Tape Line-Up (Part 1)
Know anything about bias, test-tones or calibration? If you use a tape recorder then you should! In the first of two parts, David Mellor explains the need for tape recorder line-up procedures and unravels the mystery behind this arcane science which can have a profound affect on the quality of your recordings.
If you have ever read through the 'Care and Feeding' section of the user's manual to your tape recorder, then chances are that from time to time your eyes have alighted on the paragraphs headed LINE-UP. In the first of two articles, David Mellor tells you all you need to know about the theory behind this arcane science. If you ever have the urge to delve into your machine's mystic recesses, before taking the plunge, read on...
Did you ever, in your early youth, go train spotting? You did? Well you will probably know all about the fact that in the last century, the station clock was a law unto itself. Every major railway station kept its own time, with the result that Manchester time was different to Sheffield time, which was different to London time etc, etc. Each station clock was counting the same number of hours in the day of course, it was just that when you came to the end of your journey, you had to reset your pocket-watch. Fortunately, the invention of the telegraph put an end to this nonsense, and after some decades of argument, one standard was chosen.
A similar situation existed with tape recorders when they first started to become commonplace in the late 1940's and early 1950's. There were no defined standards for line-up, which meant that although a tape may have played perfectly on the machine on which it was recorded, there could be no guarantee that it would sound anything like as good on another machine - even one of the same make and model.
That is one reason why we line up tape recorders to defined standards, which I shall expand upon shortly.
The other major element of line-up is concerned with freedom of choice - a lofty sentiment indeed. When you left the shop with your brand new TASTEX 99 recorder or whatever, the salesman probably shouted to you from the doorway the brand of tape it was lined up for, ie. for which the bias and EQ were optimised. That scrap of knowledge would be very useful if you were prepared to stick to the same brand for ever and ever. There are, however, a number of tape manufacturers, most of whom make more than one type of tape. Ampex for instance make 456, a high output tape which I have found to be compromised to a slight degree on signal print-through (the unwanted transfer of magnetism between adjacent layers of tape), and 406, which has better print-through characteristics but has a lower output level. So you see we have a choice - but if you use a tape other than that which the recorder was originally lined up for, chances are that it will sound dreadful regardless of the tape's intrinsic merits. Change of tape means change of alignment, both bias and EQ. This is troublesome but necessary, and once you have found a tape that suits your requirements then you should not need to change brands.
Summarising then, tape machines need to be lined up so that (a) tapes can be freely interchanged between recorders, and (b) we have a choice of tape type. Onward...
A bewildering array of initials, but as CCIR and DIN can be regarded as synonyms for IEC, it boils down to a choice of two sets of standards that we need to know about. (There is a third body, the AES, whose standard is used for tape travelling at 30 inches per second, but as I for one cannot afford to burn tape at that rate I shall go no further than to mention it.)
NAB stands for National Association of Broadcasters; IEC for International Electrotechnical Commission. The Americans were first off the mark with their NAB standard, IEC in Europe came a little later. Some (including me) would say that IEC is the better standard to adopt for it is, after all, used by an influential major broadcasting corporation resident in this country!
These two sets of standards are the reference points throughout the civilised world for tape recorder users. The benefit conferred is that when you label your tape either 'NAB' or 'IEC' standard, and include test tones at the head of the reel, you know with confidence that you can send your master tape anywhere in the world and the receiving person or organisation will be able to play back your tape exactly how you recorded it. If you do not think that this is important, and send out tapes of unspecified standard without test tones at the head, then you cannot expect anyone else to treat them with the respect they may deserve.
When you have read this and the succeeding practical article, you will hopefully have a firm grasp of the principles of tape recorder line-up and be ready to get your studio set-up into professional order.
So simple - but if this vital adjustment is out by a fraction on your machine, then it will really mess up your recordings, making them dull and lifeless - a bit like the shampoo ads! Fortunately, the azimuth standard is something all the previously mentioned standards organisations agree upon. So what is it?
