Single Ended Class-A Power Amplifier using 6C45Pi

The valve itself, have a simple circuit enviable any transistor amplifier, however, not many projects that employ a single active component. This construction philosophy, I like to call it minimalism.

If one speaks of minimalism, I think, that this project has all the credentials to qualify as such.

This is an amplifier cu, a final integrated (Compact) which employs a single tube (triode) signed 6C45Pi. To do this, could be used several tubes, including 437A, E55L (pentode) and the fantastic and expensive EC8020.

Two lines must for this valve: The EC8020 has an exceptional linearity, high dissipation and trasconduttanza worthy of a power triode. All factors ideal for a good amplifier cu, were it not for the prohibitive price (on the ebay Offers easily exceed one thousand dollars!).

6C45Pi even if does not reach the performance of EC8020, performs the task with dignity. The cost is a couple of orders of magnitude lower and is currently produced by Electro Harmonix and is part of his series most valuable the Gold Pins.

Schematic Diagram

The circuit used setup is classic: single-ended. This choice leads to a lower power, an output transformer more expensive, but all will benefit, at least for my taste, the sound quality.

shematic-Single-Ended-Class-A-Power-Amplifier-using-6C45Pi

About nutrition, the first idea was to use, such as straightening, a valve, but with an excessive size of the Schottky diode and filtering capabilities have also obtained excellent results (Sonic).


PCB

Not many projects are implemented on printed circuit valve, but it is my strong belief that, 's addition to a more orderly construction, provides added value to the success of the project.

The layout is as follows:

PCB-Layout-Single-Ended-Class-A-Power-Amplifier-using-6C45Pi

A fundamental rule in the design of the board is to avoid any link mass, so you should use as much as possible in a star structure. Usually the star center coincides with the negative of the first capacitor filter.

Realization

The board:



I used the usual method of iron more HCl + H 2 O 2.

Four resistors are metal film for half a watt, those from 220ohm wire wound vitreous enamel coated 6W, the Panasonic FC Electrolytic low impedance. Alps potentiometer, then the diodes, as already mentioned, type Schottky 1000V 1.5 A.

The two TU are wrapped by the Company upon specific ATC Electronix: www.atcelectronix.com/ provides the following features:

Primary: 4000 ohms
Secondary: 6 ohms
Power: 10W
Grain-oriented core AC 0:35
Bandwidth: 15 Hz - 75 KHz



I like to use for my projects output transformers wound on specific, because I have the opportunity to obtain more than I seek., The kind Mr. E. Calabrese's ATC Electronix has been immediately available, we discussed the characteristics and came quietly meeting.

Mechanically, the TU are well made, sturdy, beautiful black ice caps, we see that there is the hand of a fan.

Some photos of finished:


The cabinet is birch plywood 10mm polished with wax. I like the effect that old yellow comes out. The photos do not make the idea of greatness, but it is really small, only 230 x 200 mm high 60mm (TU and TA excluded).



I suggest to the speaker terminals, derived from a larval bi-wiring connections, it was a very economical way to get the four terminals, buying in bulk would have cost me more.



The cable used for connections between RCA, potentiometer and card is excellent Belden.

Listening

A valve, one triode and yet it sounds. Thought of having to keep the volume potentiometer can to listen to a decent level, but it was not at all. The pot did not exceed half of the race than I shouted down ... but we are talking about a "paltry" Watts!. Certainly the BLH (FE108ESigma) have done their bit (90 dB / W efficiency).

I connected everything, turned on and my room was immediately filled with music. Main features: detail, depth and dynamics.

Everything I heard sounded really good, thanks not only 6C45Pi but also of TU ATC Electronix.

The Finaly I am deeply moved, usually happens at concerts and my master amplifier.

The Medes and the voices are spectacular. I listened to all possible kinds of vocal music. Each of them seem natural, male or female is.

From Jazz to Classical, dall'Acustica the Rock and anything there is in the middle, everything comes back perfect. This little amp is very transparent.

Having experienced the one-tube amplifier, I had the confirmation: I will be forever minimal! Few watts, or only a few components in the audio chain, single screw speakers, no cross-over, no feedback, just pure music!. An amp like this, a simple player, a pair of single screw is all that a person has (or should have) need. I am extremely satisfied, so much so that I could separate me from my SE with the 2A3 and not feel greatly missed.

Faults? Sure. The sensitivity is very high, indeed it is low for maximum power output must be driven by almost 5 Vpp. The dynamic is good, but only one watt is just for musicians. However, the little man is promoted with flying colors while remaining down to earth.

(Source: http://digilander.libero.it/giovannidefilippo/progetti/GDF-UNICO!/UNICO.htm)

Promitheus Hexfet Power Amplifier

Here you can find the whole article of the Elektor magazine of the new upgraded Hexfet amp. The new amplifier uses bigger output devices, IGBTs in T0-247 Package. It has more power and uses new technology transistors. These transistors are insulated gate bipolar transistors (IGBT). They have an input like a FET so the work with voltage and they handle current like a bipolar.
Look here for The IGBT or HexFet amp.

I used the same board for my own upgrade. I used the higher power supply voltage of the second edition for IGBTs but since I didn't like to use IGBTs (I am a great fan of MOSFETs) I went on and used bigger Mosfets instead in TO-247 P case. I used the IRFP250 and IRFP9250 from Intersil/Harris. They are very cheap and reliable. The IR are the same too. IGBTs are a lot more expensive. This amplifier is great with very good sound and a lot of power for all kinds of speakers.

Modification

The original design already allowed for a higher-output version, whence the duplicated holes for the output transistors on the printed-circuit board. At that time, advance information on the IGBTs was already available, but samples were not.

Although IGBTs are quite different from HEXFETs, the board for the original design can be used without any modification. In fact, the circuit has hardly changed. The most noticeable alteration is the replacement of the fuses in the source lines of the power FETs by emitter resistors for the IGBTs. The only other changes are in the value of two resistors in the compensating circuit of the input stage, of one in the quiescent-current circuit, and of one resistor and two capacitors in the protection circuit.

This means that anyone who has built the original HEXFET amplifier can quickly modify it to the upgraded version. One item needs to be replaced, however:
the mains transformer. After all, more power can not be obtained from the same supply voltage/current. The original transformer with 2×25 V secondaries must be replaced by one that provides 2×30 V at 3.75 A. This will result in a direct voltage of ±43 V.

Ceramic speaker driver amplifier design

Today's portable devices contributed to the smaller, thinner and higher efficiency of electronic devices needs. Now cell phones have become very slim shape, so that the traditional electric-powered loudspeaker has become a mobile phone manufacturers can create as many thin constraints. Ceramic or piezoelectric speakers is rapidly becoming a viable alternative to electric-powered loudspeaker device. These ceramic speakers (drivers) can be slim and compact package provides a very competitive sound pressure level (SPL), which most likely alternative to traditional electricity-powered loudspeaker voice coil.

