Valve preamplifier won't have "valve" sound

Early valve amplifiers did produce a significant amount of distortion, as did transistor amplifiers of a similar vintage. But it is the type of distortion that these valve amplifiers produced that made them sound better.

Simple valve amplifiers introduce even order harmonics so the distortion is similar to a musical chord. On the contrary, simple transistor amplifiers produce even and odd order harmonics. The odd order harmonics were introduced by the simple class B power section of these amplifiers. A human brain is good at noticing small amounts of odd-order distortion generated by transistor amplifiers but it will ignore even order distortion generated by valve amplifiers.

So let's forget about the history and think about why somebody would want to add a valve preamplifier to their relatively non-distorting transistor amplifier. One of the main reasons is to experience the added effect of even order harmonics. Notice that I called it an "effect" rather than distortion. In particular, it must be noted that valve guitar amplifiers were around before transistor amplifiers ever existed. The sound with this added effect is the sound expected from an electric guitar!

The valve amplifier you published in the November 2003 issue employs negative feedback to remove some of the very effect that you were trying to generate. If you look at the Vintage Radio section of your magazine, you will find many examples of valve amplifier stages that employ little or no negative feedback. There is even one example of a stage with no negative feedback in the Vintage Radio section of the same issue that you published the valve preamplifier in!

Our own valve preamplifier (K188) is based on a single subminiature pentode with flying leads. The pentode has a maximum plate voltage of 30V and needs 1.2V at 10mA to power the filament. The gain is adjustable from 0 to 4 times and the output impedance is about 10kW and the frequency response extends from 6Hz to 600kHz; perfect for inserting in-line with an amplifier system to simulate valve sound. Best of all, since no negative feedback was used you would get to experience true valve sound, as it used to be.

Branko Justic,
Oatley Electronics.

Comment: while it is true that virtually all the circuits featured in Vintage Radio did not employ negative feedback, most good quality amplifiers produced at the end of the valve era did feature negative feedback, particularly in push-pull power amplifiers, and that includes virtually all guitar amplifiers.

Sadly, no simple valve circuit, whether it features negative feedback or not, can hope to simulate the overall sound of a valve guitar amplifier, particularly when it is driven into overload. The controversy rolls on!

Developing the
valve theme

Well, you said you would never do it but you did. Congratulations! As a keen hifi advocate and electronic DIYer, I believe that for a magazine such as SILICON CHIP this is a great opportunity to open up and maintain a new following of persons interested in hifi/guitar and also valves. As an instance, take a look at http://tubesall.hihome.com/tube.htm which gives some idea of the interest in valve-based audio equipment, particularly for DIYers.

Here are a couple of ideas for development.

(1). Take your existing valve preamp design and the principles of your recent guitar effects unit to provide a guitar preamplifier based on valves that provides pre-distortion (that guitar players love) and combine this with your SC480 50W power amplifier to provide a valve sound with solid-state reliability in the output stage.

(2). Develop your existing valve preamp design into a stereo RIAA equalised preamp for magnetic cartridges. Recent developments in valve quality, component tolerance and noise figures allow a very respectable amplifier to be built. Your existing power supply, properly screened, could be utilised and you already have DC for the heaters.

I use three 12AX7s (ECC83) in my preamp and I am very happy with the performance/background noise compared to solid state designs. And it sounds better!

(3). A valve power amplifier. Jaycar are selling EL34s (6CA7s) which are good for about 30W in push-pull mode and the 12AX7 could be an amplifier/phase splitter (OK, you may struggle with gain with only one per channel).

Of course, the output and power supply transformers may be the "killer" of this idea - both the availability and the cost - which you note in your valve preamp article but perhaps there are some possibilities with existing manufacturers if there was enough interest from your readers.

Looking forward to more "tubes" in SILICON CHIP.

Dean Brookes,
via email.

Valve days long since gone

Paul Rohde (Mailbag, January 2004) seems to have worked himself into a bit of a frenzy over your "valve philosophy", if there is such a thing.

