Updated from an original article by Noel Keywood, in Hi-Fi Answers, 1979.
High levels of distortion are produced on stereo recordings from errors of what is known as Vertical Tracking Angle in a cartridge, and precisely what this is is shown in Fig 1. You can roughly visualise it as the angle a stylus cantilever makes with a disc's surface, although to be more precise it is the angle between a line drawn from the stylus pivot to the stylus point of contact with a disc and the disc's surface, denoted by A in the diagram. Sometimes, and for good reason, the vertical tracking angle A is seen in the other position shown, transposed 90° anti-clockwise. This shows that the plane of motion of the stylus is in fact tilted over relative to the vertical by an amount equivalent to the vertical tracking angle. Note that the actual angle of the stylus on its cantilever may be entirely different from the vertical tracking angle and is known as Stylus Rake Angle. The two are however often confused.
So what's the significance of this angle? Its origins lie in the mechanics of the disc cutting system shown in Fig 2 where you can see that the cutting stylus must itself have a pivot above the disc's surface and therefore a vertical disc cutting angle of some sort or another.
This was of little consequence until the introduction of stereo in the 'Fifties when suddenly the cutter was expected to move up and down as well as sideways when cutting a lacquer. Now look at Fig 3 to see what you get on a record when feeding a nice clean sine wave into this system in the vertical direction. The actual cut in a disc looks very different from the original, 'distorted' in fact, and so you can see that there is a mechanical translation process at work here due solely to the presence of this inclined plane movement in the stylus cutter. Using a replay stylus at the same angle does, as you might well imagine, translate the mechanical disc signal back into a clean sine wave again so the final outcome is, all things being' equal, which they aren't, zero distortion on playback with a pickup cartridge.
I should point out perhaps that the shape of the signal on the disc is in itself of no consequence. What does matter is that the playback stylus must move in the same plane as the cutting stylus or distortion will result. Some of you may already be mentally protesting that the stylus does in fact move in an arc and so the plane of motion varies its angle. Although this is true, maximum vertical movement of a stylus rarely reaches 100µm (peak-to-peak) or 1/10 mm due to imprinted modulation (as opposed to warps etc.) and with a typical cantilever length of around 5mm change of vertical tracking angle will be about 1° maximum, which is insignificant in practical situations.
To ensure that everybody used the same angle and distortion was avoided on vertical cuts or modulation a common cutting/playback angle had to be adopted, and after a spell of technical anarchy this became 15° in the United States and 20° in Europe, a tolerance of plus or minus 5° being allowed on the European figure. In fact it does appear that Britain standardised on 15° according to BS1928 :1965 but this is of little practical consequence now since most of the cutting lathes in this country (and there are a lot) come from Europe and are aligned to IEC/DIN standards, which we tacitly acknowledge. This standardisation would appear to have solved matters with the outcome being distortion free stereo reproduction, but as I said earlier, all things aren't equal in the cutting/playback process and there was one further complicating factor to be discovered before the ideal could be approached in practice.
In the early 'Sixties it was discovered by CBS Laboratories in the United States that although cutters worked correctly in the 15° cutting plane as recommended by the Recording Industries Association of America (RIAA), replay fidelity of test signals nevertheless appeared incorrect (1). It was eventually deduced that the soft lacquer in which a cutting stylus records the electrical signal fed to it was in fact deforming under the strain of cutting and then springing back after the cutter had passed as shown in Fig 4.
With different amounts of spring-back, according to cutter depth and hence stress on both the lacquer and cutter itself this lead to the final groove shape being equivalent to that theoretically produced by a cutter with a vertical cutting angle of about 0°. This would require a cartridge with its stylus pivot lying on the disc surface to be designed if minimum distortion was to be achieved on playback, and so it was incumbent on cutting lathe manufacturers to increase the actual physical working plane of their cutters by a substantial amount in order to realise ideal playback conditions. In practice CBS found that their cutting angle needed to be increased by about another 14° in order that final modulation slant on the disc sprang back to 15°.
Although this discovery was presented at an IEEE meeting in February 1963 I have not been entirely happy that it got much further since the topic seems rarely to be mentioned compared with say, lateral tracking error, where enormous amounts of energy and time are devoted to the removal of its last effects. Measuring vertical tracking angle on cartridges tends to enforce this feeling since significant errors often exist. However, London's thriving cutting industry is probably responsible for a majority of the records we listen to in the UK and most studio cutters seem to be compensated for the effects of lacquer springback, as far as my casual enquiries over the last few years can ascertain. Neumann, whose lathes and cutters seem most popular by far, include this paragraph in their literature for the SX-74 head, for instance - 'Care was taken with regard to the vertical tracking angle to assure, on the one hand, a 15° angle to meet the DIN 45 542 standard and, on the other hand the desire of the International Electrotechnical Commission (in IEC Publication 98) to standardise an angle of 20° ±5°. For this reason the cutter was constructed to cut an 18° angle when correctly mounted. Lacquer springback is included in this calculation.' I haven't yet found the time to sort out the discrepancy between Neuman's quoted DIN spec of 15° vertical tracking angle against that published in DIN 45 500 where an angle of 20° ±5° is quoted as per IEC 98 for cartridge vertical tracking angle, except that allusions are made in the latter document to inevitable assessment errors and the need for tolerances to be kept in mind. It would appear that the cutting world is now achieving an effective modulation slant on disc of around 15° in the United States and 18° in Europe and that for minimum distortion we should therefore be aiming for these figures in pickup cartridge vertical tracking angle. As you might have guessed, the point I am about to make is that in spite of adequate passage of time too many cartridges do not appear to approach this ideal.
