Dynaco SCA 35 stereo integrated amplifier
amplifier menuMain Menu
Vintage test report
 


Taken from 'The Gramophone' December 1964

The Dynakit series of amplifiers and tuners have been well known in the United States for a number of years, so when my father heard that they were to be imported into Britain he suggested I might like to report on one of the latest models for THE GRAMOPHONE. This sounded to me at first rather like taking coals to Newcastle, but since the war even that seems to be economically justifiable!

For my own use I would prefer to have a preamplifier and power amplifier in separate units (and these are available in the Dyna line), as this keeps the principal sources of heat and hum away from the low-level circuitry. However, for this report I chose the integrated unit, partly because of its probable appeal to those with limited space or on a limited budget, and partly to see how well the problems of combining everything in one chassis and maintaining a competitive price could be overcome without sacrifice in the quality of performance.

The two-stage preamplifier uses a single 12AX7 double triode for each channel, mounted with their associated circuitry on a small printed circuit board. It is used only for the pickup and tape head inputs, appropriate negative feedback circuits being selected by contacts on the input selector switch to provide equalization for the RIAA and NAB response curves. A pair of packaged electronic circuits (PEC's) are mounted on the input sockets and provide attenuation and equalization for high- level magnetic and ceramic pickup cartridges. An unusual feature of the circuit is the use of cathode-to-cathode positive feedback to increase the gain of the preamplifier. The preamplifier outputs are fed to further sections of the input switch, which can at this point select alternative high-level inputs such as radio or tape recorder amplifiers. The switch itself is particularly ingenious, accomplishing the selection of the appropriate input and equalization, and short-circuiting of unwanted inputs and the preamplifier when not in use, all in one switch wafer for each channel. The selected pair of signals are made available at a pair of sockets for a tape recorder, and are also passed on through a stereo-mono switch and a filter switch to a two-section ganged volume control. The filter switch enables series and shunt capacitors to be inserted into the signal circuit to restrict the response at high and low frequencies when a noisy programme source is used. Associated with the volume control is a 'loudness' switch which enables additional bass to be introduced automatically at low volume control settings if desired. (This is a feature which no amplifier designed for the American market can afford to be without, though I personally deplore its use and always leave the switch in the "off" position.) It is followed by a Baxendall-type tone control circuit, the unusual feature of which is the high impedance level at which it operates. This is achieved by the use of another pair of PEG'S mounted directly on the two sections of the bass and treble controls. These are followed by a balance control which enables the sensitivities of the two channels to be adjusted or either one of them to be cut right off if desired.

Two separate identical printed circuit boards are used for the power amplifier. The pentode section of a 7199 valve is used as a voltage amplifier, its plate being directly coupled to the grid of the triode section, which is used as a cathodyne phase inverter. The operating conditions are ingeniously stabilized by a DC negative feedback circuit by feeding the screen grid of the pentode through a resistor from the cathode of the triode. A decoupling capacitor for the screen to cathode of the pentode applies positive cathode-to-cathode AC feedback to increase the gain of the stage. Push-pull 6BQ5's are used for the output stage, the screen grids being connected to taps on the output transformer primary to provide about 25% screen loading. The output transformer is of special design to minimize high frequency phase shift due to interwinding capacitance and leakage reactance. Negative feedback is applied from the secondary to the cathode of the voltage amplifier stage, and additional feedback is applied through a small capacitor from the appropriate 6BQ5 screen grid to provide high frequency compensation. A common cathode biasing-resistor is used for the output valves of both channels. Separate heater windings on the power transformer are used for the two channels, each with a hum balancing potentiometer whose slider is connected to the,output valves' cathodes. A pair of silicon diodes in a full-wave circuit is used for the common plate supply, with heavy resistor-capacitor smoothing for the early stages. (I understand that the same circuitry is used for the separate "Stereo 35" amplifier, with the exception that as less gain is required a 7247 double triode is used in place of the 7199.)

