Available in print in Common Ground Issue 4
During the course of September I was fortunate enough to experience two completely different events. When they occurred within a couple of weeks of each other they made me step back and take a fresh look at what I am trying to achieve with my audio system, and more importantly why.
The first event was the annual hi-fi show at Heathrow Airport. Whilst it should be an occasion to evaluate the best products for bringing music alive at home, past experience has taught that it is almost always a sad disappointment. This year didn't look too promising. The show sprawled across two hotels, and with typically dreadful internal room acoustics the various companies at the show made a fist of displaying a wide variety of music systems, musical styles, technical approaches, and aesthetic statements, as to what makes a great hi-fi system.
During a long day meandering around the corridors, I found it to be unusually quiet for the opening day, and was able to have plenty of time to sit and listen to the wide array of equipment on display. Transistors, valves, vacuum transistors, push-pull, single ended, horns, panels; you name it, it was there somewhere. There appeared to have been two notable movements within the past twelve months: Firstly the stranglehold of the Single Ended valve amp, witnessed over the past two years, is definitely on the decline. There appears to a big swing back to transistors (albeit some in SE operation) and push pull valves. Secondly there was a big return for the "big" sound of a few years ago. Many systems appeared to major on scale and volume, whilst completely losing the plot on clarity and detail.
The second major event was somewhat less predictable, and certainly had nothing to do with audio. At the end of September I was out walking in the Lake District with a few friends on a very breezy day with plenty of passing showers. Whilst I am not normally a big fan of getting soaking wet, we had been pretty fortunate to avoid the majority of the rain. Whilst traversing the ridge route between two of the peaks, we were treated to a fantastic display of nature at her finest. Autumnal colours in the valley, with sunlight throwing a glorious glow on the lake, whilst being chased away by a rapidly advancing rain shower. The shear range of the colours, and dynamics of the weather had a grip over all of us.
What made it singularly spectacular was the physical reality, scale and authority of the overall scene, balanced by the finesse and beauty of the individual elements as they each mapped out their own scene within the whole.
This combination of shear physicality combined with effortless clarity started me thinking about the emotional impact of music, and how any music system with any hope of truly capturing the attention of it's listener must combine these two opposing requirements. Of all the music systems I have heard over the years, very few manage both acts. Very many of the large expensive systems can manage the physical scale, but at the expense of finesse, and true dynamic clarity. Very many of the smaller systems have an amazing ability to retain detail, but are wholly unable to truly portray the required physical size of the performance.
This thought process allowed me to think back over the systems I had listened to a couple of weeks earlier, and interpret my reactions in a new way.
Whilst I was at the show, I had clearly identified two systems which gave the best approach to the ideal combination of scale, with clarity, and emotional involvement. These were the Carfraehorn system, and head and shoulders above the rest was the Living Voice system. What really set these two apart was the effortless nature of the presentation. The way that big, solid, clean images appeared before you without any sense of mechanical intervention, and with a speed and natural timbre, which belied the artificial nature of reproduced music.
There were three other systems capable of the physical presence, Wilson/Krell, B&W, and JM Labs. However all three were compromised by having either poor emotional communication, timbre, or finesse.
I hope that the start to this article hasn't put to many of you off, but I feel it is important that we should all question and re-assess what we want from a music system from time to time. Only then will we know what direction we need to travel in.
In my Power amp Arcticle (Also in Common Ground Issue 1) I reported on my design experiences whilst working on a couple of very different power amplifiers. The second of these had a low input sensitivity of over 1Vrms to full output, and I glibly suggested that a good line stage preamplifier would be able to cope with this.
Experience over the past few months has shown that good line stage preamplifiers are not as common as I had hoped. Several customers with either commercial or homebrew preamps had built the two stage transformer coupled power amps, and discovered that the sound quality wasn't as high as they had hoped. In all these cases we eventually tracked the problems down to the preamps, and with a little work helped them overcome it. With this in mind I thought it appropriate to put together a line stage preamplifier for this article.
