Valve amplifier lobbying

Your editorial in the November 2006 issue and the elegant plea by Roger Lowry for a definitive valve amplifier circuit has prompted me to write to you. I have for the last 12 months or so trawled the internet looking for a valve amplifier schematic as well as a source of components.

While there is a wealth of information out there, some of it can be downright misleading. I purchased your DVD-ROM containing all the RTV&H articles just so that I could evaluate the September 1955 Playmaster design as it is, after all, a derivative of the classic Mullard 5-20. I have come very close to building one after ascertaining that all transformers are available locally at "reasonable pricing" (Evatco Hammond transformers).

My interest in electronics goes back to the mid 60s when I started purchasing Electronics Australia. I see that in July 1967, Leo Simpson wrote the article on the Playmaster 118 valve amplifier project, so I would bow to your experience and can understand your reluctance to go back to the "bad old days of valves" when you can improve the on-paper specifications so much more using transistors.

I also would ask that you use the Mullard 5-20 as the basis of a new design. It could be a "mono block" design, allowing readers to build one unit at a time. It should use readily available EL34s and I would agree with Roger Lowry that we should replace the EF86 input with a 12AX7.

Leo, I think that your quest for perfection in all things electronic may be influencing the type of amplifier the average reader would want to or could afford to build.

Let’s put a peg in the ground and start with a more modest 20-30W per channel with an updated "5-20"; perhaps it should be a "4-20" as we would not require a valve rectifier.

Andrew Prest,
via email.

Nested feedback plus automatic bias I would like to congratulate you guys for even considering a valve amplifier project and I do agree that the cost can be very high (transformers, etc). I am glad Roger Lowry (Mailbag, November 2006) did request one though and I agree with his recommendations in the specifications. I have been considering building such an amplifier for a long time as well. Such an amplifier would certainly require lots of nested negative feedback with perhaps push-pull (parallel) EL34s (6CA7) in the output stage, a 12AU7 phase-splitter and 12AX7 voltage amplifiers, plus (due to availability) toroidal transformers, etc. It would also have to be aesthetically pleasing as well. This amplifier would also require some form of automatic bias control to compensate for valve aging. Such a design would appeal to many sensible constructors as the radical "non-feedback golden-eared fringe" has coloured many peoples’ opinions, to the detriment of valve amplifiers in general. Greg Johnson, Cooma, NSW. Is .01% THD too low? Thanks for airing your thoughts Leo, in the Publisher’s Letter in the November 2006 issue, on the design of a valve amplifier. I can see your point that there is little value in putting in a lot of time and effort into something that would possibly not have a lot of appeal and your request for a show of hands to see if it is worthwhile is a very good idea. I have made a few amplifiers over the past six years, my first being the amplifier designed by Mr Tean Tan in Electronics Australia in the Sept/Oct 1992 issues. I and others who have listened to it are impressed with its performance. This design is very quiet with just a little hiss, only noticeable with your ear up against the loudspeaker. My second effort was the Playmaster No.5, described in Radio TV & Hobbies in June 1952, using 2A3 valves. Those 2A3 valves look very impressive. This amplifier also sounds very good but is a little noisier than the previous unit. The last unit I built was the Mudlark design, featured in the August & September 2005 issues of SILICON CHIP. Leo – on your comments regarding distortion of 0.01%: is this level of distortion discernible from 0.1%? Can we really hear that difference? I think the rest of your specifications would be easily achievable. Please accept this email as a definite YES for a new project. Keith Columbine, via email. Comment: perhaps we are setting the bar a bit high by mentioning .01% but we also feel that the amplifier should not degrade the signal quality from a CD player, etc. Is such a low figure discernible? Yes, mainly because such a low figure for harmonic distortion also means very low intermodulation distortion.

Edison’s alkaline battery

Edison’s "alkaline battery" is nowadays generally called the nickel-iron cell. These rugged batteries seemingly last forever – old ex-railroad N-I cells can be easily restored to use, even after a hundred years of sitting idle.

The thing not mentioned in your article was that the N-I cell is one of the least efficient of all wet cell designs- requiring from 1.4 to 1.5 times the delivered Ah to recharge. Even a nickel-cadmium (wet) cell only requires 1.25 times capacity to recharge. On the other hand, a well-designed modern lead-acid cell, such as the sealed Sonnenschein gel/SLA will require only 1.05 times delivered capacity to fully recharge.