Sound quality is a very elusive animal and, in spite of many advances, there is still not much really good quality sound available, due mainly to the difficulty and expense of producing really good loudspeakers. Since the advent of CDs, there is no technical reason for recorded sound not to be nigh on perfect although many sound recordists or the acoustics of the recording studio still manage to wreak havoc on the final result.

When transistors first became available, I was rather suspicious of these new-fangled devices which then had many limitations and, as I thought at the time, could never replace valves. However, when I built my first complete transistor amplifier in about 1968, I could not believe the absence of intermodulation distortion and the overall goodness of the sound. I might add that the valve amplifier I had been using was one of the better designs with a claimed distortion of 0.1%. I would never now consider the use of valve amplification for one moment.

As you so rightly say, they "were great in the past (when there was no other choice) but their day is long since gone".

Loudspeakers still influence the final sound quality far more than source and amplifier and even very expensive speakers can be a disappointment. Most have colouration and most dealers seem to think that is what the buyers want. The attraction to valves is a bit like the attraction to "quack" medicines and other such fads. People are always attracted to off-beat solutions even though there is no cogent evidence of any advantage over scientifically proven procedures.

I think your philosophy on valve amplifiers is quite clear and correct but you have a magazine to produce and you must consider the requirements of your readers no matter how misguided some of them may be!

Alan March,
via email.

Miller effect
was a problem

Well, well! I thought that curiosity would finally persuade your design team, even if nothing else did, to have a go at valves! I have been making projects since the mid-1950s and I have to say that "FETs with lamps inside" haven't entered the scheme of things for me since David Tillbrook produced his brilliant amplifier design using Mosfets in the early 1980s.

On a more serious note, I recall that "Radio and Hobbies" struggled with the problems surrounding "Miller Effect" in triode stages, with some of their control units in the "Playmaster" series of the 1950s. This I think, led them to concentrate on the EF86 pentode (which had a "coiled coil" filament for hum reduction), where gain and equalisation was the aim. This valve also had its problems, noise being the main one. In a low-noise application, one sometimes had to sample several valves before satisfaction was achieved.

Where impedance conversion rath-
er than pure stage gain was the aim, "Radio and Hobbies" chose on at least one occasion (I think it was a control unit for crystal and ceramic pickups in the early days of domestic stereo) to use a 12AU7 twin triode. "Miller Effect" with high input impedances was a lesser problem with this valve, although I believe the stage gain achievable was much less than with the 12AX7 or the 12AT7.

Bruce Bowman,
via email.

Comment: the biggest problem with EF86 valves was their tendency to become microphonic, after which the slightest tap would make them "sing".

Multi-element TV
antennas can be fakes

It was interesting to read about your Penrith reader's experiences with hail-damaged TV antennas in Mailbag in the December 2003 issue. I live in Riverstone which probably copped the worst of that same storm. My backyard went from English Country Garden to Arctic Wilderness in about five minutes! I had a similar antenna arrangement, with a VHF/UHF job pointed at Artarmon and a 96-element UHF antenna pointed at Woolongong, but I didn't get any sort of picture afterwards; not surprising with the masthead amplifier lying on the ground in four pieces and buried under six inches of solid ice!

The UHF antenna was reduced to a single piece of aluminium box section sitting forlornly at the top of the pole, and the VHF antenna looked like someone had attacked it with a meat cleaver.

Well, a new antenna system, complete with a brand-new masthead amplifier has just been installed and works a treat. I've now got the most common replacement out here: a Band III VHF antenna pointed at Artarmon and the ABC and SBS on UHF from Wollongong.

VHF I ABC reception has always been marginal at best out here, with a ghost from the Blue Mountains and violent Doppler flutter when a big Hercules comes in for a landing at the Richmond Air Force Base. Plus I no longer have those huge rear elements that seem to be so attractive to big birds!

The thing that's intrigued me is that, while as far as I can see the new 96-element UHF antenna looks pretty much like my old one, this one pulls in the Wollongong UHF channels like you wouldn't believe! Running the signal through my household 8-outlet system without the amplifier still gives a pretty passable picture. The other antenna would give a barely visible picture under the same conditions.