Looking at Figure 1, you can see that azimuth is the angle between the head gap and the direction of travel of the tape, and should be set to exactly 90 degrees as in (b). You would be amazed at how many different versions of 90 degrees can be found, even in brand new machines (especially cassette decks), so just because you bought new, you cannot automatically assume that the azimuth setting will be correct. (Note that we shall be using the same version of 90 degrees as Pythagoras, it worked for the Pyramids and it will work for us!)
To find out why azimuth is so important, take a look at Figure 2. This shows a length of tape with a high frequency tone recorded on it in correct azimuth. (The shaded line areas are the high parts of the magnetic waveform, and the unshaded parts the low.) The playback head is obviously well out of alignment as part of the head is reading a high in the waveform and the other part is seeing the low. This causes cancellation of the recorded signal and produces a low output or, in the worst case, no output at all!
The effect is most pronounced at high frequencies. If you consider that at a tape speed of 15ips (inches per second), a recorded frequency of 10kHz (that's 10 thousand cycles per second) has a wavelength on tape of less than 1/25th of a millimetre (!), then you can see how accurately the azimuth must be set on both record and playback heads.
It is possible to define the azimuth standard yourself from first principles, but as this involves turning the tape oxide-out and recording a tone through the backing, I know you will forgive me if I do not elaborate on this. For as we shall see next month, there are much easier ways of checking tape head azimuth.
In addition to telling us to get our heads straight, what the NAB and IEC standards do for us is to define levels of magnetic flux on tape with respect to frequency - these are the so-called tape equalisation (EQ) curves, and this is where the two standards differ.
If we set our tape recorder to play back according to one of these EQ curves, then we know that anyone else who has aligned his machine to the same curve will be able to play our tape correctly. (It is unfortunate that, having bought a machine of a particular standard, it must always be lined up to that standard. You cannot turn a NAB recorder into an IEC recorder other than by changing components internally.)
I think the time has come when I must introduce a long word -'nanoWeber', as in 'nanoWebers per metre' or nWb/m. The Weber (normally pronounced 'Vay-ber' in this country) is the unit of magnetic flux and the flux density on the tape is measured in nWb/m. It is possible to obtain calibration tapes (or 'test tapes' as they are often called) pre-recorded to either NAB or IEC standard. Both define a 'zero' level in terms of magnetic flux density. The NAB tape defines a zero level of 200nWb/m at 1000Hz (1kHz) and the IEC tape a level of 320nWb/m, which is roughly 4dB higher.
These levels of magnetic flux density have absolutely no given relationship to the number of volts going into or coming out of the tape recorder. In particular they relate in no way to the concept of '0 VU' until a relationship is defined by the user. This is a great pitfall for the unwary and a great many people will happily equate electrical levels with magnetic levels with no justification for doing so. The standards authorities (NAB/IEC) define only magnetic levels, leaving you free to set 200nWb/m to -10dBu or 0dBu or whatever level you wish in your studio, as this has no bearing on tape exchange.
I ought to mention at this stage that NAB and IEC each have their own preferred method of measuring magnetic flux density which differ by about 0.85dB. Therefore if you measured an IEC tape of 320nWb/m by the American method, it would read only 290nWb/m. A small point, but one ought to be aware of the difference.
In addition to the different nominal zero levels, there are equalisation differences. You are probably familiar with cassette equalisation (EQ). Ferric cassette tape uses 120µs EQ and chrome tape uses 70µs. As I am not giving a lecture in the physics of tape recording, but only an outline of matters of practical significance, I do not intend to go into excessive technicalities. In essence, the problem is that high frequencies are more difficult to record than middle and low ones, and the degree of difficulty is dependent on tape speed (or in cassettes, tape type). High frequencies, therefore, need boosting.
The NAB standard sets an equalisation of 50µs for both 7½ips and 15ips. IEC on the other hand specifies 70µs for 7½ips and 35µs for 15ips. This means that at 15ips, IEC puts more high frequency signal onto tape than NAB, resulting in a small improvement in signal to noise (S/N) ratio. (There is, of course, a trade-off between S/N ratio and high frequency saturation level). At 7½ips the situation is reversed, and NAB puts more high frequency onto tape. For both tape speeds, the NAB standard also specifies an additional bass boost on record, and an identical bass cut on playback.
