through inexpensive Chinese multimedia speakers or in a Korean car, the new acquisition seemed like a fairy tale to me! I listened to music on my new headphones all evening long. Further - more, if “ears” for 50 bucks produce such sound quality that if I buy something more serious, I’ll be on fire!?
After surfing the Internet, I found out that “serious” headphones have a resistance of more than 32 Ohms (which I considered the standard for all headphones), along the way I found out that for such instances it is better to get a special telephone UMZCH in order to unlock their potential. But buying an amplifier was not part of my plans. I’ll do it myself, I thought, since my profession is directly related to electronics.

Rice. 4 Customized amplifier circuit

I also corrected my signet, here is the final version - Fig. 5. I placed all the transistors under one heatsink (they still don’t get very hot), freeing up space for my modifications.

Fig.5 Final version amplifier circuit board

Two black tracks are on the back of the board (I cut them out with a knife after etching and drilling). The board turned out to be two-layer; otherwise it would not be wired normally; the size is 90x110 mm.

Fig.6 Power supply diagram

Several questions immediately arose:
- why are there no film or ceramic capacitors in the rectifier?
- is there any real benefit from capacitors in parallel with bridge diodes?
- for what purpose were these particular resistor values ​​chosen in the stabilizer kit?
I'll collect it and see, I thought. Well, in general - not a fountain. As I thought, without film capacitors in the rectifier, capacitors in parallel with the diodes in theory reduce the level of RF interference, but without them it’s not bad, I didn’t notice any difference either by ear or by instruments. But I liked the work of the stabilizer switching circuit, I should take note. When connecting the amplifier, a 100 Hz background was heard. I won’t spoil my experiments, the final circuit of the power supply, with my adjustments (Fig. 7)

Fragment excluded. Our magazine exists on donations from readers. The full version of this article is available only

Fig. 7 Power supply diagram with my adjustments

Now I am satisfied with the power supply, the background noise from the headphones is gone. At maximum load (1A on both arms), the voltage at the output of the stabilizers drops by 10 mV.

My signet in Fig. 8

Fragment excluded. Our magazine exists on donations from readers. The full version of this article is available only

Rice. 8 Power supply signet

A ready-to-use power supply installed on the chassis is shown in Fig. 9.

Rice. 9 Amplifier power supply

A little about the design. The mole rat BP is made of double-sided 3 mm fiberglass, because after etching the tracks there was copper foil left on the back side, I decided not to tear it off (a nasty task), there will be additional shielding. One radiator for two stabilizers, again from the old one motherboard. On the right side of the board there is a connector for connecting a blue LED (for LEDs of a different color, you need to reduce the value of resistor R1, see Fig. 7). The output voltages are output through wires soldered directly into the board (blue harness in Fig. 9). The transformer is screwed to the board with an M6 pin. Board size 90x200 mm.

Frame

As always, the most time-consuming part of the project is the body. The case is completely dismountable (my specific requirement, those who work in a sensitive enterprise will understand: blush :) is made of 2.5 mm aluminum and 3 mm fiberglass. M5 brass nuts are soldered to the rectangular fiberglass plate that serves as the chassis for the device. The power supply board is screwed to the chassis with 4 hexagonal posts and 2 screws (see Fig. 9). The amplifier board is screwed to the racks, the connector from the power supply is connected (Fig. 10.

Fig. 10 Amplifier assembly.

The front and back panels are made of sheet aluminum, bent on a machine. The top cover is made of fiberglass. Side panels made of aluminum are screwed at the end of the assembly to the corners on the chassis and on the top cover; they form peculiar legs.
The assembled amplifier (Fig. 11) is completely shielded; screw connections ensure reliable electrical contact between the housing parts. Rear panel contains a standard 3-pin plug from a computer power supply, and a switch from there.

Fig. 11 Assembled device

I will briefly describe the important aspects of the assembly.

1. All wires are twisted, and those that go from the power supply to the amplifier are shielded (just in case).
2. IMPORTANT! The power supply housing is connected to the chassis at one point, where the transformer is screwed on (that’s why the pin securing the transformer is brass)
3. IMPORTANT! The nuts of the headphone connectors on the front panel (the signal housing hangs on them) are insulated from the front panel with dielectric washers.
4. The potentiometer shaft is electrically connected to the body of the device; if this is not done, when touching the potentiometer shaft there will be interference in the headphones.