The second reason is that if you want to experiment with building a valve amplifier, you will not get any simpler than a line stage preamplifier, especially if you can live without an input selector switch.
Rather than coming up with a finished design, and proclaiming it to be the answer to all your prayers (provided you purchase all the parts from me, obviously!) I would much rather set out a very simple project for you all to have a play with. Unlike so many other self promoting egotist I do not: (1) insist that you buy the parts from me, (2) claim there is only one "true" way to build it, (3) offer a kit to remove money from you, (4) suggest it is the only true path to audio nirvana.
What I want to offer is a very simple, single valve, no feedback line stage preamplifier design. Within this basic design there is a number of different valves you can experiment with, and a number of different ways of putting it together. I am hoping it will be an easy to build base, for you to actually learn what you like, why you like it, and which of these methods most closely approaches your goals for music reproduction.
The general idea is to build a basic line stage preamplifier that allows for a number of different designs and valves to be experimented with. Initially it should be built with good commercial components, nothing fancy. The aim is to allow you, with a little money and some time, to clearly demonstrate to yourself that circuit topology and valve choice have a marked and significant effect upon sound quality.
In my opinion the circuit/valve choice is the most important choice. You are then able to best exploit and refine this basic design with top quality components, but you cannot turn a pig's ear of a circuit into a silk purse by throwing expensive components at it. To my mind there are far too many articles setting out ways of applying tweak components, and not enough encouraging you to try out different circuit topologies.
The approach I would recommend is this:
1) Case: Maplin aluminium chassis, buy the largest it's only about £8. Use Qmax punches for the large holes for the valve bases.
2) Components: purchase a collection of the required components from Maplins or Electromail, and choose good commercial components: e.g. metal film resistors, cheap film/foil capacitors, low esr electrolytic capacitors etc. These will generally be significantly cheaper than any of the "audiophile" parts in the more specialist catalogues, so you won't care much if you never need them, or decide that design isn't for you.
3) Cheap hook-up wire, or PTFE silver plated copper wire (25m reels available from Electromail)
4) Sockets and switches: reasonable inexpensive items
5) Volume pot: Alps Blue, not very expensive, good sounding pot. There are much better, but most £1000 commercial products are lucky if they get an Alps Blue, so it's more than good enough for the time being.
6) Valves: get all the valves together from one place if possible. Buy four of each type so you can get a reasonable match on gain between the two channels.
7) Valve bases: chassis mount, buy three B9A, two B7 and two octals
8) Mains transformer: you can purchase this from SJS, but in the words of the BBC "other mains transformer suppliers are available".
A good many of you have almost certainly tried this one at home. Using anything from a cheap open frame potentiometer, to some amazing mega quality stepped attenuator with resistors out of satellites. Hands up! Who hasn't tried it?
The basis approach is shown in Figure 1. Obvious advantage, no circuit, therefore no distortion! This leads to a sweet, clean, nice presentation of music which is normally a true revelation next to the mud bucket, poor components, feedback strewn preamp you used to be using. Correct?
Unfortunately there tends to be a high output impedance and high capacitance. The potentiometer tends to be placed by the cd player, and yards from the power amp. In valve systems the pot may be 100K ohms in value. Lastly, and most crucially, there is nothing to drive the power amp! There is no gain! These factors all go to ensure that the passive preamp always sounds lifeless, flat, overblown and dynamically lacking. Always!
The general layout of the active line stage is shown in Figure 2. Here you have a single valve, without feedback attached to the output of the potentiometer from Figure 1. "Surely this must be worse" I hear you cry. What it does address are the negative aspects of the passive. It has gain, ability to drive the power amp, lower output impedance, and significantly better scale and dynamic contrast.
On the down side, it may have too much gain, it may suffer from noisy valves, it may slightly reduce clarity.
Shown in Figure 3, it's bit of a dog really. It does sort out the first two problems with active line stages. Unfortunately it introduces all the problems of passives.