I actually asked the guy fixing a neighbour's antenna about this and he confirmed what your Penrith correspondent suspected: there are antennas that have been carefully engineered using strict scientific principles which perform their tasks as well as is possible for a structure of those dimensions. There are also antennas which have been "reverse engineered" (ie, copied) from a working design (how accurately being anybody's guess), and then there are "counterfeit" antennas which are simply rubbish, basically flung together to resemble a proper antenna. They do work, sort of, but so does an ordinary piece of wire under good conditions.

The stupid thing is that in many cases the same materials could have been used to make an antenna that really did work!

Keith Walters,
via email.

Hybrid computers not quite dead

Rod Cripps, "Mailbag" December 2003, was interested in hearing from people with information on surviving examples of analog/hybrid computers. Well, the technology might not be as dead as he thinks.

A series of construction articles to build a hybrid computer ran in "Everyday Practical Electronics Online" towards the end of 2002. Currently, the December 2002 issue is offered as a free sample and a PDF version can be downloaded from the EPE Online website at www.epemag.com.

This issue contains Pt.2 in the construction series, which describes the "programming" of the computer together with using it with PC-based software to do calculations for a flight simulator.

Paul Gittings,
Russell Lea, NSW.

Running SC480 modules at higher voltage

The letters about the SC480 amplifier (from R. C. and J. W.) in the November issue caught my attention. I have been using three of the 1987 modules for some time with a 30V-0-30V transformer without problems. The DC voltage to the modules is ±43.2V. It is true that the BC639s and BC640 run hot but only at 50-60°C and this appears to be within the operating specs of the transistors (ON data sheets).

Do you think I have a disaster waiting to happen here? RC's transformer must be unusual if the rectified DC voltage is 47V.

Just for your information, two of the 1987 modules were built using MJ802/MJ4502 output pairs (since I had these available) and they work perfectly without any other modifications and (to my ears) sound better than when using the 3055/2955 pairs.

Incidentally, concerning the valve preamp, I have used two 240VAC mains transformers back-to-back to generate HT voltages for valves. This maintains isolation from the mains and need not be too expensive since the current involved is usually low. If the input transformer has dual secondary windings, one winding can be used for the heater supply. Of course, size and space may be a problem.

David Allen,
Aspley, Qld.

Comment: your amplifier modules should be OK with the higher voltage, especially as you are using the MJ802, etc. We would expect them to sound better too.

Solar cells not viable

I want to comment on the editorial regarding solar or gas fired power stations in the December 2003 issue. Firstly, I think you may have glossed over the issues of solar power somewhat. For years, conventional solar power (using solar cells) has seemed to be a reasonably viable way of generating electricity that is friendly to the environment, however a lot of people seem to only concentrate on how much power they can output and forget about how much power (and pollution) it actually costs to make them.

Unless a solar cell is capable of generating more energy than the manufacturing process takes, then there is little point in using them other than to make people buy them because they believe they are helping the environment or saving money. Also there is the issue of disposal, given that in my experience solar cells only work efficiently for about 10 years or so and eventually degrade due to the effects of the weather (sun, rain, hail, etc).

I can't help but think that as a country we have chosen the certainty of environmental damage through greenhouse-gas emissions for electricity generation as opposed to taking the route of nuclear generation. Whilst I will admit that yes nuclear accidents are possible, I still believe that it is also possible to reduce the risks so as to be almost negligible.

Anyway, it is good to see that the Editorials still keep touching on fairly controversial topics. Thanks for a great magazine, keep it up.

David Peters,
Bathurst, NSW.

Comment: the subject of solar payback has been well researched. Have a look at www.ecotopia.com/apollo2/pvepbtcsi.htm. There are plenty of other sites which give similar information.

We really don't think nuclear power will ever be a viable or ethical solution. The problems of decommissioning power stations and reprocessing or long-term disposal of fuel rods is extremely difficult. Nuclear power stations also cause far more "thermal" pollution than coal fired stations because they must run with lower steam temperatures - they are simply nowhere near as efficient.

Share this Article:

Tweet