Ceramic speaker driver amplifier circuit and drive the traditional electric-powered loudspeaker amplifier circuit compared to a different output drive requirements. The structural requirements for ceramic speaker amplifier is capable of driving large capacitive loads, can provide more current at high frequencies, while also able to maintain a high output voltage.

The characteristics of ceramic speakers

Ceramic speaker manufacturers use the technology is very similar to the manufacture of multi-layer ceramic capacitors used in technology. Compared with the electric-powered loudspeaker, the loudspeaker manufacturing technology enables manufacturers to more stringent control over the speaker tolerances. Speakers you want a unified and agreed upon the acoustic characteristics, the rigid structure of tolerance is very important.

As can be seen from the driver amplifiers, ceramic speaker impedance can be modeled as an RLC circuit, it comes with a large capacitor, this large capacitor is the main component of the model shown in Figure 1. For most audio frequencies, the ceramic speaker is manifested mainly capacitive. This capacitive characteristics of the speaker to express its resistance will decrease with increasing frequency. Figure 2 shows the similarity with 1uF ceramic capacitor speaker impedance and frequency characteristics.

Figure 1. Ceramic speaker models.

Figure 2. The speakers and 1uF ceramic capacitor impedance and frequency characteristics.

The resistance also has a resonance point. Speakers at the resonant frequency above the audible most efficient point. Decrease in the impedance at 1kHz speaker expressed the resonant frequency.

Relationship Between Sound Pressure, Frequency and Amplitude

The two ends of the ceramic speaker to exert an alternating voltage can make the internal speakers of piezoelectric thin film deformation and vibration, the displacement distance is proportional to the input signal. Vibration of the piezoelectric film to promote the surrounding air, resulting in the sound. Improve the speaker can increase the voltage on the piezoelectric element deflection amplitude, resulting in a greater sound pressure and thus can increase the volume.

Ceramic speaker manufacturers maximum voltage tends to be used to mark their speakers, usually around 15Vp-p. Ceramic devices, said the largest voltage reaches the maximum extension. If the applied voltage is higher than the rated voltage does not produce a higher sound pressure, but it will increase the output signal distortion, see Figure 3.

Figure 3. Frequency Characteritics of Sound Pressure Level.

By comparing the SPL and the frequency and impedance versus frequency curve can be seen in self-resonant frequency of piezoelectric speakers have a large SPL above the most efficient.

Amplifier Requirements.

Ceramic speaker manufacturers typically provides 14 to 15Vp-p of the point of maximum voltage to produce the greatest value of sound pressure. The question now becomes how to generate a single battery supply voltage of these voltages.

One way is to use switching regulators to raise the battery voltage is 5V. With a stable 5V voltage, the system designer can select a needed bridge-tied load (BTL) for single-supply amplifiers. Connected through the bridge can be automatically loads the voltage speaker to see double.

However, the BTL amplifier with 5V output swing theory can only give rise to 10Vp-p. The voltage is not enough to make ceramic speaker output value of the highest SPL. In order to generate higher sound pressure values need to be adjusted to a higher supply voltage.

Another method is to use the boost converter to the battery voltage rises to 5V or higher, but it also has its own problems, such as component size problem. Large peak inductor current will be restricted the size of the overall solution, because in the end of the inductor size must do in order for the physical size of large core is not saturated. Although the small size of the inductor-current market, there are, but the core's saturation current rating may not be sufficiently large to satisfy the high-frequency high-voltage-driven speaker when the load current demand required.

Because ceramic speakers at high frequencies with very low impedance, so when you drive the ceramic components to take into account high-current drive and to avoid the limit.

Chosen to drive the ceramic speaker amplifier must be large enough to drive current in order to avoid driving a large number of high-frequency component of signal-driven speaker into the current limiting mode.

Figure 4 is a Class G amplifier using the application circuit. Class G amplifiers are available multiple voltage rails: a high-voltage and a low voltage. Low-voltage rail used in the output small signal, when the output signal demands for higher voltage swing, the high-voltage rails were switched to the output stage circuit.

Figure 4. Schematic amplifier Class G

Thus, when the output signal is small, G amplifier efficiency than Class AB amplifiers high, due to a lower voltage rail. Because there is a higher voltage rail, G amplifier can still handle the peak transient signals.

Shown in the figure uses a MAX9788 on-chip charge pump to generate a negative voltage and the VDD opposite. The negative voltage rail only require a higher voltage output signal when added to the output circuit on the rail. Boost converter and method of using the traditional Class AB amplifiers, the device can be more efficient to drive ceramic speakers.

Loudspeaker manufacturers often recommend and ceramic speaker series with a fixed resistor (RL), as shown in Figure 4. The resistor in the signal that contains a large number of high-frequency components can limit the amplifier output current.

In some applications, if the restrictions sent to the speaker of the audio frequency response of bandwidth to ensure that the speaker is not short-circuit of the amplifier, you can not have this fixed resistor. Ceramic speakers currently on the market value of the 1uF capacitor of magnitude or so. Speaker as the impedance at 8kHz when 20W, at 16kHz when 10W. Future ceramic speakers may have a larger capacitance value, will force the signal amplifier at the same frequency to provide more current.

Ceramic speaker and the efficiency of electricity-powered loudspeaker.

The efficiency of traditional electricity-powered loudspeaker is easy to calculate. In electrical windings on the coil can be modeled as a fixed resistor in series a large inductor.

You can use the speakers resistance value calculated in accordance with Ohm's law to provide power to the load:

P = I2R, or P = VxI

The power consumption of the speaker coil heat.

As the ceramic speaker's capacitive characteristics, they are power consumption when not produce too much heat. According to the dissipation coefficient of ceramic components, such speakers consumption of so-called reactive power (blind power) is very small. Therefore, the consumption of reactive power generated when the heat is very small.

Can not use a simple calculation of P = VxI reactive power

Reactive power should it be calculated as follows:

P = (πfCV2) × (cosΦ + DF)

Where: c = capacitance value of the speaker; v = RMS driving voltage; f = the frequency of driving voltage; cos j = "current through the speakers, and speakers on the phase angle between the voltage"; Df = speaker dissipation factor, depends on the signal frequency and ceramic speakers ESR.