The design turned out to be restrained. The body is painted with matte black spray paint (2 pieces were used for 3 layers). The handle on the potentiometer shaft is made from a perfume cap. All screws and ends of the side panels are polished to a mirror finish.

Probably many of you have encountered such a problem when, having connected your headphones to an MP3 player or phone, the volume was insufficient, in other words, the power of the player or phone was not enough to provide loud, clear sound. And what to do in this case?

To do this, you can assemble a headphone amplifier with your own hands. Its scheme is quite simple and any radio amateur, no matter whether beginner or experienced, can do it, showing accuracy and attentiveness.

When creating this amplifier, I wanted to make it unusual, I wanted to move away from the classic plastic case. Remembering that fans of computer modding often make transparent cases for their PCs, I also decided to make the case of my amplifier transparent. And as a highlight - to abandon the printed circuit board and make everything surface-mounted.

The development of the scheme was carried out in the program Eagle. This is a classic dual opamp amplifier. OPA2107.

Below is a DIY headphone amplifier circuit:

List of required parts for the amplifier power supply:

• Power connector;
• LED 5 mm (any color);
• R1LED - resistor rated from 1K to 10K (1 W);
• CP1, CP2 - electrolytes 470 μF (for voltage 35 or 50 Volts);
• RP1, RP2 - 4.7K (1 W);

Amplifier parts list:

• IC1 - dual operational amplifier OPA2107;
  (note - on schematic diagram The operational amplifier is designated as OPA2132, the fact is that at first I planned to use it);
• C1L, C1R - 0.68 uF 63 V (for audio input signal);
• C2, C3 - 0.1 µF (film, to stabilize the operational amplifier);
• R2L, R2R - 100K (0.5 W);
• R3L, R3R - 1K (0.5 W);
• R4L, R4R - 10K (0.5 W);
• R5L, R5R - jumper (optional);
• Stereo jack - 2 pcs;

Since I decided to make everything hinged, I started making the frame. Here you will need accuracy and attentiveness, because... the case will be transparent and any defects will be immediately visible.

For the power bus, I used single-core copper wire, 1 mm thick, taken from cable scraps that were used for home wiring.

Any transformer power supply with a voltage of 12 Volts and an output current of 300 mA or more is ideal as a power supply. It is advisable to use a transformer power supply, since the use of pulsed ones can lead to interference (a constant hum will be heard in the headphones).

For the power connector I used this connector: (the central contact is the power plus).

In order to form identical terminals of resistors and wires, I used a regular screwdriver. You can use different diameters for larger or smaller radii.

A little lower you can see the power supply wiring. The input to the power supply is 12 Volts, which are then converted into +6 Volts and −6 Volts using a voltage divider (resistors RP1 and RP2, 4.7 kOhm each). The fact is that the operational amplifier requires bipolar power supply. The wire in the center is the so-called “virtual ground”, which under no circumstances should be connected to the real ground (at the power connector).

Two large 470 µF 50 Volt capacitors paired with 0.1 µF capacitors are necessary in order to reduce interference to the op-amp and increase the stability of its operation. To do this, you need to try to place them as close as possible to the op-amp terminals.

Here are a few more photos from different angles that show how I did the editing.

After you have finished soldering, you can begin checking the amplifier. A little advice, you don’t need to use your coolest headphones to check, some simple ones will suffice. The fact is that if you get confused somewhere and solder the parts not according to the diagram, then it is quite possible that you will ruin your headphones. But I hope that when you check everything will be fine.

Since the amplifier will later be filled with epoxy resin, I decided to raise it a little so that when pouring it would be exactly in the center of the body. To do this, I soldered small pins from the bottom.

I thought it would be nice to refine the design of the amplifier a little more and so I decided to print stickers for the audio connectors. I prepared them in Adobe Photoshop, then printed it on thin photo paper and glued it to the connectors with double-sided tape.

For some time I have been thinking about the design of the body and the material from which the mold will be made for pouring. I chose 1.5 mm plastic; it cuts perfectly with an ordinary stationery knife, leaving a very smooth edge.

