Secrets of a lamp sound

It is necessary to build the lamp amplifier? Certainly, at least to learn that represents this well-known "a lamp sound". Who cannot construct itself, that buys in shop or orders the individual project. But all amplifiers sound differently. Efforts of thousands audiophiles ways of creation of the lamp amplifiers possessing a fine sound were outlined. They do not hide results of the experiments, issue magazines (for example, "the Bulletin А.Р.А.") where publish successful (and not so!) circuit decisions, resting on rare, or very expensive accessories and materials. Much less in these editions it is given attention to theory questions, "is more shown off". It is recommended to select each element of the amplifier and to listen, listen! And here, gone crazy from councils and listenings, the reader already runs on the market and looks for condensers on 100 dollars for piece or the transformer for 500, expecting to hear with their help well-known "a lamp sound".

Enterprising people started to make p on потребу thirsting various lamp amplifiers and WHALES (sets of details). The plants making electrovacuum devices, again let out pryamonakalny triodes (2С4С, 6С4С, 300В etc.) . Curious messages are printed: members "societies of mister Sakuma" (Japanese audiophiles) ignore amplifiers, if their cost less than $10000. To put it briefly, the opinion strongly affirmed that "a lamp sound" — it is good! And for big money — it is even better!

As are compared by amplifiers on sounding? Certainly, listening to music records: phonograph records, CD, recorder tapes. Thus it is necessary to switch constantly some cables that demands certain time. Considering brevity of musical memory, comparison turns out not so such reliable. It is much better to connect a source of signals to entrances of both amplifiers, and their exits to switch on the EXPERT by means of the powerful switch.

the Block diagram of such path of listening is provided by p on fig. 1 (for simplicity one channel is shown).

Here a source of information and loudspeakers — same for both amplifiers. By means of regulators of RP1 and RP2 identical loudness of sounding of speaker systems (EXPERT) is established at different positions of the SA1 switch. The indicator of level of PV1 can be absent, but it is better, if it is used. The circuit — simple and clear.

However if we will compare p amplifiers to different target resistance, mistakes in an assessment of amplifiers are inevitable. In what here business? And the matter is that the EXPERT, as a rule, have chastotnozavisimy internal resistance of Z. On fig. 2 approximate dependence of Z on frequency for two-way the EXPERT is shown. Fazoinvertor on low frequencies has two peaks instead of one, but it is the main point does not change. If the EXPERT — three-lane, "humps" on ha-rakteristeke Z (f) can be more. R _E — resistance of a loudspeaker on a direct current, it approximately to equally "nominal" resistance the EXPERT, i.e. number Z = (1,2... 1,3) R _E . Are used the EXPERT with nominal resistance of 4 or 8 Ohm more often. Audiophiles love loudspeakers for cinema with nominal resistance of 12 and 16 Ohm for their high return. Humps on the characteristic of Z=Z (f) can surpass in 2 and more times number Z .

is abundantly clear that at different target resistance of R _{amplifiers vy} and identical EDS on their exits, tension on the EXPERT will be a miscellaneous as R _vy and Z form a tension divider. If target resistance of amplifiers are not identical, and they after all can be and frequency-dependent, the EXPERT will sound differently. Especially it is appreciable at comparison of lamp amplifiers without feedback [1] and transistor, having, as a rule, deep negative feedback. In the first case of R _vy = 2... 3 Ohm, in the second — R _vy = 0,1... 0,01 Ohm.

the Lamp amplifier will emphasize p those frequencies, on which Z increases. And it is valid, LF and VCh at it sound "better". If frequency of the section LF and VCh () in the EXPERT falls f _{paºñ on area of 3 kHz, and on this frequency there is "hump", string instruments and voices of soloists better sound. The conclusion arises that the frequency characteristic of internal resistance the EXPERT should have as it is possible smaller nonlinearity, (in an ideal — a horizontal straight line) that it was possible to compare two different amplifiers.}

is artificial having increased R _vy for the amplifier with small internal resistance, having turned on the consecutive R _{resistor д} (fig. 3), we will receive identical working conditions the EXPERT.

These reasons were checked by div in practice and completely proved to be true. Two stereophonic amplifiers were compared. The first — lamp, single-cycle, on lamps 6Н23П and 2С4С, according to Loftin-White circuit without OS. Its key parameters: R _vy ~ 3 Ohm, R _vy ~ 3 W? f= 12... 40000 Hz. Target transformers of the amplifier are executed on cores from steel of type 3409, cm2 S=15? = 0,35 mm, l 3 = 0,3 mm. The second — transistor, with OOS, R _vy ~ 0,01 Ohm, R _vy = 50 W? f= 5... 150000 Hz.

Need to be told that this lamp odnotaktnik on 2AZ lamp (2С4С) is considered almost "exemplary" UMZCh among audiophiles. However, they stipulate also additional conditions (special wires, special solder, etc.). Its sound is really good: sharp front (attack), big transparency." Through it" string and percussion instruments perfectly sound.

the Transistor amplifier was constructed by p according to the reasons stated by the author in [2]. Time of establishment of its transitional characteristic to an error of 0,01 % does not exceed 10 microsec (on active resistance of loading).

In experiments were used by p three-lane the EXPERT with passport capacity of 70 W. Fazoinvertor is adjusted on frequency of 25 Hz, the frequency characteristic of Z is provided in the table:

Comparison of amplifiers was carried out at R _{vy = 3 W. Tension AChH on EXPERTS plugs at Rвых = 2... 3 Ohm gets lifting (to 3 dB) on LF and VCh, according to Z growth. Without R д the transistor amplifier sounds more "chilly" but as soon as R д = 2,2 Ohm joins, its sounding by anything (I emphasize — anything!) does not differ from sounding of lamp Loftin-White. Offer wishing to be convinced in it.}

Having talked about entrance resistance the EXPERT, we will pass to target resistance of the amplifier. As it was already noted, it makes a great impact on quality of sounding. Therefore we will look, as it to measure. There are some ways, but we will stop on what is defined in state standard specification 23849-87 [3]. This method is based on a transmission of a sinusoidal current via target plugs of the amplifier and power failure measurement on its target resistance of Zi (fig. 4). The direction of a current of I in drawing is shown conditionally (from the generator in loading). This circuit is not intended for measurement of negative Zi. Here R1 — the active resistance equal to nominal resistance of loading for this UMZCh. It should be sufficient capacity as through it the decent current (only in 3 times smaller maximum flows). Power failure on it, measured by the PV2 voltmeter, should be on 10 dB (in 3,16 times) less nominal target tension of the amplifier. The ZCh generator too should be rather powerful (for example, G3-109).

as the amplifier for creation of a necessary current can use the second channel of the stereoamplifier or any other UMZCh of sufficient capacity. If the tested amplifier has, for example, R _{of number} = 50 W, number Z = 4 Ohm, the current of I = 1,1 is required And. Target resistance
/>