For an ideal capacitor, the voltage and current between the phase angle should be 90 degrees, while the ceramic speaker is mainly capacitive characteristics, so cos j is equal to 0, that is not part of the ceramic speaker's capacitive power. However, the non-ideal characteristics of ceramic materials will lead to lag behind the voltage on the speaker through the speaker of the current phase angle between them is not equal to 90 degrees. The ideal 90-degree phase shift and the actual difference between the phase shift is the dissipation coefficient. Ceramic speakers in the Df can be modeled as a small resistance, ESR, and an ideal capacitor in series. (Do not series resistance and the isolation between the amplifier and speakers to confuse resistance) Df is the target frequency, the ratio of ESR to the capacitive reactance (Reference 2 and 3):

DF = RESR / XC

For example, a 1.6uF capacitor and the speaker 1W ESR being 5Vrms, 5kHz signal when driving reactive power is:

P = (π × 5000 × 1.6e-6 × 52) × (0 + 0.05) = 31.4mW or 31.4mW.

Real Power

Thus, while the ceramic speaker itself does not like the electric-powered loudspeaker, as the actual power dissipation by thermal methods, but the driver amplifier output stage, and is located between the amplifier and speaker external resistor (RL) could give rise to heat (Figure 4).

An external resistor the greater the amplifier will generate more power, thus affecting the low-frequency response.

In the drive with a 10W series resistance of the ceramic speaker, we can see that the reactive power has little effect on total load power. Most of the power consumed by an external resistor, see Figure 5 shows the amplifier power and frequency curve.

Figure 5: the necessary relationship between power and frequency

Better low-frequency response for a smaller external resistor, but this will cause the amplifier output stage power consumption increase. Amplifier efficiency shows how much power consumption in the amplifier output stage. Driven right amplifier power, including class and G class D amplifier, including the demand for more efficient solution, due to the load formed by a number of series resistance, it will load on the network rather than the speaker to generate a certain power. Even if the efficiency of 100% of the amplifier, series resistance will be consumed is supposed to give the speaker power.

In this simple example, 5kHz provided to the load point of the total power is 515mW. Efficiency of 53% of the amplifier will consume 457mW of power. The necessary power amplifier the size of the device can determine the size of the package used. If you must use high-frequency sine-wave-driven speaker, it requires a great deal of power.

In short, the increasingly thin portable devices contributed to the small size ceramic speaker needs. This speaker is different from the traditional electricity-powered loudspeaker, so designers need to consider different design elements. Ceramic speaker's capacitive properties require amplifiers with high output voltage drive and large output current capability, so that the entire frequency range in order to maintain a high voltage.

Chosen to drive the ceramic speaker amplifier must be able to mix the load while providing reactive power and real power. Amplifier efficiency must be high enough to ensure small size and low cost programs.

Therefore need to use different from the traditional Class AB amplifiers amplifier topology. For example, Class G and Class D amplifiers and other more efficient solutions are becoming more attractive, in which G amplifier can provide the best efficiency.

Translate from http://www.eeworld.com.cn/mndz/2008/0625/article_52.html

LM3886 Power Sub Woofer Amplifier

I found this schematic, LM3886 Power Sub Woofer Amplifier. Maybe next time, I'll share the results this amplifier.

Dual-band CDMA-EVDO and WCDMA-HSDPA Power Amplifiers

ABIResearch According to a recent report (1), wireless mobile data for the European contribution to a large number of operators in 25-30% of their income for Japan's contribution of up to 38% of operators revenue. This wireless data services, such as music downloads, e-mail and the increasing popularity of mobile Internet users and mobile broadband data services revenues to compensate for declining voice usage per user. Therefore, operators around the world are turning their 2G voice-oriented technology, upgrade their networks to 3G and 4G technologies, to enable them to provide high-speed, broadband wireless data services to consumers and business users.

CDMA / EVDO and WCDMA / HSPA is the most widely deployed 3G technology, the two. The two techniques depends on the Code Division Multiple Access (CDMA) digital wireless technology, so that many users can be effectively used with a radio-frequency spectrum of the voice and data communications. CDMA is a species rely on a unique "cut code" assigned to each user spread spectrum radio communications, in order to distinguish in the same RF spectrum of the other users. The spectrum to expand the channel capacity increases and allows multiple users to use the full bandwidth of operators. CDMA / EVDO network operators 1.25MHz, and WCDMA / HSPA network to use a wider range of 5MHz. Data transfer rate can also be in wireless networks to further increase high-level modulation to provide a better user experience and wireless data.Multi-band CDMA and WCDMA mobile phones the need for greater RF design work in order to provide sufficient output power and to ensure the linearity and efficiency of the normal operation of each band. Design usually includes a power amplifier and favorable duplex of each frequency band, RF filters and switches to meet the multiple RF input to a single antenna. In order to reduce the multi-band mobile phone design complexity, ANADIGICS硑made a dual-band CDMA and WCDMA power amplifier module. Integrated in a single package through the two power amplifier modules, and before the use of two separate DC and RF power amplifier design of signal lines in comparison, which can effectively reduce the circuit board area. This article will describe the dual-band power amplifier module in the CDMA / EVDO and WCDMA / HSPA mobile phone designed for typical application.

This article will focus on ANADIGICS AWT6221 in UMTS paragraphs 2 and 5-band, dual-band WCDMA / HSPA mobile phone design applications. The AWT6221 has won ANANDIGICS by the Chinese Ministry of Information Industry issued by the application of information technology, communications technology Outstanding Innovation Achievement Award (2) and EDN Innovation Award, communications and network product excellence award.

AWT6221 dual-band power amplifier significantly reduces power consumption up to 75% of the average current, and increase up to 25% of talk time.

ANADIGICS of HELP3™ power amplifier, such as AWT6221 power amplifier uses the company's exclusive InGaP-Plus ™ technology, in the same InGaPGaAs die, which integrates bipolar and field-effect transistor (FET) devices. Through selectable bias modes, HELP3 ™ power amplifier in the low range and mid-range output power levels in the present the best efficiency. Intelligent bias circuit of AWT6221 down at low power levels can be reduced to 8 milliamperes current power - best in the world. By integrating two separate amplifier chains, super-mini AWT6221 in these two frequency bands to provide superior performance, and save the printed circuit board area. (Figure 1)

ANADIGICS company is a combination of popular band dual-band WCDMA / HSPA Power Amplifier Module for CDMA / EVDO applications, dual-band power amplifier module. More information can be found ANADIGICS Inc. ANADIGICS company has developed a reference design products to help customers faster, easier to achieve their design goals. In this article we will describe the AWT6221 (Figure 2) reference design advantage.