This all sounds complicated, I know, and I am tempted to write the last paragraph again so that you can have another go at it. I think it might be more easily explained as follows:
Suppose you had an IEC recorded tape and you played it on a NAB tape recorder. It would reproduce with more high frequency (treble) content than intended and less bass. If you had a 15ips NAB tape and played it on an IEC machine, the treble would be dulled and the bass boosted. This accounts for one of my reasons for choosing IEC for my own studio. I always record at 15ips, and I prefer that if my tape is played on a machine of the wrong standard that it should sound brighter and lighter rather than duller and heavier - personal preference.
Whichever standard you opt for on your recorder, the test tape you should buy will contain a variety of useful material for lining-up purposes. In addition to the 1kHz tone at zero level, it will have a frequency tone (or tones) for adjusting the azimuth and also a complete set of tones from 30Hz up to 18kHz or so, permitting a complete check-out. Whatever the frequency of the tone on the test tape, you can be guaranteed that will be recorded in correct azimuth and be at accurate zero level (or -10dB) taking into account the necessary equalisation. If you line up your tape recorder to play back all the tones at the same level, then your machine will be 'flat' within a fraction of a dB (decibel) - depending on the quality of the test tape.
Having aligned the playback side of your machine, the record side is calibrated to this. Your tape machine will then record and play back flat, ie. what goes in is what comes out! And what is more, you will know exactly what is on the tape and have no nagging doubts that errors in recording are being masked by errors in playback. If anyone then has a problem playing back your tape, you know that it will be their machine that is at fault and not yours. Wonderful feeling isn't it?
I wish I could say that having lined up your tape machine, then you could forget about it forever more. Unfortunately, this is not the case because you will have to continually adjust for head wear - the tiny oxide particles on the tape act like fine sandpaper rubbing against the heads and ever so gradually wear away the head surface where it makes physical contact with the tape. This manifests itself as a reduction in high frequencies going onto and coming off the tape. If tape heads are always kept clean then this will not prove to be a problem as a regular line-up will compensate for the wear with no noticeable side-effects. When the amount of wear has become too great, however, then you will probably run out of adjustment range and I am afraid that when that time comes you will have to put your hands into your pockets and shell out for a head transplant!
I have mentioned that having aligned the playback side of the machine to the test tape, then you will align the record side to the playback. However, there is one small step that must be done between these two line-up procedures, and that is to set the bias.
The invention of AC bias in Germany in the 1940's was a giant leap for tape recorderkind. Before AC there was DC, which worked by arranging that the tape passed a magnet before reaching the record head. The purpose of the magnet was to set a medium level of magnetism to which the record head could add or subtract. DC bias was better than no bias at all, but machines which used this system were useless for anything other than speech recording.
AC bias, which really made the tape recorder what it is today, works by adding a high frequency tone (usually around 100kHz) to the signal to be recorded onto tape. This makes all the tiny magnetic domains rattle round their hysteresis loops at a considerable rate of knots making them more receptive to the lower frequency audio signals (20Hz-20kHz ideally) that we want to be recorded. Although bias is definitely a good thing, too much bias is too much of a good thing; it has to be set just right. How much is enough depends on which brand of tape you intend to use.
Setting the bias correctly means optimising various performance parameters, such as: maximum output level, high frequency saturation level, distortion, signal to noise ratio, drop-out performance etc. There is quite a lot to think about here but fortunately there is a simple rule of thumb that will get you as close to right as will make little practical difference, as we shall find out when we come to the 'this is what you do' section next time.
Having covered the theory behind the need for tape recorder line-up this month, in next month's article I shall give you a rundown of the equipment you will need to line up your multitrack and stereo recorders, even your cassette machine if you are really keen. I shall also supply a step by step guide to correct operation of your tweaker - and tell you where to get one if you are lacking in this department! In the meantime, remember what I said earlier - it's a nice feeling when you know that you're doing it right!
Feature by David Mellor
mu:zines is the result of thousands of hours of effort, and will require many thousands more going forward to reach our goals of getting all this content online.
If you value this resource, you can support this project - it really helps!