Making the tracks on video tape visible with a magnetic fluid

Philips tech. Rev. 40,129-132, 1982, No. 5 129 Making the tracks on video tape visible with a magnetic fluid A. M. A. Rijckaert It has been known for more than fifty years that magnetic effects at the surface of materials can be made visible by means of a liquid containing a magnetic substance in suspension. A suspension of this kind is known as a 'magnetic colloid' or a 'magnetofluid'; sometimes the term 'Bitter water' has been used, after F. Bitter, who published the method in 1931[1]. Magnetic colloids have been in use for some time at Philips Research Laboratories, mainly for making the magnetic tracks visible on video tape. This is done by using an aqueous suspension of aggregates (clusters) of iron-oxide particles about 0.007 J.1min diameter. The clusters must not be larger than about 0.1 urn [2] because of the very fine distribution of the magnetic" patterns on the tape. The special feature of our method is the good use it makes of diffraction effects when the patterns are observed under an optical microscope. This provides a simple way of evaluating the operation of the servosystems that control the movements in the mechanical part of a video cassette recorder - in our case the Philips VR 2020 - during the writing process. The advantage here is that the effect of the servosystems for writing can be observed separately from that of the servosystems for reading. The video signal written on the magnetic tape is approximately sinusoidal. After treatment of the tape with colloid a pattern appears on its surface - a 'Bitter pattern'. The pattern consists of two accumulations of clusters of iron-oxide particles in each sinusoidal period; this is because the forces acting on the particles are proportional to the square of the magnetic field-strength [3]. The period of the lattice structures that form the Bitter pattern is thus half that of the corresponding sinusoidal signal. Our experimental arrangement is shown in jig. 1. The magnetic tape treated with colloid is illuminated at a particular angle (through a bundle of glass fibres) and the patterns on the tape are observed through a microscope M. The light diffracted back towards the microscope is increased in intensity if the path difference s is equal to the wavelength of the incident light or to a multiple of it; see jig. 2. The dominant wavelength in the diffracted light thus depends on s and hence on the period of the observed lattice structure. Since the incident light is more or less white, the regions on the Ing. A. M. A. Rijckaert is with Philips Research Laboratories, Eindhoven. tape that correspond to a different period appear differently coloured under the microscope. The diagram of jig. 3 shows the frequencies, dependent on the magnitude of the video signal, that occur in the frequency-modulated signal written on the tape. The information in the video signal relating to Fig. 1. Arrangement for observing Bitter patterns. M microscope. o object, a video tape treated with magnetic colloid (Bitter water). G bundle of glass fibres.!i Fig. 2. Origin of interference due to the periodic structure of the Bitter patterns; s path difference after diffraction of the light rays; 8 angle between the incident and the diffracted light, d 'lattice constant'. f t. 4.8MHz p -t Fig. 3. The frequenciesf of the frequency-modulated signal written on the tape and containing the video-signal information, as a func-. tion of time t. Wand B correspond to the 'white' and 'black' levels of the video signal. The parts of the signal corresponding to the line-synchronization pulses are denoted by p. [1) F. Bitter, On inhomogeneities in the magnetization of ferromagnetic materials, Phys. Rev. 38, 1903-1905, 1931. [2) Magnetic liquids, Philips tech. Rev. 33, 293, 1973. [3) N. H. Yeh, Ferrofiuid Bitter patterns on tape, IEEE Trans. MAG-16, 979-981, 1980. p w B