Figure 1. Schematic AWT6221RDAFigure 2. Prototype AWT6221RDA

AWT6221 is a dual-band WCDMA / HSPA mobile UMTS band 2 (824-849MH) and band 5 (1850-1910MHz) and developed. 3GPP technical specification of the user equipment level 3 power, must meet or exceed these minimum performance levels (3)

• Maximum output power: +24 dBm +1 /-3dBm
• Adjacent Channel Leakage Power Ratio (ACLR) + /-5MHz:-33dBm
• Adjacent Channel Leakage Power Ratio (ACLR) + /-10MHz:-43dBm
• Frequency :30–1000MHz:-36dBm/100kHz

• In the between 1-12.75GHz: -30dBm/1MHz
• In the between 869-894MHz: -60dBm/3.84MHz
• In the between 1930-1990MHz: -60dBm/3.84MHz
• In the between 2100-2170MHz: -60dBm/3.84MHz

Most of the CDMA and WCDMA networks, frequency-division duplex systems allow different radio frequency bands allocated to the uplink (from mobile to base station) and downlink (from base station to mobile) link at the same time. Duplexers for mobile devices to allow the UL and DL use a single antenna. UMTS band 2, we have chosen to Avago Technologies FBAR duplexer micro-high-Q response to ACMD-7403. For Band5, we have chosen to Epcos low-loss SAW duplex B-7663. The reference design also includes thin-film power directional coupler CP402A coupling AVX, for a number of top-3G chipset to provide the required RF power.

AWT6221 the development of reference design starting in the power amplifier to further coupling with the duplex S parameter measurements. In this measurement, the zero-ohm resistors are used to shorten the circuit board matching components (Figure 3).

Figure 3. Schematic Diagram AWT6221RDA

Describe the N-port S parameters for each one-port network response of the voltage signal. The first number of the subscript refers to the response to mouth, while the second figure refers to the case mouth. S21 refers to the signal side, therefore a result of two reactions in the mouth. S parameters of an equal number of rows and columns from the matrix port. S matrix along the diagonal of the parameters are known as the reflection coefficient, diagonal S parameters are referred to as transmission coefficients, because they only occur in a single port. S matrix of two-port network in the table below:

Reflection coefficient (S11) for band 2 and band 5 the frequency range of RF paths to test. Power amplifier at the output of the circuit is not in line with the formation of a series of impedance changes, the results displayed in the Smith form (Figure 4, Figure 5).

Figure 4. Cell Band: PA output to Antenna

Figure 5. PCS Band: PA output to Antenna

For any power amplifier, the output impedance have a major impact in the operation, and a strong influence on the transfer to the antenna, linear (ACLR) and operational efficiency of power. ANADIGICS power amplifiers as the vendor can help engineers to do a good job balancing the load-pull of information. Optimum output impedance will be other characteristics of RF chain, the influence of components such as diplexers and switches are not always 50Ω. Variation frequency band characteristics of the advantages of duplex and operating temperature, in well-developed WCDMA RF design, particularly challenging.

Designed with Discrete Components, Complementary Symmetry Power Amplifier

1. Power Amplifier Basic Circuit Characteristics

Complementary symmetric OTL amplifier basic circuit as shown in Figure 1. Where: C1 for the signal input coupling element, should be noted that the actual circuit in the polarity should be consistent with the potential situation. R1 and R2 bias circuit composed of BG1, BG1 to provide a static operating point, but also in the entire circuit to play the role of DC negative feedback. Asked to approve R1 is greater than the current BG1 base current of at least five times, according to β is 100, Ic1 for 2mA calculation, R1 should be no more than 6k, it is given as a 5.1k; C1 accordingly, to set at 22μ, it is 20Hz signal on the impedance of 362Ω; R2 must be in accordance with the specific voltage of the power to determine, or about R1 (E/2-0.6) / 0.6, in accordance with 32V voltage value should be taken as about 120K, the exact value of the actual debugging by making BG1 collector voltage of 15.4V to get.

C2 and R3 constitute a bootstrap circuit, require R3C2> 1 / 10, (R3 + R4) Ic1 = E/2-1.2, due to the exchange of R4 is the BG1 load resistance, as far as possible to take bigger, R3 normally taken within the 1k. In accordance with 32V supply voltage value and Ic1 to 2mA calculations, R3 and R4 of and for the 7.2k, the actual will of R3 to 820Ω, R4 to the 6.8k, Ic1 compared to 1.94mA; C2 is therefore desirable to 220μ.

R5 and D are BG2, BG3 complementary bias control circuit components, to BG2, BG3 together provide a suitable quiescent point, to eliminate crossover distortion as much as take a small value, according to the experimental results are generally obtained in the 3mA ~ 4mA; R5 Resistance to change can BG2 and BG3 base voltage drop between the changes in the work to achieve the adjustment of its static, and R5 in series D is to compensate for BG2, BG3 launch junction threshold voltage changes with temperature, preferably in two only the series diode junction fired up and compensate for the threshold voltage of complementary tube changes with temperature, so that complementary management quiescent operating point stability. Omitted to simplify the use of a diode circuit. In parallel with BG2, BG3 base between the C4, can reduce the dynamic work of the ΔUAB, generally taken to be 47μ; C3 is to prevent the BG1 the exchange of high-frequency self-excited negative feedback capacitor, generally taken to be 47P ~ 200P.

BG1 from voltage amplification role in this circuit is known as the incentive level, require Buceo> E, Iceo ≤ Ic1/400 = 5μA, β = 100 ~ 200, so it should use low-power low-noise transistor. BG2 and BG3 are complementary current amplification pole, respectively, and BG4, BG5 constitute a composite tube to amplify the output current require Buceo> E, Iceo ≤ Ic2/100 = 30μA, β = 100 ~ 200. In the BG4, BG5 using ordinary high-power tubes instead of three has been made within the composite pipe under high-power 3, BG2 and BG3 need to provide the post-class high-power three tubes of more than 100mA peak drive current, it should the use of power three tubes. BG4 and BG5 is responsible for the amplification of high-power output current control, quiescent current is desirable in the 10mA ~ 30mA, asked Buceo> E, Iceo ≤ Ic4/100 = 0.1mA, β = 50 ~ 100. BG4 and BG5 the maximum limit of current Imax should be greater than the largest amplitude of the output current times the big one, can ensure the maximum amplitude of the output current when β> 10.

R6 and R7 are BG4 and BG5 quiescent operating point adjustment of shunt resistor, the dynamic work of the streaming effect can be negligible. The Ube4 and Ube5 are equal to 0.6V standard parameters, by the complementary current amplification level obtained in the static working current of 3mA ~ 4mA, calculate the R6 and R7 should be taken as 220Ω. In fact, the power may be a difference of three tubes Ube larger, BG4 and BG5 of Ube be carried out by matching the measured use of the bootstrap circuit to work with half the total current magnification composite pipe shall be constructed without the aid of the work of the bootstrap circuit The other half of the composite pipe is smaller.

