Linsley Hood simple class A amplifier designs (1969, 1996)

No other design met has been the subject of as much discussion, debate and praise, not only for it's perceived sound quality, but it's simplicity and succinctness. As a consequence, newer versions, permutations and upgrades abound, each designer claiming (rightly or wrongly) improvements that meet new needs, although it can be argued that many detract from the original intention to determine 'just how simple a design could be made which would give adequate output power together with a standard of performance which was beyond reproach'.

10W JLH class A design, less supply. Flat from 40Hz to 95kHz, >20Hz - 200kHz (-3dB). Power bandwidth (10W) 30Hz - 70kHz. THD = 0.05% (50Hz - 20kHz @ 9W, 0.1% @ 1kHz and 10W). Quiescent current = 1A. Full and comprehensive article appeared in Wireless World, April 1969.

Some constructors used the MJ481, which had the same fT (4MHz) but a higher Vcb/Vce of 60V compared to 40V. Modular preamp for above.

John Linsley Hood's comments on transistor selection for his 1969 JLH Class A (Wireless World, April 1969)

"Some experiments were made to determine the extent to which the circuit performance was influenced by the type and current gain of the transistors used. As expected the best performance was obtained when high-gain transistors were used, and when the output stage used a matched pair. No adequate substitute is known for the 2N697 / 2N1613 type used in the driver stage, but examples of this transistor type from three different manufacturers were used with apparently identical results. Similarly, the use of alternative types of input transistor produced no apparent performance change, and the Texas Instruments 2N4058 is fully interchangeable with the Motorola 2N3906 used in the prototype.

The most noteworthy performance changes were found in the current gain characteristics of the output transistor pair, and for the lowest possible distortion with any pair, the voltage at the point from which the loudspeaker is fed should be adjusted so that it is within 0.25 volt of half the supply line potential.

The other results are summarized in Table 2. The transistors used in these experiments were Motorola MJ480 / 481, with the exception of (6), in which Texas 2S034 devices were tried. The main conclusion which can be drawn from this is that the type of transistor used may not be very important, but that if there are differences in the current gains of the output transistors, it is necessary that the device with the higher gain shall be used in the position of Tr1.

When distortion components were found prior to the onset of waveform clipping, these were almost wholly due to the presence of second harmonics."

Further comments on the later re-design shown below (Electronics World, September 1996)

"As I commented, at the time, the design gave a somewhat lower distortion if the hFE of Tr1 was greater than that of Tr2. This caused the output circuit to act as an amplifier with an active collector load rather than an output emitter follower with an active emitter load.

A simple modification which takes advantage of this effect is the use of a Darlington transistor such as an MJ3001 for Tr1. At 1kHz, this reduces the distortion level at just below the onset of clipping from about 0.1% down to nearer 0.01%. As before, the residual distortion is almost exclusively second harmonic. Also, as before, it fades away into the general noise background of the measurement system as the output power is reduced."

Re-design, September '96. 15W (sine) into 8 ohms, freq response = 7Hz - 50kHz (-3dB). Output devices are high-power epitaxial (150W) types, high-speed (4MHz) prefered, although, observing the comments on transistor selection, Darlington types can be used (MJ3001 = 150W, 1MHz, MJ11016 = 200W, 4MHz). 4k7 pot sets quiescent and the 22k pot sets the output DC. Quiescent current = 2A (88W quiescent dissipation per channel).

Alan Jackson offers PCBs for this project at apj audio.

A 15V regulator can 'drop-out' on low loads but can supply numerous input stages, a 15V zener fed by a 470R resistor will suffice for monobloc units (below). Some 78x/79x series regulators can be very noisy compared to adjustable types like the LM317/37 types. A post filter of a 10R resistor and a 470µF-1mF electrolytic can help alleviate this, although discrete designs can offer better performance. Additional caps can improve, doubling or more the value of the input (0.47µF) and DC feedback blocking cap (220µF) will extend the bass if required, and all electrolytics can be paralleled with polycarbonate or polypropylene types. Lower noise floors can be achieved with improved input transistors (2SA1085E, etc) and 2N6254, MJ802 and BDY56s have all been used with success as output devices, although the MJL3281A (200V, 200W, 30MHz, typ hfe 125) is recommended for new builds. Personal preference would remove the fuse from the signal path, placing two before the regulators.

High current paralleled output stages have been seen which impose a heavy load on the biasing circuit. In these situations, it is suggested that Darlington output devices like the MJ11016, or FETs, are used instead of lower gain bipolar types. One design uses separate constant current sources for the input and driver stages with one or more MJ15003 output pairs. Other variants include;

   Nelson Pass' PLH amplifier which explores the removal of the input transistor, reducing the overall feedback and replacing the bipolar devices with MOSFET types.

   Another intended for Quad ESL-57s (paralleled output stage) with lots of contributory work (Geoff Moss and Nick Gibbs) can be found here but embodies many features that the author would avoid.

   Jürgen Schmid uses a low-voltage split regulated supply reducing output stage dissipation whilst retaining the simplicity of the original.

Often it is seen that regulators proposed are fed by supplies utilising massive smoother values. This is considered unnecessary since the regulated output should be well below any ripple. Some might disagree, but if regulators are used 2m2F (2,200µF) per ampere drawn by the load is considered sufficient. Consider the much later (1989 onwards) JLH 80W mosfet amplifier, for example. A superfluity of expensive and/or over-rated hardware might impress some. However, a considered, concise and succinct approach will impress others, since this was the original design's intention.

Valves versus Transistors

Contact me at
especially if you want additional content to this page
or if you find any links that don't work. Don't forget
to add the page title or URL. Take care!

Google search for John Linsley Hood.

Some more JLH designs.
Back to index, sound, tips or home.