Taken from 'Gramophone' November 1969

Ferrograph F307 Stereo Amplifier. Price: £56.
Manufactured by The Ferrograph Co. Ltd., Mercury House, 195 Knightsbridge, London, SW7.

The F307 stereo amplifier is Ferrograph's first venture into the solid-state domestic amplifier field and, visually as well as performance-wise, the amplifier appears to carry out the traditions of its tape recorder ancestors.

Its overall dimensions are 16 x 9¾ 5 inches including a well finished solid teak cabinet. The amplifier proper is carried on a pressed steel chassis, the front control panel being built up from aluminium extrusions. Slots in the base and sides provide adequate ventilation and the decor is austere and dignified, there being but four control knobs and a power switch (apart from the phone socket and indicator light) visible on the front panel. These controls are selector switch, volume, bass and treble; the variable controls are of the concentric knob type, thus enabling the operator to vary each channel independently. The 4-position input selector switch is pickup 1 (magnetic), pickup 2 (ceramic), radio and auxiliary.

Additional controls are hidden behind the flap underneath the main controls. These are loudspeaker/phone on/off switch, low-pass filter switch, pre-set tape input level controls (one for each channel), mode selector for stereo or mono from either left or right source and, finally, tape recorder switch for connecting amplifier either to tape recorder output or source. The facilities supported on the back panel are equally comprehensive: mains input is via a miniature 3-pin socket and is fused at 1 Amp; two additional switched mains sockets of the American 3 Amp flat-pin type are provided, whilst the mains transformer is tapped for 200-210 V, 220-230 V, and 240-250 V input. The DC supply to both channels is separately fused with 1 Amp miniature fuses. Loudspeaker connections are DIN non-reversible sockets; the input phono sockets are auxiliary, radio, ceramic pickup at 2 Megohms and magnetic pickup which is switched for an input load impedance of 47, 68 or 100 Kohms. Additionally, two pairs of phono sockets are provided for the input and output connections to a tape recorder.

The circuitry is mounted on a number of printed circuit boards, the right channel power amplifier board running from front to back along the right hand side of the chassis with the left hand channel symmetrically disposed on the other side. The power transformer, rectifiers, reservoir capacitors and power switch are contained in a shielded box on the main chassis frame. They are of generous proportions and the magnetic hum field is gratifyingly low; obviously, the manufacturers have drawn widely on their tape recorder experience during the design of this amplifier. The tone and gain control circuitry is on its own panel adjacent to the controls and mode switch, whilst the preamplifier is mounted underneath the main frame adjacent to the input controls. The layout is logical and should make for easy servicing in the unlikely event of this becoming necessary.

The input from the magnetic cartridge is via a two-stage preamplifier with overall feedback applying the necessary replay correction for a magnetic (velocity) cartridge. The output of this preamplifier is then connected via the input switch to an FET which forms the high impedance input circuit for the ceramic cartridge, radio and auxiliary circuits. The input of the FET is protected by a diode against transient overloads that could otherwise destroy this particular type of device. This FET operates as an emitter follower and the signal is applied via the switched low-pass filter and the feedback gain controls to the volume control and output transistor driving the power amplifier boards.

Unlike many transistor amplifiers, the low- pass filter, with -3dB points at 5, 7 and 10 kHz, has a steep slope of approximately 30 dB per octave and infinite rejection points at 8, 9.5 and 18 kHz. It is a passive network, using inductance and capacity instead of the more usual RC feedback network. The tone controls call for no particular comment, except that they are of exceptionally wide range.

The tape recorder socket is connected to the input of the main amplifier, whilst tape replay is taken via the pre-set gain control to the main volume control. Thus, although both tone controls and filter circuits are used for recording, the output from the tape recorder is taken straight to the main amplifier.

The power supply consists of two separate rectifier smoothing systems, one for the two output amplifiers and the other for the preamplifiers; the power supply line for the output circuit is 70V with transistors type 39285 arranged in quasi-complementary symmetry. The power amplifier circuit is straightforward and follows good engineering practice. The output transistors are driven by a complementary pair, in turn fed by a single ended transistor, DC feedback being applied. This network is AC coupled to the preamplifier with overall feedback from the loudspeaker to the emitter of this stage. Each loudspeaker is fed via a 1050-microfarad capacitor and 0.35 Ohm resistance.

Operation of the amplifier is simple and straightforward and all the controls are 'positive'; there is no slop or shake of the rotary controls and the push buttons and slide switches have well defined operating positions. Sometimes, even in the most expensive equipment, an otherwise excellent piece of apparatus is marred by lack of attention to small details such as these.

