Valve amp transformer repair

One of the most common failures of valve amplifiers (apart from valves and capacitors) occurs when the output matching transformer ceases to match. Under certain conditions of failure it is possible to give the amp a new lease of life by a simple re-wiring job on the transformer in question.

Most transformers have several taps to enable a wide variety of speaker impedances to be matched. These are arranged as shown in (a). If the coil between 0 volts and 4 ohms fails (usually) the output may be re-wired as (b) making use of another 4 ohm impedance winding on the transformer.

Having said this it is worth noting that this coil may be wound with lighter gauge wire than the original windings, and will thus be more prone to fail if the amp is run at full power for an extended period. Note also that it is important to ensure that the original winding is not shorted to earth at the point of failure (in which case it must be disconnected) or that the failure did not cause several turns to be shorted together, which would result in large eddy currents flowing within the transformer, giving rise to overheating, and loss of efficiency.

Pseudo phaser

Dave Rogers-Walton

Tone stops allow the electronic organ player to select from different fixed filters whilst playing and so are really a type of performance control. There is no reason why a similar facility should not be added to the more usual performance controls (e.g. pitch bend) found on synthesisers. What is needed is some form of one-knob, complex tone control which, unlike standard tone circuits, should be able to shift the balance within the important mid-band as well as at the extremes, so varying the whole character of the sound. The circuit shown does exactly this. It has also been found to have a useful effect at the output of guitar overdrive circuits.

RV1 simultaneously alters the time constants of several high and low-pass signal paths which are then remixed at the output. One of the results of this is the formation of a notch which varies in centre frequency as well as depth, depending on the setting of RV1. L1 forms resonant circuits with the capacitors over most of RV1s range but is effectively faded out of circuit towards either end, thus adding another variable element to the frequency response contouring.

Used with a reasonably bright audio input, the perceived loudness is then similar for all settings, so a small turn of this single control can be used to obtain a distinct change in timbre without having to readjust the volume, while a back and forth rotation during a sound gives an effect like a manually swept one-notch phaser. A by-pass switch is not needed since with RV1 fully anticlockwise, the tonal response is almost identical to a 'straight through' path.

For best results this circuit must: (1) feed into an impedance of 47k or over, for example, a non inverting op-amp, which if given a gain of x5 will also more than make up for the insertion loss of the passive circuit. (2) be driven from a low impedance stage, designated 'A' in the diagram. This may already exist within a piece of equipment (emitter follower, op-amp, etc) in which case the present circuit can simply be added on passively. If the available output impedance is more than a few hundred ohms, however, a simple buffer stage should be included.