130 A.M.A.RIJCKAERT Philips tech. Rev. 40, No. 5 the successive lines of the television picture is divided up by line-synchronization pulses. (These pulses are used for synchronizing the flyback of the electron beam in the picture tube after a line has been written.) The lowest level in the signal corresponds to 3.2 MHz, the highest to 4.8 MHz. Since the magnetic heads in the VR 2020 recorder move at a velocity of 5.08 mis, these frequencies correspond to signal wavelengths on the tape of 1.59 urn and 1.06 urn respectively. The 'lattice constant' d of the Bitter pattern thus varies from 0.80 urn to 0.53 urn. Assuming that the light is incident on the Bitter pattern at an angle e = 60 and that the interference pattern of the first order is observed in the microscope, then the wavelength of the diffracted light is equal to Fig. 4. The Bitter pattern, without interference. Two video tracks (with no space between them) are shown over their full width. The arrow indicates the direction of movement of the magnetic heads. The track width is approximately 22.5 urn...'---.'.-.-,--... ------- '...,-.'-..,----...,U,. " M'.'. --'--.,_.,..... 1 n, 1, u,._ ~~~:::::::::::::::::::::=====::::~::::::=.::.:=.=.2:~==-~-iii... ~- ~...!'~~.!!!..._----,._-.." _---_.,_._.. _..._.'"._.. ~------_.- _._._ ~... ----.,,-..,.,,,~ '_''_I_t-..~==~==~._.._.._.._-,_..._.._.._._._._,_._-_._._._._ " _, _ M... ""._.... ". '_'_'---. _._~;.~~_--=::=:!==;:==.=:_;=-._..._.~-. '"._.... "... --_._...._,.._-,~-_...._-,.- -. '.----_._--..-.-. -'~----_.-_._--.. '~'--.-_,._-- _..._._------...._-~-_,_,._ --...----.._._-....._-.. - -.......-.._....~..._--_ -.. - -.._-----_......._--~._---_.._------------ - -------._-.._--,.._---,.._-_._-- --_.._--_._-_.._-_ ~._.._..._-_..._-_.._-_._-_.._-----.-.._------.. M 4 ~...._..._-._- _-_ _-- -.,--.. _ -._- --._._- --.. --'._ --.. _._ --- _'_ - '-'-.. h. ',.. _ -_..._._....._._._ -._.-._.-._ --._._ -_._--._...._ '---...,,... _.._.~_._._._.- ~..--....._~...-- ~--,..._._--,.. _--"--_ _--,.._- _.._-_..---.._-_..-_M -_.. " _..._ ~._.,--_.----... -._-._- _.._- _... -._ -- - _ --...- -._ -_ --- _- _._-._._-...-._..-..-.._.._...._'.._ -.._..- --,...- - _ -.....-.. M _..,.... n......... - -.._ -....... - _.... _.._ _M.....-... -_ -..,.._ - - -..... - _- ~._....._._. _.._._ --_.._..,---... -.. -_ -_.._.. -.._ -,_.--.._.._.._..._.._._-_. --_._.-. --.._-.._._......._._.._._. --_._-_._..-_._._-_...._.._._ -_.._.._._...- ~"-..._._-_._._-,._._-_..._...._-,_._.._~--_._--_._--_...-..._-,._.._.._.._- _-_..._--..._-_..-..._. -_.-- --,._._-_ --_..._-.._-._...._-._.._-.-- '_._-.--_.- _.._...._._-_.._-... -._-.._.-,_...~.._.~.._--_..... -._.._-_......._-._. -.._-.._. _..._.._-..._-,.._..._-,.._.-.._-_.._- _._ - --....-.. _-,_.._._. --.._._-- _-.._.. -._.._.-..-.._.. --_.._-..- -_._-..-._.'.._.._-.._-...._ ---_.. -._.-..--,.._.._.-.,......-.._.-...'_-_'''_.-.._._-.._..-._.. _._..._ o_.._._, _,.. _.._.. _. _._- _..-._'''-.--..-..-.-"-..-,--..-._- _.- -.._-...-..-.._._..._..,. -_.,- '_."-..._-_ -_.'_._.._-_._- -...,_._._.._,._..._,..._._,_....-...._...._,_._--.._._.._.._,_._...~. -- _.._... _ --_,- _.._._-... -..._-_.._.._--~ -.._._- _..._.._._ - _- _._.......~..".-... - ~_..._~-_._.-.._- - _-- - _.- _.._..-.- -._. --..~.._-,.._- _.._.~-,_- _.- _ -_.-._.-.._-.._.. -._--._-,-_.._--. -..._- -.- _---.._.._- - _--.._.-.-.._-,,--..."._..-..._--...-.._-- _-_u..-.- ~-.._-~- _ "._,,_.. _-.._.. -.._.._-.~...,.-..- _.- -.._..- _- -.'.- ~--.- _.., ~ _--,,----. -..-.._ -.~..-.-.. ~.-..-..-.-.. -.._- -._..-.-._-- _ - -._._- -.- - _--...._.._- --'-"-- - - q---_.._'.- - _." _-.-..._.-..-......_- _. _ "0_'-.. _._..._. _._"._ "_ -- - -._.-- -._._-_._...._..-.- -.._-._-..- -..._... _.-..,. '''_' "_.. _... -.._..._- _.- _- -._.._. _ -..._. - _...... -..._. -..._ -.._ '_"_' -"_ - _ ::::::. :::::::: _.-:;.::::: :: ::::::. :::::::. -.._.-:: :::::::.._ -"._._ :::::::=~:: ::::.._ -:: :::;: :: :::::: -:: ::-:"";..~ ~~.._.._._.- _'..-_. -.._. -.._. - -..._.-..._..-._.-._.u._._._.., - _ -- "" --_..._._._- _. _.~._..-._._.-._- - _ ~._..,......_.-_ ~. ~...,.._~ _---- _.., _- _ '"._..,. _ -....- u. ~....._._.... _._.._._ '._.~.. ~ "~""'_ M ~.. ~ _ _ ~ _._._. _.._. _ ~.~.. _ --._ - _... _ -..- _ -"_.-.._ ~- -.._-.- _. u... '--,"" t _...'.,_, _..<. ::=:~::::=:::: -. -....:-~.:'::::~..::;:~::::.-:.::::,:::::.:. -..,...." I... ti......h.o. h... ",.. "Ol "._, :::::: ".."... ~ _ ~.. - _ ".~. "..... - _ ~ _... _... _.".. ~ '.... _-,_ -._ -.._'-.._.-.._..-..._. " -", _. _. 'It _... ~ ",~. '... h \._... _ -.... ~..-_ ".._..-...... -_._._..--.._.._._,... - _ - - - -..._._...,,_._._. --._-_._ -- '.-.-.".~ -..--._ ~.~.._..-..-.._.... - -..- -..._._ -,._'''_._-_.._.._... -_-..,_ - _ -'_.-..- --.._'-._ - _ - - - -.._.._-_._- -_._ --... -..,._._-- -,......._-..-.~-_..--......-......,.....:7:":'::-:;':~!;'''''"., ",., """""1, ".,.. :;:::::::::::::::::::::::::::::" 10 " ' 'H" UI.. Fig. 5. The Bitter pattern with interference. The line length on the tape is 325 urn.