Then I designed the fill form using the same Eagle. Having cut out all the parts, I started assembling. In order to facilitate this procedure, I first grabbed all the corners with superglue, then taped each seam 2 times, which ensured complete tightness.

The easiest way to find out the volume of epoxy resin to pour is to fill the mold with water, then pour the contents into a cup and find out the resulting volume and weight. Of course, you can measure the volume using a ruler - but the method with water seemed easier to me.

I used clear epoxy resin to fill it. For this particular resin, the ratio of hardener to resin should be 1:50. It was quite difficult to measure such a small amount of hardener; jewelry scales were useful for this. In general, for different brands of epoxy resins the ratio of hardener to resin varies, see the instructions.

The mixed resin must be poured slowly down the side of the mold to avoid bubbles. The picture below shows that when pouring the resin, I poured a little more than required, but the resin did not spill out due to surface tension. This is necessary because epoxy resin shrinks slightly in size as it hardens.

When the epoxy resin hardens, a lot of heat is released (in my case the temperature was 62 degrees). The mold is then covered to prevent dust and debris from reaching the surface.

I left the epoxy resin to cure for a day. After this time it dried and I started removing the mold. For this I used a belt sander.

Then, using a router, I ground off the chamfers and all the sharp corners.

To polish the body, I first used 600-grit sandpaper, and did the final wet polishing with 1200-grit fine sandpaper.

And finally, here are a few more photos of the finished do-it-yourself headphone amplifier:

Now you know how to make a headphone amplifier with your own hands.

Dissatisfaction with the quality of music playback sound card computer forced me to start making a desktop amplifier. I decided it would be simple. homemade amplifier for headphones, assembled according to the classical scheme on one .

However, there is a note. This amplifier will be suitable only when the input signal does not require voltage amplification (for example, an MP3 player or computer provides a sufficient output). Also, any noise generated in the power supply will go straight through the amplifier. For this reason, it is necessary to use only a stabilized power supply. Output voltage range 10-20V and current 750mA. Here we use an N-channel MOS transistor with a reverse diode for operation in switching and linear modes IRF610. During the manufacturing process of the amplifier, the use of other transistors was tested: IRF510, IRF611, IRF612 and IRF710, all without exception worked well. I recommend not using IRF530 and IRF540 (commonly found in power supplies). The used LM317 stabilizer with adjustable output voltage allows you to very accurately adjust the output parameters of the power supply.

Since this amplifier will be sitting on a desk in a production office, it needs to fit in with the work environment. I was lucky that there was a failed external CD-ROM, its design was perfect. In addition, its case already had a switch, power adapter, RCA socket and inputs on the rear panel, as well as a headphone jack on the front panel.

In the manufacture of the amplifier, only those electronic components and components that were available were used. Conventional resistors and film capacitors. Capacitors with a capacity of 1 µF, 0.47 µF and 0.1 µF are polypropylene. But no one is stopping you from using higher quality parts.

Cooling radiators have a relatively small volume of cooling area, but I draw your attention to the fact that they are screwed directly to the metal body, which also takes part in heat dissipation. The smaller radiator has a volume of approximately 1.75 square inches. Be sure to isolate the MOSFET and regulator from the heatsinks.

The operation of the amplifier was tested using an regulated power supply; it was turned on at low voltage. The bias is set using a 100 kOhm variable resistor. The amplifier showed good job throughout the entire voltage range from 10 to 20 V, but still high-quality sound reproduction began at a supply voltage of more than 13 volts.

Next, the operation of the amplifier was checked using a USB oscilloscope. This is a DSO-2150 with 60 MHz bandwidth and maximum frequency sampling rate 150 μ/s. The sine wave we saw showed itself at its best from 20 Hz to 20 kHz.

Square wave 100 Hz

Square 4800 Hz

Green is the input signal, and yellow is the output. The signal power of my generator is not high and this is reflected in the quality of the original waves. If you compare the input voltage and output voltage you will see that the circuit gain is about 0.8. It can be seen that at 100 Hz there is a slight slope. The slope gradually decreases and the frequency increases and beyond about 300 Hz the squared wave response is excellent up to 20 kHz - the limit of the oscillator signal. Since the music consists mainly of sine waves this is not a problem. Since an MP3 player or computer will be used to adjust the volume, there is no need for a potentiometer. Another ULF, but using lamps, is possible.