Heavy though it is, so far there is no competition in efficiency to the lead-acid cell; nothing else even comes close. In the most critical uses, like prime movers and alternative electrical systems, the lead-acid cell is simply the only choice.

Bear Stanley,
Thereon, Qld.

Comment: older readers will recognise N-I cells as NiFe cells.

Hydrogen booster is a con

I read the letter from J. S., of Geelong (Ask SILICON CHIP, October 2006) with amusement. I am a fully qualified auto electrician with 20 years’ experience. The hydrogen booster is typical of those schemes to extract as much money as possible from those who don’t understand how things work!

My other favourite is the two shaped magnets (which look not unlike the ferrite beads we use to prevent RF interference) which are clamped onto the fuel line of the vehicle and which promise to increase the magnetic flux and so increase the power output of the fuel flowing through the pipe. Again, nothing but a con!

Dave Sargent,
via email.

Hydrogen is not a catalyst I have been buying your magazine for a number of years and think you do a really good job. You manage to research and compile an extremely wide field so as to keep things interesting without repeating yourselves. The circuit diagrams and explanations you publish are informative and best of all easy to follow. From the standard set in your publication I will now assume your staff are able to differentiate between fact and opinion. Now if we can refer to "Ask Silicon Chip" October 2006 letter from (J.S., Geelong, Vic) on a hydrogen booster. How much of your answer is fact and how much is opinion? Here is a quote from Roy McAlister, president of the (American Hydrogen Association: "In fact, introducing a small amount of hydrogen, two to five percent, into internal combustion engines that currently run off gasoline, diesel or natural gas increases the efficiency, improves mileage and reduces pollutants quite remarkably". Roy is not the only one respected in the scientific community who has been published expounding the virtues of "hydrogen enrichment of fossil fuels". Your answer to the aforementioned letter seemed to revolve around the efficiency of the process - good thinking on your part but perhaps not the whole story. The principle of hydrogen enrichment is to use hydrogen as a catalyst. In a nutshell, this quote from a book by Michael A. Peavey, page 175: "This is because hydrogen breaks down larger hydrocarbon molecules, creating a larger surface-to-volume ratio and allowing oxygen to more completely burn the components". You are right in assuming the process of dissociation is not 100% but the point is that it makes the fuel burn faster and cleaner (recognised scientific fact). Eric Basill, via email. Comment: Sorry Eric, anyone who thinks hydrogen is a catalyst is wrong. By definition, a catalyst is a substance which accelerates a chemical reaction without itself being consumed by that reaction. Nor does hydrogen break down hydrocarbon molecules or lead to more complete combustion of hydrocarbons. You might have seen it in a book but it is mumbo-jumbo.

Please resurrect Murray valve amplifier

In the early 1960s, Cyril Murray at Sydney University published an amplifier design that was different – high performance (distortion of less than 0.02% under class A1) with no overall feedback. It could be used without an output transformer (with a 500 or 600-ohm speaker).

I made several of these in 1964 and they produced music with great clarity and dynamic range (we had no distortion meter). Because the 600-ohm speaker had so many windings, the combination was extremely efficient and a 12W amplifier was capable of filling a small hall full of people without any sign of distortion.

If you decide on a new valve amplifier design, don’t bother with ultra-linear or single-ended triodes (there are plenty of good designs already) but give Murray’s amplifier a good look. Because it arrived on the scene at the end of the valve era, it slipped under the radar. It was so good that it deserves to be resurrected and I at least would be very keen to build another, perhaps with a bit more power output than the original.

Murray wrote two articles on the amplifier in Proceedings of the IRE (Aust.) in March 1960: p129-133 (measurement) and 134-137 (design). There are a couple of websites that mention Murray’s amplifier.

I hope you can do something on this design, as it was a real breakthrough at the time.

Mick Carrick,
via email.

Comment: there were a number of high-quality output transformerless designs produced in the 1960s which were essentially single-ended push-pull circuits with an output capacitor coupling to a 600-ohm loudspeaker. We could take the same option today but instead of using a high-impedance speaker, we could perhaps use a quality line output transformer to drive a conventional low-impedance loudspeaker.

Incidentally, the 600-ohm speaker was not efficient just because of the high number of turns. Speakers in those days tended to be more efficient anyway because amplifier output power was usually limited.

