Armstrong Model A20 & PCU 25 control unit

amplifier menuMain Menu

Vintage test report


Taken from 'The Gramophone' April 1962

This is an entirely new stereo amplifier and preamplifier by Armstrong Ltd. The two main amplifiers are mounted on one chassis designed for fixing to the floor of the cabinet, while the control unit fits on the front panel. Push buttons are used for six functions of the control unit, resulting in a neat and simplified appearance. When the preamplifier was unpacked it was found to be what is known in space parlance as 'in go condition'. The instruction book, however, states that it is packed with front panel and control knobs removed. On the other hand, the main amplifier should have come out of the carton (according to the book) set for a mains voltage of 250 volts, but in fact the voltage selector plug was packed separately at the bottom of the box. I mention this because it might easily be possible for someone, having read the instruction book, to decide that the amplifier was in 'go' condition only to find that the count down must be halted while a feverish search is made of the bonfire heap!

Each main amplifier consists of a high gain triode stage coupled to a double triode phase splitter, followed by two EL84s in push-pull, using the familiar ultra linear circuit. A pre-set hum control is fitted for heater balance and power supply is by valve rectifier with resistor capacitor smoothing. The control unit has five knobs, which are, from left to right, input selector switch, bass, treble, volume and balance. Below the knobs and to the right are six push- buttons having the following functions: on/off, high frequency filter, low frequency filter (rumble), stereo, mono A-B, mono A. The last two mono functions are: mono A switches channel A inputs to both amplifiers so that a monophonic source such as radio will be reproduced through both loudspeakers. Mono A-B switches channel A and B inputs in parallel to both amplifiers, for use when a mono record is being played with a stereo pickup. For each channel there are six inputs. Tape 1, for direct connection to tape heads and compensated for CCIR characteristic. Tape 2, for reproducing the signal from a tape recorder or tape deck having its own electronics. Input 3, auxiliary, will accommodate a wide range of crystal pickups. Input 4, equalized to RIAA characteristic, is for low output magnetic pickups, Input 5, radio, and input 6, microphone. The sensitivity of these inputs is quoted in the specification but it is worth noting that the input voltage can be approximately 25dB above the figure quoted before overloading occurs. Thus, for example, input 3, auxiliary, can be anything between 100mV and 2 V without causing distortion in the input circuit of the control unit.

The treble control works in conjunction with the treble filter so that the turnover point is progressively lowered as treble cut is introduced. With treble at level, the filter produces a slope of 12dB per octave starting at 9 kc/s. Treble at 11 o'clock produces a slope of 9dB per octave starting at 7 kc/s. Treble at 10 o'clock gives a slope of 5dB per octave starting at 5 kc/s. This system provides, in effect, a variable frequency filter giving a wide range of control. The rumble filter comes into operation at 80 c/s with an initial slope of 6dB per octave, increasing at lower frequencies.

On test, the amplifier nobly supported the manufacturers' specification. The overall frequency response, taken by feeding a signal into the radio input at 1 watt level was 20-20,000 c/s ±1dB. Power output measured 12 watts at 1 kc/s, 9.6 watts at 50 c/s and 10.5 watts at 10 kc/s. The balance control gave a change of 10dB in each channel. In the specification the manufacturers give the range of the treble control as +15dB to -20dB at 20 kc/s. There seemed little point in measuring this outside the audible range and I therefore measured it at 10 kc/s where it gave an increase of +12dB and a decrease of —11dB with the H.F. filter out. The range of the bass control at 50 c/s was +13.5dB to -11dB. At 10 kc/s with treble control at level the filter produced a drop in high frequency response of —5dB and with full treble cut — 23dB. The low frequency rumble filter was checked at four frequencies: 100 c/s, -3dB; 50 c/s, -7dB; 30 c/s, -12dB; 20 c/s, —26dB. Both channels were checked with identical results. Negative feedback measured 29dB and stability was excellent. A - square wave at 1 kc/s was observed on the oscilloscope using a resistive load and showed no trace of ringing until a 0.5 mfd condenser was shunted across the loudspeaker terminals. This is remarkably good with such a high order of negative feedback and pays tribute to the design of the loudspeaker output transformers.

The construction of both units is rather different from that of many other amplifiers. Large tag boards with components neatly arranged are not used and my first impression was that the general appearance was untidy. On closer examination, however, the layout proved to be logical and efficient, and the components used are of high quality throughout. It is to be expected that an amplifier that per- forms so well on test will prove excellent under listening conditions and this is no exception. Listening tests were therefore conducted primarily with a view to finding out the effect of the treble/filter control when worn and noisy LPs were being played. The degree of control was entirely suitable for (these rather difficult conditions for it was possible to remove a large amount of unwanted noise whilst still retaining a good musical balance. Before I finish I should like to make two small criticisms. There are four screws securing the panel of the control unit to the chassis. On my unit they were nylon screws, obviously designed to merge into the Perspex panel. This they did not do, and since they were not symmetrical, tended to be more noticeable than perhaps if they had been of contrasting colour. The other criticism is concerned with the selector knob. This is stiff in action, as one would expect from the number of switch wafers and the type of index spring used. The knob is so small that a very firm grip indeed is required to turn it, a situation which, I am sure, could be easily remedied without spoiling the appearance. Happily, neither of these two points detracts from the performance of the amplifier, which is excellent.