R8 and R9 are to prevent the BG4 and BG overcurrent limiting resistor, generally take between the 0.2Ω ~ 0.5Ω. Will use the 200mm long, diameter φ0.08 both ends of the enameled wire were welded over the 1k resistor at both ends to fold up enameled wire can be wrapped around the resistor. Is equivalent to the role of fuse fuse belongs to the simplest non-smart Limit blown protection methods.

C5 and C6 is the signal output capacitor, with a small capacitor in parallel with the large capacitor up and use, it can eliminate large-capacity capacitor has a larger internal inductance for high-frequency signals obstacles. Note that it is actually played the role of the mid-point of floating power supply, so power is not in accordance with the capacity to fight with a right pass the bottom of AC impedance of the signal should be how to calculate, but according to how much energy consumption of the output power required is calculated. In the mid-point of the floating power supply voltage fluctuations as a result of the output current cut-off wave output signal, they will have serious clipping. According to the energy storage capacitor voltage is proportional to the square with the relationship between the midpoint of the floating power supply output capacitor, the total power supply capacity should be 4 times the energy storage capacitance.

C9 and R10 is the exchange of negative feedback network, and R2, R1 together form the voltage in parallel with negative anti-ashamed. R2 and R1 anti ashamed constitute a negative DC voltage allows the total magnification approximately equal to R2 divided by 1.2k (equal to R1 with the launch of BG1 junction dynamic resistance in parallel), the design parameters in accordance with Figure ① about 100 times, adding C9 and R10 After the exchange of negative feedback network, the total voltage magnification approximately equal to R2 and the R10's parallel resistance divided by the 1.2k, is about 18 times. Practice has proved that the work in this way the voltage performance of parallel negative effects of anti-ashamed very good.

2. Improvement of Basic Circuit of the Power Amplifier

As shown in Figure 2, complementary symmetric OTL amplifier basic circuit, the signal input level resistance incentives only 1k, impedance needs to be done in order to transform the vast majority, the high impedance signal source match. Incentive level direct the signal input into a composite pipe is the easiest way to composite pipe connected in a variety of specific circuit. Additional pre-class practice is equivalent to a simple voltage-controlled current-mode operational amplifier, BG0 the base and the emitter is equivalent to op-amp positive input and negative input, positive input of the dynamic resistance has been increased to more than 10K. Meanwhile, the output from the power amplifier receiving the negative input emitter of negative feedback resistor R10 and sampling resistor R11 determines the ratio of the total voltage magnification.

Debugging is also the first points of the circuit R5 adjusted to short-circuit resistance status of 0 to BG2 ~ BG5 in a cut-off state, respectively, with two 1K/2W resistance received from the total power output at both ends to obtain the midpoint voltage. Instead of using a 200K potentiometer R1 or R2 connected to the circuit board, with wires to the input signal input capacitance C1 and ground short-circuit.

Connected to power, measure BG1 of the collector to emitter voltage drop Uce, regulation allows Uce equal to 200K potential E/2-0.6; in the total supply voltage of 32V, when, BG1 static Uce should be equal to 15.4V(0.1V). Then measured the actual 200K potentiometer in which the resistance value, replace the fixed resistor with resistance potentiometer replacement, and then measuring BG1 static Uce should 15.4V(0.2V) within. To determine the static Uce good BG1 again after small to large BG4 and BG5 adjust R5 so that the quiescent operating current of 15mA. For insurance purposes, R8 and R9 can be for access into 100Ω/2W resistors, R8 and R9 on the first measurement of the static voltage drop should be 1.5V.

Disconnect the power supply, measuring the actual resistance of R5 in which adjustable resistor values, resistor R5 into the same fixed resistors, removed earlier from the output side are connected to the power supply at both ends of 1k/2W divider resistors. Re-connect power, R8 and R9 on the measurement of the static voltage drop should be maintained at between 1.2V ~ 1.8V. Measuring the output level should be the midpoint between the 16V(0.5V). The C1 input capacitance signal input terminal and the earth floating disconnected, measure the voltage drop R8 and R9, C1 encounter with the screwdriver when the input voltage on R8 and R9 dropped significantly larger. And then replace the 0.3Ω resistors R8 and R9, connected to the horn Lyrics. Connected to the power due to C0 charge, midpoint voltage of the output needs to rise slowly from zero, which produced only a slight impact on sound. When the hands touch the C1 input speaker will issue a "woo" the exchange of sound. The C1 input and ground (negative power supply side) short circuit, speakers should not sound, the actual will be issued by a slight background of white noise or very low voice communication hum.

As shown in Figure 2 complementary symmetry power amplifier improvement type OTL circuit, there is a clear disadvantage is that the input signal DC level than the midpoint of the output voltage is lower 2V ~ 3V, not in the public use and transistor op amp IC components, the high cost of the 20th century, early 80s, it has very good power amplifier using a single power supply practical circuit. The mid-20th century, 80, op amp IC, starting, people began to use pre-op amp IC to serve as a very great and inspiring.

A typical circuit shown in Figure 3 due to op amp IC do not need to adjust the static working point, as long as BG2 ~ BG5 adjust R5 so that the quiescent operating current of 10mA ~ 20mA can be. Note that, although the op amp IC do not need to adjust the quiescent point, but in BG2 ~ BG5 in a cut-off state, from R8, R9, and BG3, BG5 emission knot is wizard-pass will be the negative input of op amp IC is set to high, transport, Z IC output is low, then fired through the BG3 junction to the negative input of op amp IC is set to low, the output of op amp IC turning into a high, resulting in the output variable of the low-frequency oscillation state can not provide a stable The reference to the midpoint level.

In this situation, adjust the BG2 ~ BG5 quiescent operating current, op amp IC output is high when the regulator R5 is invalid; while IC op amp output to '0 BG5 can not turn to regulate the R5 can only BG2, BG3 , BG4 to enter the work area, BG2 actually only played the role of the diode by BG4, and BG2 the current direct all poured into the op amp IC output, resulting in BG2, and op amp IC damage due to over-current! (I have to hand at that time owned by a few domestic op amp IC and a dozen medium-power three pipes all the damage, but also failed to adjust the quiescent operating point out.) Must first IC op amp with a wire to the negative input and output connectivity, temporarily connected to the negative feedback resistor R6, so that op amp IC in follower mode output stability level of the mid-point of reference, in this state, adjust R5 so that BG2 ~ BG5 the quiescent operating current of 15mA, the R5 resistance into the same fixed resistor confirm BG2 ~ BG5 quiescent operating current of 10mA ~ 20mA in between, and then the negative op amp IC input and output end to end open, the feedback resistor R6 access circuits.

Use as a pre-op amp IC and motivate highly polar, the best to BG2 ~ BG5 quiescent operating current bias mode into resistance from the transistor and the voltage divider composed of regulators, so that there is a greater power supply voltage can be changed under the to maintain almost the same quiescent operating current.