I was impressed with the thought that had evidently been given to the chassis design and layout. The assembly is in three main sections. The three printed circuit boards have' the components already in place and are mounted in cut-outs in the main chassis, which also carries the output transformers and the power supply circuitry. The front panel on which the operating controls are mounted and the to, panel on which the input sockets, output terminals and hum controls are mounted air spaced at a distance from the un-turned from and back edges of the main chassis which form an effective screen for these components.

I found the instructions very clear and easy to follow, and all wire lengths accurately specified. The only awkward connections were those from the output transformers to the output terminals; other leads from the front or back panel to the main chassis are attached at one end while these three main assembly units are still separate, and there is plenty of room to work.

My father arrived on his recent visit to the USA just in time to participate in the excitement of the first listening tests. I believe he was not disappointed. (That is a considerable under-statement. I thought the quality A1 - P.W.) We were impressed by the clarity of reproduction and the complete absence of over-sibilance. Everything seemed to be working as it should, with the exception that the hum level was a little higher than I had expected (more about this later). We found it very easy to adjust the tone controls for a good square wave response at the radio input (the equalization circuits would prevent this at the low level inputs), and the balance between the two channels was very close. There was a slight spike on the leading edge of the square wave which was attributed to a small increase in value from normal of one of the capacitances in the tone control PEC, since it did not appear when the power amplifier was tested by itself. The effect did not show up in any of the frequency response measurements, which demonstrates how severe a square wave test is, and had no noticeable effect on listening performance. (It can be compensated for in the power amplifier by a small increase in the value of the screen feedback capacitor, but I would not advise anyone to attempt this adjustment without the use of a square-wave generator and oscilloscope.)

The results of the more detailed lab tests were equally gratifying. It faithfully delivered the specified 174 Watts sine wave power into a 16 Ohm resistive load for the specified input level with less than 1% distortion — except at the highest frequencies. (The distortion level rose to 2½ % at 10,000 c/s, mostly third harmonic; "though the ear cannot hear as high as that it might attract any passing bat!") At normal listening levels the distortion was about 0.2%. The intermodulation distortion at full output was well within specifications and the channel separation was 38 dB at 10,000 c/s, approaching 50 dB at 100 c/s. The tone controls gave the desired 12'oB bass boost or cut at 50 c/s, and 12 dB treble cut, but only 9 dB boost, at 15 Kc/s, and were remarkably well balanced between channels throughout the range. With the filter circuit switched in, there was an additional attenuation of 3 dB at 80 c/s and 5 Kc/s, falling to 6 dB at 40 c/s and 8 Kc/s. The only difficulty was in the hum and noise level, which Was rather high. I found on careful examination that I had not accurately followed the pictorial diagram of the lead routing on the wiring to the printed circuit boards, nor had I twisted the heater leads as tightly as is shown. I found a considerable improvement on correcting this, and also on dressing the heater leads so that they would stand away from the circuit boards and in contact with the bottom cover. I found then by interchanging valves that one 12AX7 gave more hum than the other; Dynaco supply selected valves, made in West Germany, but a poor one must have slipped in somehow. Fortunately, I was able to go through half a dozen valves of two leading US brands and found one nearly as quiet as the better German one. I ended up with a hum and noise figure with the volume control flat out and the hum controls properly adjusted about 55 dB below full output. In listening, the hum was barely noticeable, and there was only a very slight hiss. With the controls set at a normal listening level I cannot tell whether the amplifier is switched on; that, of course, is why a pilot light is provided!

This is an amplifier of attractive appearance and superlative performance for its modest price. Its success is due to careful design and advanced circuitry which keeps the number of valves and other components to a minimum. It was a joy to build, and is a pleasure to hear.

GEOFFREY L. WILSON.

[I thought that the performance was remarkably like that of the Quad at its best (with the exception, of course, that there is no variable slope treble filter). So I made copies of the square wave responses at 100 c/s/1 Kc and 10 Kc/s and when I got back home compared them with those of the Quad. Apart from the tiny spike they were identical in all three cases! In short, a first-class integrated amplifier and control unit. P. W.]