If you are the lucky owner tube amplifier , then most likely, if you want to listen to your favorite songs alone, through headphones, you are faced with the inconvenience caused by the lack of output to headphones.

And owners of expensive or not very expensive smartphones and tablets also have a hard time - these devices are most often not able to pump high-quality high-impedance headphones. Therefore, your favorite compositions sound completely different from how they sound on professional equipment.

Of course, if you are a true music lover and music is more valuable to you than money, then nothing will stop you from buying preamp for $6000, a headphone amplifier for $5000 and the headphones themselves for $2000. And plunge into nirvana... However, if the money situation is not so rosy, or you like to do everything yourself, then it turns out that you can build a high-quality headphone amplifier for only... $30.

Why do you need it???

Do you need a precision amplifier? It depends on your musical preferences and habits. If you are used to listening to music “on the run”, that is, from portable devices while walking, jogging, in the gym and other similar places, then the project described below is not for you. Just try to choose headphones that match your device with the most suitable characteristics and sound.

You should do the same if you like musical styles where there is strong signal distortion, such as rock, heavy metal and the like.

However, if you prefer to listen to music in a quiet, comfortable environment at home or in the office, and your tastes gravitate toward live and natural music such as classical, jazz, or clean vocals, then you will appreciate the sound quality and accuracy of the mix. precision amplifier plus high quality headphones.

Options

Let's say you decide that you need a headphone amplifier. What's the next step? On the Internet you can find a lot of projects using the ubiquitous LM386. The microcircuit has become popular due to its high reliability, low cost, ability to work with single-supply power and a small number of external elements. Such amplifiers usually work well with inexpensive headphones, but all these advantages pale when compared to the noise and distortion levels of the LM386 and a well-designed discrete or ASIC amplifier.

If you have about $30 and are not afraid of working with surface mount elements (SMD elements), then the project presented here is exactly what you need.

Ideas and scheme

When designing this scheme, the following points were taken into account:

• The amplifier must be driven by the relatively high impedance output of a tube preamp or electric guitar amplifier. In other words, the input impedance must be easily tunable for sources with different output impedances.
• small number of components. Therefore, microcircuits were chosen instead of transistors.
• low gain and power. Needs to be rocked sensitive dynamic headphones, not the speaker system.
• The amplifier must be able to handle high impedance headphones. The author uses Sennheiser HD 600 (resistance 300 Ohms).
• get the lowest possible noise and distortion.

Schematic diagram precision headphone amplifier shown in the figure:

Click to enlarge

When developing this design, microcircuits from such manufacturers as National Semiconductor, Texas Instruments and others were considered. Weight useful information was found on Headwize resources and DiyAudio forums.

As a result, the choice fell on a precision headphone driver from Texas Instruments TPA6120A2 and operational amplifiers AD8610 from Analog Devices for the input buffer.

The circuit turned out to be relatively simple, with bipolar power supply. If you are sure that there is no DC component at the output of your signal source, then the coupling capacitors (C24 and C30) can be excluded from the path using jumpers H1 and H2.

The power supply provides ±12V output voltage at a load of up to 1A. Its diagram is shown in the figure:

Click to enlarge

Often in audiophile designs, the cost of the power supply is several times higher than the cost of the amplification part itself. Here it turned out a little better - the cost of the elements for the power supply is approximately $50 and the most expensive elements here are the transformer and electrolytic capacitors. You can save a little if you replace the toroidal transformer with a regular W-shaped one, abandon the LEDs and fuses at the output of the unit.

We tested a version with separate stabilizers for each TPA6120A2 channel (the microcircuit has separate power pins for each channel). The difference could not be heard or measured, which made it possible to significantly simplify the power supply.

Since all the microcircuits used in the amplifier have low sensitivity to noise and interference along the power supply circuits, as well as a high level of suppression common mode interference, then the use of standard integrated stabilizers in the power supply turned out to be sufficient to obtain high performance.

TPA6120A2

The Texas Instruments TPA6120A2 is a high-quality, high-fidelity headphone amplifier. It uses an amplifier architecture with differential input, single-ended output, and current feedback. It is thanks largely to the latter that low distortion and noise, a wide frequency band, and high performance are obtained.

