Testbench - AVAB FQ10G Graphic Equalizer
Test Report on: Avab FQ10G Graphic Equaliser
Date: April 1975, £120.00
The AVAB FQ10G 10 channel graphic equaliser is a compact unit containing 10 active filters which work at octave intervals through the audio range! The lowest responding to 30Hz and the highest to 15.4KHz. Each filter is controlled by a slider control calibrated to give 12dB boost or 12dB cut at its characteristic frequency. A slider level control can cut or boost the whole spectrum by ±12dB. With all controls central the unit has flat frequency response and unity gain. Inputs and outputs are via standard jack sockets and a switch is provided which switches out the filter. Power is supplied by a battery eliminator (small mains operated power supply) which is provided. The equaliser front panel carries an on/off switch for the D. C. and a pilot light.
The unit is smartly finished in black paint and measures only 119/16 in. x 33/8 in. x 2¼ in. deep. This is very small for a system of such electronic complexity. All controls are accurately calibrated and work at the filter frequencies marked. The control layout is good and their functions obvious.
The FQ10G is designed for low level signals (nominal 10mV). There are two other equalisers in the range, both designed to work at higher signal levels (0dBm or 775mV) one a mono unit like the FQ1OG, and the other is stereo.
All the electronic components for the filters, including the slider pots, are mounted on one glass fibre printed circuit board, The input stage and level control slider are on a second board. The circuit uses 14 integrated circuits and two transistors which accounts for the price tag. The components are of good quality and correctly mounted. The standard of soldering on the circuit board was fair but the wires to the board were very badly soldered. One soldered joint was remade before power was applied because it was touching a mounting post. No service manual was provided so fault finding could be very difficult.
The first test was to feed a square wave, which contains a wide spectrum of frequencies, through the system.
The controls were then all set so that a square wave in gave an undistorted square wave out. This is the setting which gives flat frequency response. It also confirms the absence of phase distortion. To achieve this all controls were virtually on the zero dB mark.
The second test was to increase all the controls by the same amount. This should have boosted the whole level but still retained square wave response; but it was clear that something was wrong at high frequencies. Investigation showed that the 3.84 KHz filter was working at 17.2KHz. Obviously a fault, possibly a wrong component; but we were not able to confirm this. The following table sums up the filter performance.
|Measured Frequency||35Hz||60Hz||116Hz||254Hz||495Hz||1.01KHz||1.96KHz||17.2KHz||8.0KHz||15.9 KHz|
|Range at Calibrated||+10.7||+11.3||+11.3||+11.0||+11.0||+11.0||+11.4||+4.4||+11.4||+11.3|
Measured with 10 K ohm dummy source was -93.8dBm (16uV RMS) with all controls level. Therefore the signal to noise ratio relative to limiting is -91.6dB or, relative to a 10mV signal, -56dB.
Total harmonic distortion, which includes noise, measured with a signal level of 100mV RMS @ 1 KHz was 0.02% with the filter switched out; 0.03% with it switched in and controls level and, with the oscillator removed, the background noise gave the equivalent of 0.017% THD. These figures are obviously satisfactory.
Performance is good and the design well considered but the excellence in these aspects is wasted if soldered joints are bad, components are wrong or the equipment is not fully tested before leaving the factory.
Previous article in this issue:
Next article in this issue:
mu:zines is the result of thousands of hours of effort, and will require many thousands more going forward to reach our goals of getting all this content online.
If you value this resource, you can support this project - it really helps!