Philips tech. Rev. 40, No. 5 TRACKS ON VIDEO TAPE 131 À = dsin8 = 0.87d, as illustrated in fig. 2. The wavelengths of the light that makes the Bitter pattern visible in the microscope therefore vary from 0.69 IJ.m, for the lowest level of the line-synchronization pulse, to 0.46 IJ.m, for the highest level W. In the microscope image the synchronization pulses are thus observed as red areas, and the 'white' parts of the signal are observed as violet areas, and all the levels in between appear as other colours. The colours observed in the Bitter pattern are therefore completely unrelated to the colours in the corresponding television picture, but they are related to the value of the luminance. The video signal in colour television contains the information about the colour value or chrominance and the brightness or luminance Ey. This quantity E; is a 'weighted mean' of the local intensities ER, EG and EB of the red, green and blue components of the television picture [4J, as expressed in the well-known relation Fig. 7. The test pattern transmitted by the Dutch television stations. The photographs in figs 4, 5 and 6 give the video tracks corresponding to this pattern. E; = 0.30ER + 0.59EG + 0.11 EB. The information about the individual quantities ER, EG and EB is contained in the chrominance signal, whose carrier signal at 4.43 MHz is converted to 625 khz in the VR 2020 video recorder, so that the chrominance is not visible in the interference pattern. Fig. 4 shows the Bitter pattern observed in the normal way, without interference, at the - rather highmagnification of 700 X. Fig. 5 shows the Bitter pattern observed by means of interference and at a much lower magnification, 30 X ; here both the start and the end of the tracks written on the tape are visible. Fig. 6 shows a detail of this pattern, at a magnification of 50 X. The video signalof these photographs relates to the test pattern transmitted by the Dutch television stations; see fig. 7. To clarify figs 4, 5 and 6, the method of writing the video signalon the magnetic tape in the VR 2020 recorder will now be briefly described. The method follows the VIDEO 2000 system. The video signal is written on the tape in the form of oblique tracks by two magnetic heads rotating at high speed [5]. This is done by making the tape travel around a drum in a helical path through an angle of 186 at 2.44 cm/s. The part of the drum carrying the two magnetic heads, mounted diametrically opposite, rotates at a circumferential velocity of 5.08 mis. In the VR 2020 recorder the two halves of the 12.7 mm (half-inch) wide tape are written one after the other (reversible cassette, maximum playing time 480 min). In fig. 5 half of the width of the tape can therefore be seen. To prevent luminance-signal crosstalk between two successive tracks, the write gaps of the magnetic heads are not located perpendicular to the direction of travel (the gaps are 0.5 IJ.m long [6] and 24 IJ.m wide). The gaps are mounted at an angle of 15 (the azimuth angle) to the normal to the direction of travel, with the two gaps rotated in opposite directions (see fig. 4). The tracks can therefore be written on the tape without an unused area between them and have a width of slightly less than 24 IJ.m. Fig. 4 also shows that the width of two successive tracks is practically identical, indicating that the head servosystem for writing operates correctly (this will be briefly discussed later). Each oblique track contains the information relating to the field formed by the odd (or even) lines of a television picture. In each second the electron beam traces out 50 fields or 25 complete frames on the screen of the picture tube. The field consisting of the even lines starts with a half-line, the field with the odd lines ends with a half-line. In the VIDEO 2000 system the tracks are written on the tape in such a way that the start of each track is always displaced by one and a half lines with respect to the previous one; seefig. 8. Fig. 6. Detail of fig. 5. o 200 400 pm [4J R. Theile, Fernsehtechnik; Springer, Berlin 1973. [SJ H. Bahr, Alles über Video; Philips Fachbücher, 1980. [6J It is customary to call the dimension of the gap in the direction of the tape travel the 'length', even though the other dimension of the gap is much larger. BIBLIOTHEEK NAT. LAB. N.V P 'llips' GLOEILA. ~~AB""rE<EN POS I.. S Po.r' 5600 JA L OHOVEIII