Performance tests
At first sight the claimed performance of 0.25% distortion at the rated output does not compare favourably with amplifiers claiming 0.1 and 0.02% distortion; but these very low distortion figures can be misleading because, if the distortion is due to 'crossover' asymmetry, it will contain mostly high order harmonics which are much more distressing than second or third. It must be remembered also that tracing distortion from the pickup in the higher frequencies, and non-linear distortion in the loudspeaker, especially at low frequencies, can and often do exceed 10%, although in both cases the distortion is usually of low order harmonics.

The claimed 0.25% distortion at the rated output over a frequency range of 20 Hz to 25 kHz is more than met (see Fig. 1). The power output between 100 Hz and 10 kHz is in excess of 21 watts, dropping to 18 watts at 25 kHz and 15 watts at 20 Hz.

The distortion versus power graph is shown in Fig. 2. It will be seen that the manufacturer's claims are very conservative indeed. More important is the fact that the 'knee' in the overload characteristic occurs at almost 50% greater power than claimed by the manufacturer. Again it will be noted (Fig. 2) that the distortion is principally second and third harmonic which, as before mentioned, are much less distressing than high order harmonics. Crossover distortion could not be detected at levels above 100 milliwatts and, below this level, the distortion products were below the noise level of the amplifier.

I have dealt in some detail with harmonic distortion because it is my opinion that this form of distortion is much more important than deficiencies in frequency response, etc. As before stated, the claims are realistic in practical terms and this excellent performance is due, no doubt, to the care which has been lavished on the power amplifier proper. In addition to the generous DC and AC feedback in the circuit, provision is made during test to adjust each amplifier individually for minimum crossover distortion and this, together with close tolerance components, should ensure the initial performance: being maintained throughout its life.

Hum and noise are claimed at -65 dB on high inputs and -60 dB on magnetic pickup input. These claims were met with about 3 dB in hand, and it should be noted that the background noise is almost entirely due to noise in the input circuits, hum being at least 12 dB below the general noise level. Crosstalk between channels is likewise good, being 65 dB at 1 kHz and below, degenerating to 50 dB at 15 kHz. This is much more than adequate when compared with an average of 26 dB for the best stereo tuners and pickups at mid-frequencies, generally deteriorating to 10 dB or less at frequencies above 12 kHz.

The frequency response is shown in Fig. 3. Curve A, 'flat' was obtained by setting the controls to the mid-point, no attempt being made to equalise the response at each end of the band. This meets the manufacturer's claim of ± 1 dB. On the same graph appears the range of bass and treble control. These are exceptionally wide, being approximately ±20 dB at 20 Hz and ±15 dB at 20 kHz. The equalisation for magnetic pickup again is excellent and is shown in Fig. 4., which plots the difference between the theoretical response and the actual amplifier response.

The variable low-pass filter response is shown in Fig. 5 and exhibits the familiar 'notch' of half-section, M-derived filters. It should prove useful in reducing surface noise on bad records, although in this day and age noisy records should be but a memory of the past. Just a thought, and not a criticism - if the infinite rejection point on the 10 kHz filter had been made at 19 kHz, this would have resulted in excellent suppression of the pilot tone on FM tuners and made life a little easier for the tape recorder enthusiast. (It may be noted in parenthesis that I do not know of any domestic amplifiers commercially available which do possess this feature.)

The input sensitivity was 2.2 millivolts for 15 watts output on magnetic cartridge, and 80 mV for radio, auxiliary and ceramic pickup inputs. A check on the overload capacity showed an acceptance of 80 mV al the magnetic pickup input and 7 Volts at the ceramic pickup/radio/auxiliary inputs for 0.5% distortion, these values being obtained with the gain control reduced to avoid overloading the power amplifier. Under all normal operating conditions it should be impossible to overload the input stages of this unit.

All measurements were made with an input mains voltage of 250 V, the mains voltage selector being set to 240-250 V. Quoted power output figures refer to true watts, not speech and music, IHFM ratings, or what have you. Under user conditions the amplifier handles well and the fact that the seldom-used controls are tucked away under a flap makes operation much easier by non-technical members of the family. The overall decor in light teak, brushed aluminium and light grey, make a pleasing contrast, and the unit is visually unobtrusive.

A well prepared and produced handbook is provided; the installation and operating instructions are precise and foolproof, whilst maintenance problems are minimised by excellent photographs in which all the components are numbered to correspond with the circuit schematic provided. Packing is excellent, the amplifier being supported by polyurethane sponge inner supports in a standard corrugated carton.

To summarise: this excellent amplifier carries on a long established tradition of sound engineering and good workmanship formulated over the past 20 years by Ferrograph; it is a stable companion of the Ferrograph 722P Stereo Recorder which was reviewed last month, and can confidently be recommended.