Figure 4 that is, after the improved circuit, BG1 launch junction threshold voltage and BG2, BG3, BG4 the threshold voltage with temperature change, itself, can play a role in temperature compensation. In order to reduce the output of op amp IC quiescent operating current in the output of op amp IC gifts added to the ground side of the shunt resistor R10. With the shunt resistor, the adjustment of BG2 ~ BG5 quiescent operating current can not access the first op-amp IC, directly from one of R7, R8, and R10 divider out the approximate mid-point of reference level. Start with 0 to large adjust R5 so that BG2 ~ BG5 quiescent operating current of 10mA ~ 20mA in between, and then access to op amp IC, the circuit that is able to work properly.

In addition, the output of op amp IC 1 1k current-limiting resistor in series R15, can guarantee the output op amp IC is '0 BG5 will not enter the cut-off state.

Used as a pre-op amp IC and motivate highly polar, the biggest advantage is that the output DC level input signal DC level strict consensus, a difference of not more than 0.05V. This can create OTL power amplifier consisting of two inverted output BTL power amplifier, while in the output DC level and DC level input signal are significant differences between cases, the two OTL power amplifier positive, inverting output DC level tend to be higher than 0.5V, significant effect on the work of speakers equilibrium position. BTL power amplifier positive, RP-output DC level difference between the DC level must be less than 0.1V, trumpet the work of the apparent equilibrium position will not deviate from the free position of equilibrium. Horn's work significantly deviate from the equilibrium position of the free position of equilibrium, the positive and negative direction of the mechanical vibration amplitude asymmetry, issued by the sound waves will produce distorted unnatural.

In addition, the output DC level input signal DC level strict consensus, which makes the use of positive and negative dual power supply OCL power amplifier become a reality. Otherwise, the result of the output DC level and power level of the mid-point difference between the large, will lead to a good speaker does not work properly.

As most of the op amp IC's operating voltage is not high, good performance of high-voltage op amp IC varieties of small, high prices, people can also be used with the same level pre-op amp IC differential amplifier circuit to achieve the same purpose. Figure 5 differential amplification method that is used to do most of the typical pre-circuit than shown in Figure 5 complementary symmetric OTL amplifier basic circuit characteristics of multi-purpose two identical transistors required than shown in Figure 5 improved complementary symmetric multi-purpose OTL amplifier circuits a practical transistor.

Said that they going to, not that people do not know how to design power amplifiers, but by the device of choice constraints, in different historical periods can only use the appropriate design of the circuit. In the 20th century, the late '80s, people began to more easily find properties matching using the same transistor. Due to differential amplification circuit can be extremely quiescent current and accurate design parameters for a given, do not adjust the differential amplifier tubes with a quiescent current. The circuit in Figure 5 the case of the use of 32V power, pre-differential amplifier tubes with a static current of 0.51mA ~ 0.52mA, as long as the first adjust R12 so that BG1 collector to ground terminal of the voltage dropped to 15.4V, and then adjust R5 so that BG2 ~ BG5 quiescent operating current can be between at 10mA ~ 20mA.

BG1 in adjusting the quiescent current, the same they must be adjusted to short-circuit 0 resistor R5 conditions make it BG2 ~ BG5 in a cut-off state, withhold access to the negative feedback resistor R10, with the wire will be BG6, BG0 base short-circuit.

Connected to power, adjust R12 so that the first BG1 collector to ground terminal voltage reduced 15.4V0.2V, and then adjust R5 so that BG2 ~ BG5 the quiescent operating current of 15mA. For insurance purposes, the first R8 and R9 for access into the 100Ω/2W resistors R8 and R9 on the measurement of the static voltage drop should be 1.5V. Disconnect the power supply, measuring adjustable resistor R5 and R12 in which the resistance of the actual value of their fixed resistance into the same resistance. Connected to power, R8 and R9 on the measurement of the static voltage drop should be maintained at between 1.2V ~ 1.8V. Measuring the output level should be the midpoint between the 16V0.3V. Disconnect the power, the BG6, BG0 base Qudiao between the connection wires to the negative feedback resistor R10 access to the circuit. Re-connect power, R8 and R9 on the measurement of the static voltage drop should be maintained at between 1.2V ~ 1.8V. Measuring the output level should be 16V0.2V the mid-point between the difference in charge current magnification the greater the output DC level and signal the smaller the difference between the input DC level. C1 touch with the screwdriver when the input voltage on R8 and R9 dropped significantly larger. And then replace the 0.3Ω resistors R8 and R9, connected to the horn Lyrics. When connected to the power output of the midpoint voltage of the need to rise slowly from zero, which produced only a slight impact on sound. 2 seconds later, the hands touch the C1 input Speaker will issue a "woo" the exchange of sound. The C1 input and ground (negative power supply side) short circuit, speakers should not sound, the actual will be issued by a slight background of white noise or very low voice communication hum.

3. The Improvement of the Power Amplifier Circuit Useful

Bootstrap circuit design using the power amplifier while the circuit is relatively simple, but there is lower frequency cut-off point. The introduction of the bootstrap circuit is in order to avoid when you zoom in on a half-wave there is not enough current available to complement each other to control the use, without lack of transistor used in cases, constant-current source can be used to ensure the right to zoom in on a half-wave when there is sufficient Current available to complement each other to control the use. In the meantime, the differential amplifier is also designed to provide work from the constant current source current, can greatly enhance the suppression of common-mode noise voltage ratio and relax the accuracy requirement.

Figure 6 is to use a typical constant current source power amplifier circuit, in which: BG3 and BG4 constitute a standard constant current source, the former to the front differential amplification to provide extremely 1mA constant total current, 2 only a thin charge of BG1, BG2 be 0.5mA each of the quiescent operating current; which provides a constant current of 2mA, and inspire a very BG5 quiescent operating current of 2mA equal, so that the static Q amplifier output voltage is entirely by the mid-point resistance of the same partial pressure of R13 and R14 set out, are not so large deviation from E / 2. R14 series at the bottom on the D1 is a composite pipe to compensate for the threshold voltage drop above the bottom of a single complementary than the threshold voltage drop across the tube over a PN junction voltage drop, ensuring the same from the resistance voltage divider R13 and R14 determine the midpoint voltage out of more accurate.

