Classic Articles that did not quite make the Cut
The following magazine articles could have shed some light whenever you found yourself puzzled by all the hype surrounding tube amplifier electronics. I didn’t include them as whole articles because either they were simply too large, had a poor scan quality, or they buried a few ‘gems’ amongst a lot of ‘fluff’. None-the-less, they make for great reading, and if enough people request the whole article, I will oblige some time in the future.
The January/February 1967 Elementary Electronics (above left) article entitled ‘C is for Capacitance’ gives 12 pages on not only capacitor operation theory, but also how different types of capacitors are manufactured, along with the ‘benefits’ of each type. Believe me, if I had the extra space to give you the whole article, I would. It taught me a lot, and ultimately the ‘lesson’ to be learned from the 12 pages was that capacitors types were chosen by the industry based on a criterion that didn’t include ‘tone’. The July 1966 Popular Electronics (above center) article ‘Capacitor Know-How Simplifies Electronic Circuits’ has 7 pages giving the experimenter/builder the savvy to know the acceptable limits in substitution with ‘called for’ capacitors (“Do you accept a substitute ? Or do you abandon the project?”). After explaining the pros/cons of paper/plastic/mica/ceramic types (amongst others) we get a primer on forming/reforming electrolytic capacitors. And how much more thorough can you get than almost an entireissue? Above right is the cover for the July, 1965 issue of Electronics World. Inside are pages and more pages explaining yet once again the benefits of each type; the low inductance and temperature compensation capabilities of ceramic disc as only one example. The moral to these tales was that in high frequency work capacitor type can and does matter quite a bit, while for audio work it absolutely doesn’t matter at all. See ELECTRONIC PARTS for ideas on how capacitors were chosen by early manufacturers.
‘Cheap scopes often produce fuzzy patterns with prominent retrace lines, especially at higher frequencies.’
The header above from the July 1959 Radio-Electronics article looks cute, but the reading is anything but. It is one of those articles that you need to do a lot of reading ‘between the lines’. In Television and Radio work, the technician can easily be misled by his interpretation of waveforms on his oscilloscope. We are cautioned that… ‘a faulty waveform at the plate of the first stage of a section usually shows that stage to be faulty. And this is usually so unless there is feedback from a following stage.’ Other tips include the fact that capacitance in the oscilloscope leads can throw off a reading, as can using equipment that is not in 100% peak operating condition. A similar article in the February 1968 Radio-Electronics states that… ‘A poor scope may suffice for signal tracing, where the absence of presence of the signal is all we need to check. For pulse and sawtooth waveforms, however, the high-frequency signal components will be diminished in a scope having poor response, and the waveform will then be distorted.’ I then thought about where and when I bought any oscilloscope I have ever owned and the condition it could have been in. If you are like me, you probably bought your scope secondhand at the ham swap-meet, or through the local ‘Buy & Sell’ type paper. Also, from the book The Audio Designers’ Handbook (H.A. Hartley) we are shown that a normal 1kHz sine wave can appear to have a cross-over notch if there is distortion present in the signal (we are shown various harmonic distortions at various phase relationships), or now factor in the above mentioned scope troubles! No wonder today’s tube guru doesn’t like using an oscilloscope; learning how to use one takes time away from writing ‘bullshit’ articles!
The article entitled ’97-cent Hard Rock Fuzz Box’ tries to teach us exactly what ‘Fuzz’ is, and that for 97-cents (at the time of publication) the tube technician can add ‘the latest sounds to your current amplifier’… and ‘…where the commercial fuzz boxes provide the conveniences of adjustable fuzz quality and a footswitch, the price range of $12 to $40 often puts it well outside the budget…’. Although the circuit offered is simple diode clipping, interestingly enough this very same idea was used by ‘modern-day’ commercial units from the Tom Shultz Rockman to the Marshall Silver Jubilee amplifier; an ‘advanced’ clipping circuit is also found on the JCM900 (Oh, those incorrigible engineers!). If you know what you are doing, diode clipping sounds as good as any vintage fuzz box.
The article ‘Can An Amplifier Tube Have Zero Transconductance?’ is one of the painfully rare cases where electronics magazines from the 1970’s have anything that interest me. Sure, they have a plethora of ‘Fuzz-Box’ circuits, solid state Reverb and Tremolo circuits, and I even have an issue featuring a predecessor to the Morley optoisolator Wah-Wah circuit, but tubes were a dead issue (no pun intended) about the mid 1960’s! The article shatters the often held belief that varying the grid bias in a triode will vary the plate current. The moral of the story is that the plate resistance changes, and by experimenting with a triode tube and a constant current power supply, you ‘prove’ this. Where this is important is in the analyzing of preamplifier circuit Thevenin equivalency and optimum output tube impedance matching (believe it or not, varying the bias on your output tubes will vary the plate-to-plate impedance).
The age-old question ‘Why amplify the frequency if you can’t hear it?’ was tackled by Radio-Electronics.
Radio-Electronics magazine asked in March 1966 ‘The Wider The Band The Higher The Fi?’, and a reprinted/’updated’ article appeared in October 1967. The ‘panel’ consisted of Robert E. Furst, the vice-president of engineering at Harmon-Kardon, and Victor Brociner, assistant to the president at H.H. Scott. These were learned men, with considerable electronics education and training. Yet these two experienced engineers were diametrically opposite as to the validity of having extended bandwidth capabilities in not only the amplifier, but the output transformer and speakers as well. The crippled bandwidth of vinyl records and tape are explained on one side of the argument, along with the threat of turntable rumble, switching transients, stereo subcarriers, and vinyl record eccentricities. These affect the sound by adding noise, as well as modulating the audio. The other side of the debate held that the normally accepted audible hearing range is antiquated and a restricted bandwidth fosters undesirable phase shifting between audio signals in the middle of the audio band, and these would be clearly noticeable regardless of the argued hearing range. I wonder if Leo Fender thought about this stuff, or if he just asked his supplier how cheap he could get 10,000 output transformers.