It is included for completeness, so you can decide for yourself.
|A, G & K are Anode, Grid and Cathode for each active device|
|A1, A2 etc indicates there are sometimes 2 active devices in a valve|
|H is the Heater connection. Hct in the centre of a 12.6v heater. A 12.6v heater may be used at 6.3v by connecting H and H together and connecting 6.3v across H and Hct.|
|NC means there is no connection|
|SC means it is an internal screen, which may be earthed if desired|
|EZ80 is the rectifier valve used in the power supply, not an audio triode|
|For 280V supply use 330-0-330V taps on mains transformer, for 180V supply use 230-0-230V (see Figure 4)|
|See Figure 5 for valves base info|
|Rp to be 5W wirewound resistors|
|Other resistors to be 1 watt|
I will be concentrating on different ways to build Figure 2. Table 1 shows a number of different valves, with appropriate connection details, and circuit values, for you to try out. Note: that where possible I have chosen 6.3V valves, so you only need to build one filament supply, and valves with compatible pin configurations have been grouped together.
The valves grouped together in Table 1 have different gains. The line stage will thus offer a different level of gain with each valve. This should not be a major problem in most cases. There are, however, a few potential problems, which should be noted.
Firstly, some of the valves, particularly the 417A and 6GK5, have rather high input capacitance. This means that with a potentiometer over 20K ohms the grid circuit impedance will be high enough to cause noticeable high frequency loss. The way of avoiding this is to use a 10K or 20K ohm pot.
Secondly the same valves may suffer from some grid current i.e. current flowing in the grid circuit. This appears as a rustling sound when the pot is moved. The best way around this is to site a capacitor between the pot and the grid.
Thirdly, if the circuit does have too much gain, the easiest way of reducing it is to site a resistor between the input and the top of the potentiometer (as shown in Figure 2).
Lastly, if instability occurs when the pot is on minimum volume (a buzzing noise) then the 100R grid resistor should be increased until it no longer occurs. (There is no hard and fast rule for this as it varies from unit to unit).
Once you have carried out your experiments, and you have chosen the circuit you want to spend more time with, you can now go on to really wring the most out of it.
There is one trick with the circuit worth playing, if it is appropriate. If your chosen circuit has a bias voltage (Vk) of about 1.5V you could try replacing the Rk and Ck in the cathode circuit with a small nicad battery. Or you could use a Duracell in the grid circuit to give -1.5V on the grid, and ground the cathode. (You will need a capacitor between the pot and the battery if you try this.)
These two methods effectively fixed bias the valve! It works for power tubes, and certainly works here as well.
|You can also start throwing some high quality "audiophile" components at it to bring about improvements. My suggestions would be:|
|Hovland Musicaps for coupling|
|OSCon for Ck|
|Rhopoint precision wire wound for Rk|
|Mills or Dale wirewound for Rp|
|Holco or Vishay for the other resitors|
|Panasonic pot, or switched resistor network for volume control|
|Replace all the power supply dropper resistors with inductors|
I will not set out a long discussion on what I feel the relative sonic merits are of the above valves, as I feel this would prejudice your results. I will say that you will observe quite a large range of sonic characteristics, and it may well truly surprise a number of you.
What I would like to hope is that the ideas put together here will help you to spend a few winter evenings in a productive learning experience, and you will benefit from having a more personal and better sounding system at the end of it.
The most important thing to remember is that there isn't really a right or wrong version to go for, it is all personal preference, system synergy and musical taste. The one massive advantage with this preamp is that you get to choose it, without spending too much money.
I am more than willing to enter into discussions with people who do try it out, and am very interested to receive feedback about your efforts. I would also like to collate peoples' experiences to publish them on the web site. If this interest you, then please get in touch.
These HTML documents, and associated graphics, are Copyright 1999, SJS Electroacoustics. These documents may not be reproduced in any manner without the permission of the copyright owner. The schematics presented here are Copyright 1999, SJS Electroacoustics and are offered for construction on a non-profit basis only. Any persons wishing to profit from these designs, or requiring similar should contact SJS Electroacoustics for licensing. We invite the audio and music community to link to this home page, which will be periodically revised.
This document has been revised and updated April 4, 1999.