Please don’t do a valve amplifier

Please, please, don’t sacrifice the interests of the vast majority of your subscribers and readers to waste the limited resources of Australia’s only remaining general electronics magazine on designing and developing a "high-end" valve amplifier.

First, as you correctly anticipate, extremely few among your readers would build it. However, that alone is not reason enough to dump the idea. The BIG reason to dump the idea is that the world of audio electronics is currently saturated with valve amplifier designs.

Those few readers who genuinely intend to spend the vast sums required to experience the valve sound have many sources for excellent valve designs. Apart from all the information available on the internet, there is the current US publication "Audio Express" (audioXpress.com) which has an innovative valve amplifier project (up to 250 watts RMS, believe it or not) in every issue. From the same publisher came the widely known and highly regarded "Glass Audio", now discontinued but back issues are available on CD from their website at audioXpress.com. Every imaginable valve audio amplifier, including many with detailed construction articles, is also available through this same website.

I am not anti-valve or anti-analog. I own a number of valve amplifiers which I do listen to from time to time. I have a very large collection of LPs and a large collection of turntables.

So please, dear Editor, give us a class-A high-end, 50-60W per channel solid-state amplifier to showcase the genius of SILICON CHIP’s audio designers. Apart from its usefulness as a hyper-low distortion general amplifier for those who really appreciate the very best, it would be brilliant for driving high-end tweeters and midrange drivers in higher-power multi-amped systems with electronic crossovers.

Come on all you sensible SILICON CHIP subscribers and readers who are serious music lovers, write to the Editor to stop this crazy push for more valve audio amplifier designs in SILICON CHIP when the world of DIY audio is already awash with them.

Otto S. Hoolhorst,
Brisbane, Qld.

Fake battery-less torch My wife recently returned from the local $2 shop with a new emergency torch for the car. This was the "Environment-protective torch for the 21st century". It claimed that it was to be used for emergency purposes because it didn’t rely on batteries – simply shake the torch and it would charge it. I could see what appeared to be a magnet that would slide through a coil in the body of the torch and a small circuit board under the switch. Intrigued about how efficient the torch was, I disassembled it to see what was inside. Surprise, surprise! The white high-intensity LED was connected via the switch to two CR2016 3V lithium coin cells. The ends of the coil were un-terminated; just wrapped around the batteries. There were no components on the circuit board and the "magnet" was just a lump of iron! The whole thing is a technology fake. So don’t be taken in by these torches, especially if you think it’s always going to work in an emergency. There is, however an upside if you want a couple of cheap CR2016 lithium cells and a high-intensity white LED for around four to five dollars. Roger Forsey, Southbank, Vic.

SMD soldering is more convenient

The UHF prescaler in the October 2006 issue is much appreciated. I have recently changed to SMD as much as possible for home projects, because they are more convenient than leaded components. Bending leads, stuffing rows of them through pairs of holes, soldering, then finding the best approach angles for clipping a forest of leads was a pain.

SMD ICs are just so fast to solder; eg, with a 44-pin PLCC the steps are:

(a) Apply a run of no-clean gel flux, by syringe, along the four rows of pads.

(b) Align the chip with the pads and tack-solder two diagonally opposite leads.

(c) Fill the (nib-like) hollow of a Weller C-0551-6 or equivalent iron tip with solder and wipe over the pins of one side. Eleven pins on 0.8mm pitch can be soldered in one second!

(d) Repeat for the other three sides.

On 1.27mm and 0.8mm pitch devices, there’s rarely need for a pin to be cleaned up. There are claims of success on 0.5mm pitch leads but I haven’t tried. Similarly, it’ll be interesting to try it on J-leaded devices like IC3 in the UHF prescaler.

There’s also an easy way to solder chip resistors and capacitors: take a syringe of solder paste and squirt a little on one pad for each component. Place the components, allowing the paste to help hold them in place. Then hold down each component in turn with tweezers while melting the solder paste. Finally, solder the other ends. Reverting to ordinary solder is quicker at this point.

Some hobbyists use toaster ovens or frying pans to solder whole boards at a time but there are serious risks of component damage and rupture of the PC board vias (barrel separation) due to differential expansion. Bottom preheat to 150°C followed by hot air on top would be safer.

My 50-year-old eyes benefit greatly from a magnifier lamp. While electronics suppliers have them, a large one from a hardware store serves very well. For handling, skidding the board around on a cheap conductive mat works OK for me, when doing just a couple of boards.

Erik Christiansen,
via email.