ENCYCLOPEDIA OF RADIO ELECTRONICS AND ELECTRICAL ENGINEERING Secrets of lamp sound. Do I need to build a tube amplifier? Encyclopedia of radio electronics and electrical engineering Encyclopedia of radio electronics and electrical engineering / Tube Power Amplifiers Do I need to build a tube amplifier? Of course, at least in order to find out what this famous "tube sound" is. Those who cannot build it themselves buy it in a store or order an individual project. But all amps sound different. Through the efforts of thousands of audiophiles, ways have been outlined for building tube amplifiers with excellent sound. They do not hide the results of their experiments, they publish magazines (for example, Vestnik A.R.A.), where they publish successful (and not very!) Circuit solutions, focusing on rare or very expensive components and materials. Much less attention is paid to questions of theory in these publications, more "thrown dust in the eyes." It is recommended to select each element of the amplifier and listen, listen! And now, crazy from advice and listening, the reader is already running to the market and looking for capacitors for $ 100 apiece or a transformer for 500, hoping to hear the famous "tube sound" with their help. Entrepreneurial people began to produce a variety of tube amplifiers and KITs (sets of parts) for the needs of the thirsty. Factories producing electrovacuum devices again produce direct-heated triodes (2C4C, 6C4C, 300V, etc.). Curious reports are being printed: members of the "Society of Mr. Sakuma" (Japanese audiophiles) ignore amplifiers if they cost less than $10000. In short, the opinion is firmly established that "tube sound" is good! And for big money - even better! How do amplifiers compare in sound? Of course, listening to musical recordings: records, CDs, tapes. In this case, you have to constantly switch several cables, which requires a certain amount of time. Given the short duration of musical memory, the comparison is no longer so reliable. It is much better to connect the signal source to the inputs of both amplifiers, and switch their outputs to the speakers using a powerful switch. A block diagram of such a listening path is shown in fig. 1 (one channel shown for simplicity).
Here the information source and loudspeakers are the same for both amplifiers. Using the RP1 and RP2 controls, the same loudspeaker sound volume (AC) is set at different positions of the SA1 switch. The PV1 level indicator may be absent, but it is better if it is used. The scheme is simple and clear. However, if we compare amplifiers with different output impedances, errors in the evaluation of amplifiers are inevitable. What's the matter here? And the fact is that speakers, as a rule, have a frequency-dependent internal resistance Z. In fig. 2 shows an example of Z versus frequency for a two-way speaker. The phase inverter at low frequencies has two peaks instead of one, but this does not change the essence of the matter. If the AC is three-way, then there may be more "humps" on the Z(f) characteristic. RE - loudspeaker resistance at direct current, it is approximately equal to the "nominal" AC resistance, i.e. ZMr. = (1,2...1,3)RE. Most often, speakers with a nominal impedance of 4 or 8 ohms are used. Audiophiles love 12 and 16 ohm rated cinema speakers for their high output. The humps on the characteristic Z=Z(f) can exceed Z by 2 or more timesMr..
It is quite obvious that for different output impedances of the amplifiers RO and the same EMF at their outputs, the voltage across the AC will be different, since RO and Z form a voltage divider. If the output impedances of the amplifiers are not the same, and they can be frequency dependent, then the speakers will sound different. This is especially noticeable when comparing tube amplifiers without feedback [1] and transistor ones, which, as a rule, have deep negative feedback. In the first case RO \u2d 3 ... XNUMX Ohm, in the second - RO = 0,1...0,01 Ohm. The tube amplifier will emphasize those frequencies at which Z increases. And it is valid, LF and HF at it sound "better". If the crossover frequency of LF and HF (fsection) in the speakers falls on the 3 kHz region, and there is a “hump” at this frequency, then string instruments and the voices of soloists sound better. The conclusion suggests itself that the frequency response of the internal resistance of the speaker should have as little non-linearity as possible (ideally a horizontal straight line) so that two different amplifiers can be compared. By artificially increasing RO for an amplifier with low internal resistance by including a series resistor Rд (Fig. 3), we obtain the same operating conditions for the AU.