This article came really close to being included in its’ entirety here in the Tone Lounge. Many ‘engineers’ insist that if you should remove two output tubes from your 100-watt Marshall amplifier, you had damn-well better adjust the output impedance selector to one-half of the load connected to the amplifier. What’s supposed to happen if you don’t follow these dogmatic warnings? Not much, according this May 1962 article published in Electronics World. Why the full article wasn’t included isn’t a simple discussion. Number one; it did not address resultant reflected impedance, and its’ effect on plate current. That seems to matter to me, although it wouldn’t enter into the picture with a properly designed amplifier. However, since many people like to ‘improperly’ read between the lines, skewing any derived conclusions to support their own agenda, I chose not to include the full article. Number two; the tests were done with a low wattage amplifier (15 watts worth of Class A, using two 6BQ5’s). Again, to avoid the argument that these test results could not be repeated with a 100-watt Marshall, the article was not included. And thirdly; the circuit shown is a distributed-load type (Ultra-Linear). Back at concern number one I mentioned the plate current. The screen current increase may be problematic with a poor tube/design/bias setting. Those issues are of no concern with a distributed-load amplifier, but only the Marshall Major has an Ultra-Linear output stage. So, to avoid that argument, the article wasn’t included. The best way to teach yourself what the article may have shared with us all is to do the experiment yourself with a 100-watt Marshall. The test is to measure into a dummy load the output wattage, and frequency response, with various mismatch combinations. I will include a chart below, that shows the condensed test results from the magazine article.
Of course, by following the ‘proper’ loading of your tube amplifier, the original power output, bandwidth, and frequency response is maintained. However, by ‘cheating’ a little we can obtain interesting results. You should know exactly what you are trying to achieve, and make sure the output circuit is designed and set up ‘correctly’. The word ‘correctly’ opens up a can of worms most gurus would rather avoid.
‘Distortion In Loudspeakers’ has a bit of everything that makes me not include the whole article here.
Electronics World ran this ‘Distortion In Loudspeakers’ article in their June 1962 issue. I chose not to include it here because it has too many pages that scan poorly. The pages are filled with charts and oscilloscope photographs explaining that it is usual to have the highest distortion levels in the low bass frequency range. This is so because of large voice-coil excursions necessary to reproduce these low frequencies. Other causes include non-linearity of the cone suspension and the non-uniformity of the magnetic field over the path of the voice-coil travel. I’ve just told you in three sentences the crux of six pages worth of highly technical explanations accompanied with graphs, charts, and oscilloscope photographs. You can thank me later. Similar explanations are seen in the book The Audio Designer’s Handbook(H.A. Hartley), along with the added simplified explanation/pictures of deformation of the cone material itself. As I read both publications, I chanted…. ‘But in a guitar amplifier it sounds wonderful!’ between every sentence. You should do the same; it may make reading a chore, but the lesson is well learned.
The April 1974 issue of Radio-Electronics contained a lengthy article entitled ‘All About Transformers’. Nothing was mentioned about interleaving windings for improved efficiency and tone, but a mention was made about core saturation. That brief paragraph and accompanying picture had a profound effect on how I thought about output transformers. To see that picture, go to the ‘Vintage Advertising’ section about transformers. You can get also there by just CLICKING HERE
Guitar and/or speaker cables are thought to be a ‘secret ingredient’ for some players. The popular consensus is that expensive designer cables from people like George L. can really make a difference in your tone. Well, I have always contended that past a certain point, all cables are good enough for me. GUITARS, ETC. has more to the story, but the header from the February 1991 issue of Radio-Electronics is shown below. I won’t include the full article because it is far too long, explaining the tests conducted and how the tests were construed as being relevant to the task at hand. The bottom line is that for audio work there is no difference between ‘good’ cables.
Speaker impedance, and the concept of what it really is, seems to baffle (pun intended) a lot of folks. Well, Radio-Electronics tackled the very subject back in May of 1951. Below is a glimpse of the title page.
There is a lot to digest from these pages, but here in a nutshell is what you should know. First of all, the rated impedance (4-ohm, 8-ohm, 16-ohm) is the nominal impedance. The actual impedance has a wide variance, and depends on the frequency. The interesting tidbits? Unfortunately, the writer does not elaborate any further than the following.
… it is interesting to study the variation in impedance with type of baffle. Even if a hand is placed in front of the speaker, a distinct change will occur …
Oh, oh. Maybe a baffle made from cryogenically frozen mouse dropping can make a difference in tone. I have a hunch the bit about waving your hands purportedly changing the impedance means the (unshielded?) test leads are picking up a little too much. But, this is up to you to investigate, and spend some time with.
There are many articles throughout the long history of electronics magazines aimed at teaching the neophyte and advanced student as well, reminding us all that you can learn something new every day. Many are great supplements in rounding out your education, with titles like ‘Understanding Push-Pull Drivers’. Look through your local used bookstore and buy up the issues prior to 1970 from such names as Popular Electronics, Radio-Electronics, Radio-TV Experimenter, Electronics World, and even some issues of Science and Mechanics. The articles will deliver as promised, and the advertising is a definite bonus.