|
Arabic
Bengali
Chinese
English
French
German
Hebrew
Hindi
Italian
Japanese
Korean
Malay
Polish
Portuguese
Spanish
Turkish
Ukrainian
Vietnamese|
ENCYCLOPEDIA OF RADIO ELECTRONICS AND ELECTRICAL ENGINEERING The truth about the circlotron. The whole truth, and nothing but the truth. Encyclopedia of radio electronics and electrical engineering
Encyclopedia of radio electronics and electrical engineering / Tube Power Amplifiers Not so long ago, on dvdworld.ru, a discussion broke out, not without the participation of the author, about transformerless amplifiers in general, and the circlotron in particular. The author was in the minority ... the majority asserted the collective point of view that ...
These are the statements that are on the verge of the third stage of statistics. Let's go through the points. To begin with, let's figure out what a circlotron is and what a transformerless amplifier is ... persons who do not compromise on principles may not read further.
The circlotron is based on a push-pull bridge power stage, in which the currents of the power sources are cross-circuited through the load. The resulting load current is equal to the difference between the currents of the two arms. This is what the budget Electro-Voice A20 circlotron of 1956 with an output power of 20 watts looks like (exactly like this) (output and pre-output stages). A similar design on domestic devices was published in Radio, N9, 1963.
Well, where is the transformerless cascade here, the opponent will ask? And who told him that the circlotron is necessarily transformerless? Well, it’s definitely not me, it’s the gentlemen who invented the opponents themselves, all the questions are for them ... As well as about the transistor topology.
The load can be directly an acoustic system (as in modern circlotrons Atma-Sphere, Tenor Audio). Maybe - autotransformer (it is used both in factory designs, and by many users of "purely transformerless" circlotrons). You can finally close the load through the anodes,
and make the circlotron itself single-cycle, like this:
Well? Doesn't it remind you of anything? Well, then we will only talk about the classical symmetric circlotron. With a load in the cathodes. We have already mentioned the date - 1956. Events developed like this (I warn fans of new chronologies - the dates are real!)
The patents of Wiggins and Koikka were immediately realized in industrial products under the brands Electro-Voice (USA) and Voima Radio (Finland). The story is told in more detail on circlotron.tripod.com, from where this information was obtained by the author. Fortunately, there are still people in the world who transmit information taken not from the ceiling, but from patent libraries ... Indeed, newfangled technology ... Why didn't the scheme spread all over the world at the time? In the original transformer-pentode version, its only advantage over traditional push-pulls is the low output impedance on the cathode side, which simplifies the design of the transformer. All other "advantages" of a pentode push-pull are obvious (mandatory environmental protection, Williamson cascades, at least two pairs of separating capacitors, etc.). And a significant disadvantage - a double set of windings, rectifiers and filters - did not allow competing in price with traditional designs. After all, then there was no handicraft, and the struggle was for every dollar, and not the number of zeros in the price. A quantum leap to a completely transformerless circuit required a transition to a qualitatively different price level, especially with the then components - let me remind you that the voltages in a transformerless amplifier are tube, and the currents are transistor, so the cost of a full-fledged power filter (10-40 thousand microfarads * 200V per channel ) and today is not at all childish ... In general, the child did not take root. The new life of the circlotron began around 1982 (Brezhnev died, the Boeing was shot down, Pershing was deployed, Novacron was released). By the way, about a double set of power supplies. It is almost inevitable in power amplifiers, but in Ralph Karsten's balanced preamplifier (US patent 6242977) - a full-fledged circlotron with a direct output (120V peak-peak, no joke!) To a 600-ohm line - it cost one set of rectifiers. How? not easy, but very simple ... whoever didn't guess, go to the patent library, it's not for me to teach you. This is also possible in a lamp terminal ... a pair of capacitances and a pair (better - two pairs) of MOS transistors on good radiators. Now let's deal with the stems. It is difficult to say why such botanical knowledge has settled in the minds of opponents (the developers of the "stems" preferred ethnographic terms from the life of the indigenous peoples of the United States). As the investigative experiment showed, the Futtermann-Rosenblit circuit is called the stem (practically, only the Rosenblit version is currently being produced - the original Futterman circuit turned out to be unreliable and did not adequately use the low output resistance from the cathode). Here it is, the stem, which has nothing to do with the circlotron.