The microcircuit contains two independent channels with separate power pins. Each channel has characteristics:

• output power 80 mW into a 600 Ohm load with ± 12 V power supply at distortion + noise level 0,00014%
• dynamic range over 120 dB
• signal/noise level 120 dB
• Supply voltage range: ±5V to ±15V
• output voltage slew rate 1300V/µs
• protection against short circuit and overheating

For comparison, the distortion + noise level of the “folk” LM386 microcircuit is 0.2%. Although, of course, high parameters do not guarantee high-quality sound. To obtain the maximum result, you must take into account the manufacturer’s recommendations on the selection of external elements and PCB topology. All this can be found in the technical documentation for this chip.

AD8610

The AD8610 from Analog Devices is an operational amplifier with field effect transistors at the input, which gives low voltage offset and drift, low noise, low input currents. In terms of noise level and slew rate of output voltage, these operational amplifiers are in perfect harmony with the TPA6120A2.

However, don’t be lazy and try replacing them with other op-amps. According to the pinout arrangement, the AD8610 is compatible with other audiophile microcircuits. Moreover, many music lovers claim that they hear a difference in the sound of the op-amp!

Passive Components

Not all resistors are the same! And if your budget allows, use metal film resistors in this design, which are somewhat more expensive, but have lower noise and higher stability. If you want to save money, metal film resistors should be installed at least in the input circuits (for the AD8610), where the sensitivity to noise is the highest.

It is better to install film capacitors on the signal path C23, C24, C29, C30. The manufacturer recommends ceramic capacitors for power supply circuits of microcircuits.

The main requirement for signal connectors is reliable contact. In his design, the author used a regular “jack” to connect headphones and gold-plated RCA connectors with Teflon insulation to connect the signal cable.

The circuit diagram shows a version of the amplifier for operation from a tube preamplifier, in which the volume is adjusted. If the design is intended to be made more flexible and universal, then, of course, it is advisable to provide its own volume control at the input. To achieve maximum quality and not to degrade the characteristics of the amplifier, a high-quality potentiometer should be used here.

The budget version can be products from Alpha or RadioShack costing about $3. For $40 you can purchase an audiophile-grade product from ALPS. The best solution will use a strip attenuator from DACT or GoldPoint. Their cost is approximately $170. By the way, on eBay you can find similar Chinese-made attenuators for only $30. The potentiometer rating can be in the range of 25-50 kOhm. The use of a step attenuator, in addition to the convenience of volume control, additionally guarantees identical adjustment in both stereo channels, which is especially important in a headphone amplifier.

Design

All structural elements (except for the power transformer) are placed on one printed circuit board. If you decide to use an external power supply or assemble it in a different way, about 70% of the PCB will remain free.

The layout of the elements is shown in the figure:

Click to enlarge

The figure shows a drawing of the printed circuit board from the parts side:

Click to enlarge

The figure shows a drawing of the bottom side of the printed circuit board:

Click to enlarge

Printed circuit board drawings in the popular SLayout format can be picked up

The main installation feature: on the case on the bottom side of the TPA6120A2 there is a contact pad of approximately 3x4mm. She must be soldered to the area on the printed circuit board under the chip, which serves as a heat sink.

Photo of the finished structure:

When you turn it on for the first time, you should remove the two fuses at the output of the power supply and make sure it is working. If the output voltages are normal, replace the fuses. The amplifier itself does not need adjustment.

The board can be placed in a case of suitable dimensions, preferably metal, to shield it from external interference.

Conclusion

Subjectively, the amplifier sounds on par with professional studio equipment. When compared to the LM386, this design showed a smoother, cleaner and more detailed sound.

The scheme turned out to be quite flexible and easily customizable to suit various needs. For example, the author himself assembled two copies of the amplifier. One according to the above diagram for operation in conjunction with a tube preamplifier. The second copy was designed to work with a smartphone and a guitar amplifier, so it was supplemented at the input with a high-frequency interference filter and a volume control. In addition, to increase the gain (the smartphone produced an insufficient signal level), the values ​​of resistors R6 and R14 were changed to 2 kOhm.

By changing the values ​​of these resistors, you can change the gain within a wide range.

A variant of the amplifier printed circuit board from our “Martian friends”, designed for installation of elements in “standard” packages (there are no DIP packages used in the design of microcircuits):

Animated demonstration of the board from all angles