Both CBS Laboratories in the United States and Germany's technical standards institute Deutsche Industrie Normen issue test discs suitable for assessment of vertical tracking angle or distortion due to error in this parameter on a pickup cartridge. Although from entirely different sources, results obtained from these discs tally well, correct vertical tracking angle as indicated by the DIN 45542 disc consistently leading to low distortion on vertical modulation from CBS STR-112, optimised to possess a 15° modulation slant. All measurements referred to here have been taken from these two sources.
Because the practical subjective consequence of vertical tracking angle errors on sound reproduction are really of most significance I don't intend to dwell too long upon measurement technicalities, but a few consequences of incorrect vertical tracking angle are presented for illustrative purposes. Classic in measurement terms is the waveform to be seen in Fig 5. The upper trace alternately shows rounded and sharp peaks of significant second harmonic distortion and the spectrum below shows the harmonic structure. From the same disc can be seen significant intermodulation distortion in Fig 6 due to this cause. Lower trace shows two signals, one of 4kHz and one of 200Hz from the disc; above can be seen the 4kHz component filtered out. The change in level is due to presence of distortion sidebands from amplitude modulation which, in more straightforward terms, means that the 4kHz tone is now accompanied by other tones of significant level that weren't there in the first place and wouldn't be there either but for a vertical tracking angle of just over 30°.
Actual level of intermodulation distortion here reaches around 14%, no less, and can be heard as a wavering grating buzz behind the 4kHz tone. In fact the ear can pick up about four to six distortion sidebands at this level, and they are spaced at multiples of 200Hz above and below the 4kHz tone. It is sometimes argued that changes in tracking force and stylus movement over warps introduce too much variation into the vertical tracking angle to make any one figure valid, but this is not true. The oscillogram Fig 7 shows for instance that the 10% distortion component varies by about ±5% of its own value once every disc revolution (1.66 secs or 1.6 divisions on the x axis) over a common dish warp, or by about I % of the total signal level. My own measurements are made on cartridges at or near the recommended tracking force where any error will be at a minimum but even so I find that vertical tracking angles commonly extend upward from 18° to around 30° on modern cartridges. The graph in Fig 8 shows the results of measurement on a Shure VI5-111 using the DIN 45542 disc. The purpose of this disc is to determine vertical tracking angle rather than measure its effects and as you can see the VI5-III shows an effective angle of 26 ° which is too high. Distortion when tracking vertical modulations on STR-112 was as a result 6·7%, whereas a properly engineered cartridge will manage typically 3% or less on this disc. From experience I notice that in general Shure cartridges hover around this vertical tracking angle figure of 26° whilst ADC and Empire designs generally display angles of 26° plus, often reaching 30°. Such angles yield harmonic distortion figures of 6% to 8% from the STR-1l2 test disc and are not confined to budget designs. I have measured two ADC XLMs at 30° and two Empire 2000s at 30 ° , neither design being exactly cheap to buy. Stanton generally manage 22° and so do Pickering, since they are associated with Stanton, this figure being within IEC limits. Generally, European and Japanese cartridges measure within IEC 98° limits too, of 20o±5° and this leaves Shure, Empire and ADC out in the cold. The only possible excuse these three large manufacturers of cartridges can make concerns accuracy of the CBS and DIN test discs, for in any measurement work concerning the use of discs accuracy of the disc itself must always be kept in mind. However, if the discs are both wrong, then a majority of cartridge manufacturers are also wrong too, and I think both these possibilities unlikely.
The subjective effects upon music of incorrect vertical tracking angle are known but little talked about. You may have wondered by now why, if they generate so much distortion, expensive cartridges can sound any good at all when suffering this fault. I have purposefully left this point to the end in order that a sense of perspective can be obtained. Needless to say, such distortion levels are not generated all the time on all signals. If they were ADC, Shure and Empire cartridges would, for instance, sound as awful as you might imagine, but this is not the case. The contribution made to fidelity by achieving correct angle is minor relative to many other design factors but nevertheless important and easy to achieve. Pure centre images in the stereo stage are produced by lateral stylus movements and remain unaffected by vertical tracking angle errors. Full left or right images in one speaker or the other are produced by the stylus moving both sideways and upwards (or at 45° upward from the horizontal) and the upward component will be distorted. Pure out-of-phase signals resulting from vertical only movements will also be distorted.
By connecting a high quality speaker across the positive terminals of an amplifier's left and right output channels vertical components within a music signal can be listened to, and with this technique the advantages of getting the vertical tracking angle right become obvious. I have noticed in particular that cymbals possess less of a grey rasp and become better defined and natural whilst spread choral works also show less edginess and considerably greater freedom from what can only be termed raucous background muddle. In general left and right images are cleaned up and become aurally more palatable. The amount of distortion generated rises considerably with recorded level and these effects are not pronounced at low levels. Others have noted the same effects (2). Raising or lowering the pickup arm so that it is angled makes little difference since it will change matters by a few degrees at the most, by the way.
1 'The vertical tracking angle problem in stereophonic record reprOduction' B.B. Bauer, CBS Laboratories
2 'How we judge stereo cartridges'. High Fidelity Magazine, Vol 19, No 2, Feb 69. Daniel Gravereaux (CBS Laboratories) Further information on vertical tracking angle in cartridges:
'A study of tracking angle errors in stereo disc recording' Woodward and Fox, RCA Labs.
'Measurement of distortion due to vertical tracking angle errors in stereo disc system tracking angle errors in stereo disc systems.' Halter and Woodward, RCA Labs.
'The design of high quality stereophoniC gramophone pickups for minimum wear and distortion.' D. Jaquess, Decca Record Co.
'Pickups - The key to Hi-Fi' J. Walton
BS 1928:1965 Specification for Processed disc records and reproducing equipment. para - 4.3.21.
DIN 45500 sheet 3. Requirements for disc record reproducing equipment. para -3.9 IEC Publication 98.