These considerations have been tested in practice and fully confirmed. Two stereo amplifiers were compared. The first one is a lamp, single-cycle, on 6N23P and 2S4S lamps, according to the Loftin-White scheme without OS. Its main parameters are: RO ~ 3 Ohm, RO ~ 3 W, ∆f = 12...40000 Hz. The output transformers of the amplifier are made on cores of steel type 3409, S = 15 cm2, δ = 0,35 mm, l3 = 0,3 mm. The second is transistor, with OOS, RO ~ 0,01 Ohm, RO = 50 W, ∆f = 5...150000 Hz. It must be said that this tube single-cycle on a 2AZ (2S4S) lamp is considered almost an "exemplary" UMZCH among audiophiles. True, they also stipulate additional conditions (special wires, special solder, etc.). Its sound is really good: a sharp front (attack), great transparency. "Through it" strings and percussion instruments sound great. The transistor amplifier was built in accordance with the considerations outlined by the author in [2]. The settling time of its transient response up to an error of 0,01% does not exceed 10 μs (on the active resistance of the load). In the experiments, three-way speakers with a nameplate power of 70 W were used. The phase inverter is tuned to a frequency of 25 Hz, the frequency response Z is shown in the table:
Comparison of amplifiers was carried out at PO = 3 W. The frequency response of the voltage at the AC terminals at Rout \u2d 3 ... 3 Ohm acquires a rise (up to XNUMX dB) at LF and HF, in accordance with the growth of Z. Without Rд the transistor amplifier sounds drier, but as soon as R is turned onд \u2,2d XNUMX Ohm, its sound is nothing (I emphasize - nothing!) Does not differ from the sound of a tube Loftin-White. I suggest that those who wish to verify this for themselves. Having talked about the input impedance of the speakers, let's move on to the output impedance of the amplifier. As already noted, it has a big impact on the sound quality. So let's see how to measure it. There are several ways, but we will focus on the one defined in GOST 23849-87 [3]. This method is based on passing a sinusoidal current through the output terminals of the amplifier and measuring the voltage drop across its output resistance Zi (Fig. 4). The direction of the current I in the figure is shown conditionally (from the generator to the load). This circuit is not designed to measure negative Zi. Here R1 is an active resistance equal to the nominal load resistance for a given UMZCH. It must be of sufficient power, since a decent current flows through it (only 3 times less than the maximum). The voltage drop across it, measured with a PV2 voltmeter, should be 10 dB (3,16 times) less than the nominal output voltage of the amplifier. The AF generator must also be powerful enough (for example, G3-109).
As an amplifier to create the necessary current, you can use the second channel of a stereo amplifier or any other UMZCH of sufficient power. If the amplifier under test has, for example, PMr. = 50 W, ZMr. \u4d 1,1 Ohms, then a current I \uXNUMXd XNUMX A is required. output impedance
The input of the amplifier can be short-circuited, but it is better to put a resistor instead of a jumper, the value of which is equal to the resistance of the signal source. Zi measurements are carried out at a frequency of 1 kHz. This scheme, for all its simplicity, allows you to reveal another secret of "tube sound". The voltmeter PV1 then needs to be replaced with a sensitive oscilloscope, and the frequency of the AF generator should be changed from 20 Hz to 100 kHz. For a single-ended tube amplifier operating in class A, we will see the voltage U1 as a pure sine wave over the entire operating frequency band. Amplifiers operating in class AB, especially in class B, and covered by feedback, can greatly distort the shape of the sinusoidal current flowing through Zi. This suggests that Zi is non-linear. This is true for the vast majority of transistor amplifiers. Moreover, at the lowest frequencies, the voltage U1 can be sinusoidal, and as the frequency increases, it becomes distorted, and at frequencies of 20 kHz or more, the distortion can be very large, up to frequency doubling. And if you measure the harmonic coefficient of such an amplifier using the usual method, it can be quite small, for example, only 0,01%. Publication: cxem.net See other articles Section Tube Power Amplifiers. Read and write useful comments on this article. Latest news of science and technology, new electronics: Machine for thinning flowers in gardens
02.05.2024 Advanced Infrared Microscope
02.05.2024 Air trap for insects
01.05.2024
Other interesting news: ▪ Two Families of Inexpensive High-Speed USB Controllers 18.03.2005/XNUMX/XNUMX ▪ Hybrid generator saves up to 93% energy News feed of science and technology, new electronics
Interesting materials of the Free Technical Library: ▪ site section Preamplifiers. Article selection ▪ article by Euler Leonard. Biography of a scientist ▪ headhunter article. Job description ▪ article 144 MHz swiss square antenna. Encyclopedia of radio electronics and electrical engineering
Leave your comment on this article: All languages of this page Home page | Library | Articles | Website map | Site Reviews www.diagram.com.ua |