132 TRACKS ON VIDEO TAPE Philips tech. Rev. 40, No. 5 The result is that the line-synchronization pulses (seen from the writing head) are always side by side and an adjacent synchronization pulse cannot interfere with the picture information. In figs 5 and 6 the line-synchronization pulses are red areas, grouped along an approximately vertical line. The straightness of these lines shows that the other servosystems controlling the mechanical operation of the VR 2020 during the writing process operate satisfactorily. In our experimental arrangement (see fig. 1) the incident light beam is given a direction perpendicular to the lines of the Bitter patterns relating to one magnetic head; see fig. 4. The light is reflected by the Bitter patterns from the other head in such a way that it does not enter the entrance diaphragm of the microscope objective. In figs 5 and 6 the Bitter patterns from the other head thus form black lines, so that these figures only show the tracks relating to one magnetic head. The relation between the pictures in figs 5 and 6 and the corresponding test pattern in fig. 7 should now be clearer. Since identical video signals from the stationary test-pattern picture are repeated at a frequency of 25 Hz, the information relating to one test frame is also present in a direction perpendicular to the track, i.e. approximately in the vertical direction in the photographs. Since the adjacent tracks are displaced by one and a half lines, the Bitter patterns in this direction show (for example) the information relating to the 1st, 7th, 13th, 19th,..., 625th line; the 3rd, 9th, 15th,..., 621st line; the 5th, 11th, 17th,..., 623rd line, and so on; see fig. 8. The test pattern can thus be recognized in the vertical direction in fig. 5, although greatly distorted. As stated at the beginning, the colours do not correspond to those in the test pattern, but to the magnitude of the luminance signal. Fig. 8. The start of the oblique video tracks, with the numbers of the odd lines of successive television frames. The order of writing the tracks on the tape is from right to left in this figure (and also in figs 4, 5 and 6). The heads, however, move from left to right in relation to these figures. In the upper part of fig. 5 patterns can be seen that consist of a series of five red areas corresponding to a total duration of 2i lines. Together these successive areas correspond to the field-synchronization pulse, which is used for synchronizing the flyback of the electron beam in the picture tube after writing a complete field. Below these pulses there are greenish areas with a duration of 1i lines, followed by a black area of the same length. The green areas correspond to a signalof 223 khz, superimposed on a frequency of 3.6 MHz (the level B in fig. 3), which is used for the head servosystem for writing mentioned earlier. If no information is written on the tape (the black area), the magnetic head reads the crosstalk originating from the 223 khz signalon the previous track, and stores its amplitude temporarily in a memory. When the next track is written the same procedure takes place, and then the two crosstalk signals are compared. The position of one of the magnetic heads is then corrected by the appropriate piezoelectric element in such a way that the 223 khz crosstalk signals of the successive tracks are equal. As noted earlier, this produces tracks on the tape that are virtually constant in width.

Philips tech. Rev. 40,1982, No. 5 133 Diamond die The photograph shows a model of a single-crystal diamond die, about 4 cm in diameter. The real diamond, a natural product, is about 40 times smaller. Models like the one shown are used in the Philips diamond die factory at Valkenswaard (in the Netherlands) for demonstration and instructional purposes. The passage for drawing the wire can be seen. On both sides it has an accurately defined 'bell' shape, which determines the ratio of the diameter of the wire before passing through the die to the diameter after it emerges. In the situation shown the wire would pass through the die from top to bottom. The diamonds used for the dies have weights between 0.05 and 3 carats. The corresponding wire diameters after drawing are 8 urn to more than 2 mm. (For making wires of diameter greater than 0.12 mm synthetic diamonds are often used today, in a sintered and hence polycrystalline form.) The polished surface visible in the right foreground of the photograph is used as an inspection window during the production of the passage. In recent years laser drilling and mierospark machining have been increasingly used in the manufacture of diamond dies. We shall shortly publish an article on a number of special applications of spark machining, including the 'sparking' of diamond.