Stimulate the quiescent operating current is very BG5 has been on a 1V drop from the R4 and R12 resistance 200Ω identified as 2mA, also do not adjust. Therefore, in the regulation of BG7 ~ BG10 quiescent operating current when the first non-access BG4 and BG5, directly in the R13 and R14 divider out of the mid-point reference voltage and provide 0.4mA ~ 1.1mA of bias current to the BG6 working conditions, Starting from the smallest zero resistance BG7 ~ BG10 adjust R10 so that the quiescent operating current of 15 mA can be. Then replaced by a fixed resistor R10, will BG4 and BG5 access to the circuit board, amplifier immediately work properly. Although the existence of discrete element parameters, which may cause a very real incentive BG5 quiescent operating current and BG4 a small amount of difference between the current source current, differential amplification most will be the output terminal Q of the steady-state voltage deviation from the midpoint of the actual situation of automatically change the static BG1 operating current, so that the actual BG5 quiescent operating current and BG4 constant source current exactly equal. Of course, the actual quiescent operating current BG5 automatically adjusted, the differential amplification extremely quiescent operating current does not allow any one of them significantly reduced too much.

In accordance with Figure 6 in the component parameters, as long as you can make BG5 change 0.1mA quiescent operating current change 1mA, sufficient to achieve the right BG5 quiescent operating current adjustment.

However, due to constant-current source limits the incentive is a very cut-off state can provide maximum current to improve power supply voltage can not be a corresponding increase in the output amplitude. While the corresponding increase in the constant current source current can increase the output amplitude, but to inspire a very quiescent operating current will also increase accordingly, stability deteriorated.

A better approach is to introduce mirror circuit, using the upper and lower symmetrical differential current-mode driver to enlarge the back of the work of complementary symmetry power amplifier tubes. Figure 7 that is used to enlarge the upper and lower symmetrical differential current-mode power amplifier to inspire a very practical circuit, due to higher output power in order to avoid overload damage to the device, circuit added to limit the maximum output current protection. Which, BG4 and BG5 constitute a mirror circuit operating current Ic5 can BG5 and BG4 operating current Ic4 remain completely equal, then the drive BG6.

Achieved by BG6, BG7 constitute the upper and lower symmetrical differential current amplification method. In this way, you can ensure that in the last half-wave signals need to stimulate greater drive current when a very, BG6 also simultaneously be able to export more to the posterior pole of the drive current power amplifier tube. To achieve the same purpose, people can also be taken in parallel and then a pair of complementary symmetrical front differential amplifier, which it has realized BG6 drive. However, the mirror circuit components requirements have not pre-high differential amplifier, using two pairs of differential pre-amplifier and the circuit can not increase any performance, no need to spend the cost of using the kind of awkward. Way to adjust the circuit shown in Figure 6 the use of the same constant current source power amplifier circuit.

From the working principle to consider, using the upper and lower symmetrical differential current amplification methods for the circuit is no longer a very encouraging defects. However, the characteristics of high-power transistors are not ideal, the output current reaches 1A above, the current magnification is only 10 ~ 25, will make driving high-power triode tube complement the work of the above must provide more than 200mA current to the posterior pole.

Complementary control its own power, often exceeding 2W, severe fever, complementary management also need to install another heat-sink. In the electronic component manufacturers have developed to produce high-power Darlington circumstances, the switch has been made within the Darlington composite pipe to make the final stage current amplification tube, we can greatly reduce complementary tube output drive current. Such as the SGS produced TIP series of high-power Darlington, in the output current to 2A above, the current magnification can reach 500 or above, which only need to complement each other with providing 20mA drive current to the posterior pole following the work of complementary tube itself power consumption to below 0.2W.

Need to modify the design parameters only based on the threshold voltage of Darlington transistor threshold voltage is equal to two ordinary times, to provide quiescent bias current shunt resistors R18, R19 doubling of resistance in order to maintain control of the quiescent operating current of complementary does not change. At the same time complementary tube BG9, BG10 base-level voltage drop between the previous increase in the threshold voltage of an ordinary transistor, its quiescent current adjustment circuit means no impact.

As Darlington is not designed for audio power amplifier developed device, the working frequency of the ceiling is not high. Limit the frequency of regular high-power transistor reached only 1MHz, developed specifically for audio power amplifiers to achieve high-power transistor can only 10MHz, preferably no more than 100MHz. Although only the audio range 10Hz ~ 20kHz, but transistor current magnification and the frequency of related, at the upper limit of operating frequency, the current magnification will drop to 1-fold.

This makes the low frequency limit of 20kHz treble amplification transistor capacity than 2kHz Alto amplification of our ability to make low, thus leading to the open-loop state, treble and tenor of the current magnification is no longer remain the same. The closed-loop negative feedback to maintain the same throughout the audio sampling magnification does not change the current mixed-signal inside pitch magnification than the tenor current situation of the low magnification, so that mixed-signal inside the treble actually higher than alto's lower magnification. Therefore, the use of high-frequency high-power transistor limit will enable mixed-signal inside Treble current magnification magnification than the tenor current decline was less.

If you are using only reached the maximum frequency of 1MHz production of high-power transistor audio power amplifier, will feel more than treble 8kHz component is seriously insufficient. Therefore, the foreign manufacturer of electronic components, has been 90 years in the 20th century developed a superior audio performance, dedicated high-power transistor amplifier. Japan 3 3 Ken Ken pipe manufactured by well-known audio power amplifier is the first dedicated high-power transistors, but they are not Darlington require superior performance of the power to do the same before the driver is a very, but also give the driver the front pole of the Power to install heat-sink.

To the 20th century, the late '80s, it developed a higher performance of high-power FET. Any high-power FET operating frequency limit can also be reached 100MHz, but at first the lack of high-voltage high-power FET, manufacturers produce more than 40W power amplifier output power or in selected high-power transistor.

In fact, making use of high-power FET amplifier than the use of high-power transistor amplifier made more convenient. But require special attention to the fact, although the effect of pipe is voltage-controlled device, but the high power FET input gate and the source exists between the large junction capacitance can be achieved around 800P, and therefore the higher the status of the operating frequency to be provided under the same charge and discharge 5mA ~ 10mA drive current.

Channeling Alliance in front of the gate resistance will affect the input junction capacitance charging and discharging, as far as possible to take a small resistance. Figure 11 that is, the practical use of high-power FET amplifier circuit, due to some high-power FET gate voltage there is no built-in limit to protect voltage regulator tube, specially in the circuit by adding a limit-voltage protection voltage regulator tube. With no built-in limit-voltage protection regulator tube high-power FET, welding wire must be used to gate and source short-circuit, welding good power FET and current-limit-voltage protection regulator tube before the gate and the source short-circuit between the wires removed. Use of high-power FET amplifier design, debug mode with the use of high-power transistor amplifier design is identical.

Should be noted that the high-power FET threshold voltage between 2V ~ 3V, (produced by Samsung, the threshold voltage of high-power FET, mostly 2V), the actual high-power FET operating voltage should not exceed the maximum allowable voltage is half the value of the maximum working peak current should not exceed the allowable current of 2 / 3 can only work to ensure safe and reliable.