The F-R circuit works confidently only with feedback (at least 12dB). Without OOS, it is inoperable - the output impedance from the cathode and anode sides is different, there will be a lot of second harmonic even by khaend standards. But only 3 preliminary cascades are needed, and one is enough in the circlotron. And, on top of that, the final stage in the F-R circuit sees completely different load capacities. In the circlotron, both arms are symmetrical, and there are no problems with different phase shifts. Kilohertz up to hundreds. For direct current - both in the circlotron and in the "stalk" - two independent sources of output stage bias are needed. Indeed, with a direct connection of acoustics, the difference in the currents of the shoulders closes through it. But in practice, with a maximum current of a shoulder of 0.5A (eight 6H13C or 4 6C33C per channel), even with a complete failure of one shoulder, exactly half an ampere will flow through the load. In life, even the most deserved opponents and radio destroyers will not be able to achieve an imbalance of serviceable shoulders of more than 1/3 of the quiescent current with serviceable lamps. Is it possible to kill acoustics with a direct current of 100-200 mA? In extreme cases, if one shoulder failed, and in the other - the nets sat on the ground, then excuse me - the fuses should work. Opponents, do you know what it is? And with an autotransformer connection, the question of a constant in the load is generally inappropriate. With a full winding resistance of 1 ohm from each cathode to ground - exactly half an ohm, and at the output terminal - a quarter of an ohm ... multiply by 0.5A, we get 125mV in the worst case. Now about OOS. Circlotron without NFB on traditional "stabilizing" lamps
The first - transformer - Electro-Voice circlotrons worked only with OOS. For the sake of economy, they used pentodes, and with cross-powered screening grids, they squeezed everything that was possible out of them. A modern circlotron removes the same 20W not from a pair of 6P6S, but from eight 6N13S. So the question of non-linear distortion, the notorious third harmonic is not at the first watt, and not even at the tenth ... And, by the way, what will happen at the tenth watt with a single-cycle at three hundred? This is not a scolding for the sake of it, it's just to represent the difference in scale. Now about class A and AB. Here both incurable opponents and even quite literate people get confused. Further - for the literate! Consider a real circlotron (Mammoth 1), 8 6H13S lamps per channel, load 8 ohms. Let's set the quiescent current to the triode - 75mA (total - 1.2A, the offset is about -60V). At what output power will the cascade go from class A to class B? We will restrict ourselves to a sinusoid at the input for simplicity of the example. Modeling in EWB 5.12 accurately reflects the essence of the process. Traditional logic says - with an instantaneous load current of 0.6A (effective load voltage 3.4V, power - 1.5 W), one arm will completely close. 6W is not enough. And now let's see how the currents of the shoulders actually behave (excitation 9.2V eff, output 3.4V eff):
Nothing is closed! After all, under the cathode - not the ground and not the cathode capacitor, but half the load! Have you forgotten the Law of Three Seconds? We increase excitement, we approach the cutoff.
Oops! now you can turn on the stopwatch. On the grids - 20V eff, on the load - 7.3 V eff, load power - 6.6W. This is approximately the border of classes A-AB. Now let's increase the load resistance to 16 ohms with the grid excitation unchanged. The current shape will return to class A (approximately as in the first graph), at the load - 10.7V eff, or almost the same 7.0 W. The A-AB border will shift to 13W at the output (14.4 V eff at the load). Yes, the circuit likes high load resistances, I warned you. Who doesn't love them.. And no problems with the transformer in the cutoff. The cut-off in life, by the way, is less sharp than on ideal models - the lamp closes not so willingly. And finally, what does it sound like? Opponents, tell me honestly - which circlotron, when and in what system did you listen? Mammoth - always ready for your service. Come, let's fight together... Links and thanks
Publication: klausmobile.narod.ru
Machine for thinning flowers in gardens
02.05.2024 Advanced Infrared Microscope
02.05.2024 Air trap for insects
01.05.2024
▪ The smallest twin-engine turbojet aircraft ▪ Heat reduces school performance ▪ Detection of ultraweak radio waves using a laser
▪ section of the site The most important scientific discoveries. Article selection ▪ article Look for a woman. Popular expression ▪ article The functional composition of Orion TVs. Directory ▪ article Glue for the repair of rubber products. Simple recipes and tips
Comments on the article: Vladimir The author is well done!
Home page | Library | Articles | Website map | Site Reviews
www.diagram.com.ua |
See other articles Section 
Latest news of science and technology, new electronics:
All languages of this page