Home   Projects   For Sale   Contact     News  
     

 


 

Amplifier sound character

 


Vintage sound

Many people have a romantic idea of vintage audio sound as all warm and fuzzy, soft deep bass with a smooth lower midrange and a slight roll-off at the top end of the treble - much like how classic valves often sound. In reality, the sound of vintage audio gear varies greatly, depending on brand, relative price and era.

For example, Technics tend to have a very heavy, almost 'glassy' sound with strong mid bass and projected treble and a lean midrange, this trend gets stronger after about 1980 as ICs became more widely used by them. Quad are very understated in a typical English way - pure, uncoloured, almost 'soft' sounding, imaging can be superb - especially on vocals - but the overall sound might be too reserved for some people. Pioneer vary more than most (perhaps I've just heard too many), but most have a deep sound with strong midrange and good bass dynamics - the higher end receivers sounding 'warmer' than their integrated amps, the earlier amps in particular sounding brighter than the later ones or the receivers.

Also, these days people will often run vintage equipment on either modern speakers, modern music formats, or both - vintage speakers and older music have a completely different sound to anything modern.

 

Brightness

Vintage audio gear (and even a lot of modern transistor audio gear) gets a reputation for being 'bright', I don't know whether I just have exceptionally sensitive hearing but I find a lot of lower end modern audio equipment literally painful to listen to because of this.

Many people perceive this 'brightness' as being an issue in the treble region - actually, what I perceive as 'brightness' is rooted in the midrange, usually around 2-4kHz. I tried using mild notch filters to remove this brightness but then you're removing portions of the 'presence' frequency band of the music.

Resonances play their part, but sometimes this brightness is present in the original recording and some equipment and room acoustic combinations just brings it out more than others - load a portion of the offending song into a sound editor like Audacity and plot a spectrum graph, most likely you'll see peaks in the midrange region. Try applying a mild notch filter in this region and then replay the sample and it will probably sound much softer and easier on your ears at higher volumes.

Some people blame modern metal tweeter drivers, but these aren't solely the culprit for brightness - having soft dome tweeters may make the sound more 'warm' and less fatiguing to some people but will just nudge any apparent brightness firmly into the midrange. Metal tweeters are there to project dynamics and 'sparkle' in a way that a soft domed tweeter just can't, they do that well - but inevitably that will give the sound a different character that people may, or may not, like.

 

Dynamic range and compression

Most 'better' vintage audio equipment used discrete transistors in the low level signal stages - while not as 'perfect' as modern op-amps, a discrete transistor circuit can take a much higher supply voltage (on vintage Pioneers, often somewhere between +/-25v to +/-40v opposed to around +/-15v for most op-amps). This higher supply voltage means that the voltage output on peaks can swing much higher in relation to the rest of the sound signal, meaning better dynamic range and 'headroom'. This all depends on the magnitude of the input signal and the gain of the amp though, so you can't just assume that an amp stage with a +/-30v supply will sound better than one with a +/-15v one.

The same goes for power amplifiers - to a trained ear, an amplifier which is being pushed hard and running out of power reserve will noticeably compress the sound long before it's driven into clipping (clipping = audible distortion). This is the reason that a great deal of higher end audio equipment has high power ratings, it has more to do with dynamics and compression than it has with outright loudness. Hence the reason why a good high powered amplifier will always sound more dynamic than a smaller one, even at low volume levels - there's more room for the peaks of the signal to expand without being compressed, so the resulting sound is more dynamic and powerful.

This is the same as the compression which you'll get with cheaper loudspeakers - when asked to provide higher sound pressure levels (SPLs), a loudspeaker which has a lower capability will compress the peaks of the resulting sound, partly because of the mechanics of the suspension and cone but mainly because the voice coil heats up and increases in resistance during musical peaks, which means less force gets to the cone - resistivity of conductive materials such as copper increases in proportion with temperature, and higher resistance means less current passed and hence less power and less SPL.

Magnetics also play a part - this is one reason why more expensive loudspeaker drivers tend to be heavier and have larger magnets than cheaper ones (although control of resonant frequencies also plays a large part in the need for large magnets).

On power amplifiers (and, to an extent for signal stages), the two factors which are the most important for good dynamics at high power levels are the voltage and current capability of the power supply.

 

Input levels

This really leads on from the section above. Back in my early days in audio, I used to wonder why some companies sold 'input attenuators' which would sit inline with the audio outputs of a CD player etc and reduce the volume level of signal getting into the pre-amp - surely, if a CD player could output a higher signal than most (vintage) audio gear was designed for, that was a good thing for noise levels?

Most modern CD players will output around a volt of signal, many pre-amps (especially older ones designed before the CD era), will be designed for half of this or under. Putting a really high signal into an amplifier doesn't necessarily mean you'll get distortion, but it will have an effect on the dynamics and linearity of the sound output.

 

Effect of components on sound

Yes, components do have their own sound. At the extreme, you have op-amps which contain complete amplifier circuits complete with dozens of miniature transistors etched into their silicon, many IC devices such as op-amps are designed to be 'modular' - for example, the ubiquitous 4558 dual op-amp used in huge amounts of audio gear could often be directly replaced by many more modern, higher quality chips with little or no modification - with the result of much better sound quality, especially on high gain circuits such as phono stages.

Discrete semiconductors such as transistors also have an effect on sound - capacitances, resonances and linearity of a device (well, its linearity at the point it's being operated at) all factor in here. When you get to small signal transistors, things such as noise levels become important (think 'white' or 'thermal' noise), a modern audio grade transistor will easily outperform a vintage one, although the difference that will make in a well-designed circuit might not be anywhere near as large as you'd think.

Then you get passive components. Capacitors in audio equipment are of varying types, depending on their size, the job that they're meant to do, and their cost. Many capacitors in transistor audio gear will be electrolytic (for more information on electrolytics, see this page) - these capacitors are, by their nature, imperfect, but a necessary evil when you consider their size advantage over 'better' types. Electrolytics vary widely, with frequency related changes in inductance, ESR and overall differences in value altering their performance and hence the response of the circuit that they're part of. The golden rules with electrolytics are to use good quality ones which are suited for the job you want them to do, make sure they don't get reverse biased, keep well below their voltage rating, keep them cool, and finally aim to renew them every 10-15 years.

Inductors don't see much use in audio gear outside of tuners, speaker output networks and crossovers, so don't really factor in here. Some people talk about DC resistance, but that might well have been factored in by whoever designed the circuit in the first place. Personally, I leave them alone unless I have a really good reason to do otherwise.

Resistors in most audio gear tend to be pretty good, purists talk about certain types of resistor colouring the sound of amplifiers and I wouldn't argue against this, but as long as they're of good quality, they're not a major issue. Wire wound resistors might be another issue because of inductance in the coil, but you're only going to meet those inside power supplies, amp output stages and loudspeaker crossovers, they don't really have much use elsewhere - most resistors used in small signal circuits are going to be metal or carbon film, hopefully with good tolerances and low temperature coefficients.

 

Go back to previous page

 


All photographs and text on this website unless otherwise stated are © 2008-2014

Briarsfield Hi-Fi - www.briarsfieldhifi.co.uk