This requirement is already higher than the three requirements of a wide number of machine control, machine control the actual work of the three voltage can not exceed half the value of the maximum allowable voltage, while the three-machine tube can not exceed the maximum operating current of the maximum allowable peak current of 1 / 3 before being work.

High-power FET has a great advantage is temperature-stable performance is very good, from 25 ℃ ~ 125 ℃, almost exactly the same job characteristics. Therefore, the use of high-power FET, the heat-sink can also be appropriate to allow the temperature as high as 90 ℃, while the transistors are still two of this breakdown may actually allow the temperature of the work should be limited to below 70 ℃.

4. Power Supply "High-Performance" for Power Amplifier

Did not strictly control the output voltage of the midpoint value in requesting an ideal situation, power amplifier can only use a single power supply, the midpoint of the floating power supply with automatic follow ways. Just give a sufficiently large capacity storage capacitors, the actual output capability with the use of dual-power supply there is no difference between OCL output mode.

The reason for using OCL output mode, in addition to surface design a better performance can be further power amplifier, the greater practical significance is the use of positive and negative dual power supply output mode can further reduce OCL circuit background noise. In the pre-power amplifier input stage uses a differential amplifier circuit, the output DC level has been input with the signal DC level remain basically the same, a difference of less than 0.2V. In such a situation, the input signal DC level bias resistor connected to the positive and negative dual power supply potential on the midpoint, you can put a single power supply of the OTL output mode using the positive and negative dual power supply into the OCL Output way, no longer used automatically to follow the mid-point of the floating power supply.

In fact, the use of pre-op amp IC input stage can make the output DC level and signal input DC level remained nearly the same, a difference of less than 0.02V, it is precisely because the internal op amp IC also used to do a differential amplifier input stage , and generally the way of composite pipe to do the input stage differential amplifier circuit, so that flows into or out of IC positive and negative input quiescent current less than 0.1μA, the negative feedback resistor on the static DC voltage drop has been less than 0.01 V.

Identifying characteristics are always matching pipe, composite pipe can be used of course, the way to do input stage differential amplifier circuit, so that the output DC level and signal input DC level remained nearly the same, a difference of less than 0.02V, very consistent characteristic need a pair of tubes made of semiconductor materials, this is the op amp IC production process advantages. In short, the only way to change the OTL output OCL output mode, in the circuit design does not have any increase.

In fact, with work produced in class AB type power amplifier complementary symmetry there is a flaw, which is the last stage high-power amplifier tubes, static current is near the end zone, regardless of the use of high-power three tubes, or the use of high-power FET , in the end zone in the vicinity of the dynamic resistance were significantly higher than the linear dynamic resistance area should be large enough number, the actual can vary by several times to 10 times. Quiescent current is smaller, the greater the dynamic resistance. When the amplifier output voltage is zero when the speaker vibration basin will continue to be damped oscillation to stop. Voice coil moving in a magnetic field generated by currents will hinder the free vibration speaker vibration basin, if the amplifier in series with the voice coil resistance relatively large, it will cause voice coil moving in a magnetic field generated by currents decrease, reducing the role of resistance Nepal, vibration basin Vibration damping is not easy to stop down, the voice of a "dragging its feet" in divergence Shoubu Zhu situation. At the same time, the voice coil woofer speaker moving in a magnetic field induced currents produced by power amplifiers that can not be shorted out as much as possible, it would become alt prevent interference with the work unit loudspeaker drive signals. A better reason why the replay amplifier sound quality, mystery lies in the fact that it has a very low quiescent output impedance.

A power amplifier due to large, fever serious, should not be used under large operating voltage. To do this, you can use the high and low supply two sets of positive and negative manner on the final stage power amplifier tubes with current working status of implementation of the dynamic bias, so that amplifier output voltage amplitude is less than 4V when the high-power amplifier tubes with working in a current class status, the output amplitude is greater than 4V, when transformed to B status. As the rotation of working in the high-power current amplification tube is always in high-current mode of operation, the actual effect is the same way as the work of a purely A.

Figure 10 that is designed using high-power Darlington A high efficiency power amplifier typically dynamic bias circuit, in order to achieve a better dynamic offset, T1, T2 and down two high-power Darlington Complementary tube, in order to increase the threshold voltage bias circuit. Require two complementary parameters governing the same, the actual difference between the current magnification should not exceed 20%.

Dynamic bias is the result of each one and a half-wave output signal through the 4V reference value transform, require dynamic bias changing the speed limit must be more than 20KHz frequency output signal of at least 100 times higher, photoelectric conversion device isolation frequency response to be at least 1MHz, The diodes used must also be high-speed pipe. When the output signal voltage is less than 4V, the optical output transistor is cut-off state, two complementary high-power current amplification tubes were biased in the quiescent current of 1A to work, and when the output signal voltage exceeds 4V, the optical output transistor in the conduction state of two complementary high-power current amplification tube was biased at 10mA quiescent current at work. However, as the output signal voltage exceeds 4V, the current high-power amplifier tubes operating current must exceed 0.5A, 4Ω load must exceed 1A, is also equivalent to the CPI actually work. In the meantime, the output signal when the voltage is 6V or less, BG11, BG12 in the cut-off state, T3, T4 Darlington switch is also closed, T1, T2 two complementary high-power amplifier tubes with current low-voltage power supply from 8V. In the output signal voltage exceeds 6V, when, BG11, BG12 in the conduction state, T3, T4 Darlington switch also turns on, T1, T2 two complementary current amplification tubes replaced by high-power 30V high-voltage power supply, so that High-power current amplification tubes to reduce power consumption.

In the N groove and high-voltage power FET P Erdaogou easy to buy cases, can switch to high power FET to produce high-efficiency power amplifier A dynamic bias. Similarly, T1, T2 and down two high-power FET to be used complementary tube to require two complementary parameters governing the same, the actual difference between the current magnification should not exceed 20%. Bias due to the use of dynamic methods of work, bias circuit to adjust the parameters slightly more complicated.

The specific way the same way as described above, first T1, T2 from the R11, R12 series set out quiescent current regulator out of 1A, then the appropriate allocation of both the actual resistance, so that when the R12 in the short-circuit T1, T2 of the quiescent current 2mA ~ 10mA. That is not completely closed, there is no need to transfer large.

A dynamic bias in view of the main purpose of PA is to reduce the output resistance amplifier itself, in the upper and lower high-power amplifier tubes in the appropriate series current limit protection resistor, the amplifier's maximum output current limit special design in the power part of the change among the circuit.

In this way, matching power amplifier with dynamic offset the use of the CPI level of the two groups also have positive and negative power supply designed to power supply.