EP0003712B1 - Process for reducing the magnetic perturbations in rows of high amperage electrolytic cells - Google Patents

Description

The present invention relates to a new method for reducing magnetic disturbances in the series of long, high-intensity electrolytic cells, intended for the production of aluminum by electrolysis of alumina dissolved in molten cryolite. It applies to the reduction of disturbances due to the own field created by each tank and by its neighbors in the same queue and the adjacent queue when the latter is at a distance relatively close to the queue considered.

We know that, to reduce investments and increase yields, the trend is to increase the power of the tanks, which, supplied with 100,000 amps twenty years ago, now reach 200,000 amps. It is also known that the tanks arranged transversely to the axis of the line have smaller magnetic effects than the tanks arranged lengthwise, despite the complication of operating conditions and the deterioration of resulting work. From this point of view, longitudinal tanks do not have these drawbacks, and the aim of the invention is a method making it possible to reduce the magnetic effects of longitudinal tanks to a level lower than that of transverse tanks, hence it results in considerable energy savings while retaining the operating advantages due to the long layout.

In all that follows, we will designate, according to the usual conventions, by B _x , By and B _z the components of the magnetic field along the axes Ox, Oy and Oz, in a direct right-angled trihedron whose center O is the center of the cathode plane of the tank, Ox is the longitudinal axis in the direction of the file, Oy, the transverse axis and Oz the vertical axis directed upwards.

According to the usual convention, the upstream and downstream positions are designated by reference to the conventional direction of the current in the series.

• Figure 1 shows in vertical cross section, passing through point 0, a long electrolysis tank.
• Figure 2 shows, in schematic horizontal section, passing through point 0, a long electrolysis tank.
• Figure 3 is a diagram of the field B _z along a long side of the tank.
• FIG. 4 represents the shape of the metal-electrolyte interface according to the distribution of the field B _z of FIG. 3.
• Figures 5 and 6 schematically show two possible arrangements for feeding the tanks lengthwise: by a head and central risers (Figure 5) or by the two heads (Figure 6).
• Figures 7, 8, 9 and 10 show the position of the negative conductor according to the invention, as a function of the coefficient α, fraction of current supplied to the upstream head.
• Figure 11 explains the influence of the field of the neighboring queue on the tatal field of a given tank on its minor axis Oy.
• FIG. 12 shows the position of the compensation conductors for the field of the neighboring line in an electrolysis room comprising two relatively close lines.
• Figures 13, 14 and 15 show the position of the negative conductor as a function of the coefficient α when the influence of the neighboring queue is taken into account.

Statement of the problem

dans lequel By_" est la compcsante horizontale du champ magnétique selon l'axe Oy (axe horizontalFrench patent 1 143879, filed on February 28, 1956, in the name of "PECHINEY" gave the conditions to be respected to reduce the magnetic effects in long tanks, and most of the tanks built in the world, since that time, have used the provisions of the recommended conductors, in order to satisfy the double condition:

in which By _" is the horizontal component of the magnetic field along the Oy axis (horizontal axis

perpendicular to the line axis) and dB _Y ° is the gradient of this potential along the vertical axis at the center dz of the chip.

In all that follows, reference will be made to the diagram of an electrolysis cell, as it appears in FIG. 1 in cross section.

However, the conditions recommended in patent FR 1 143 879 only concerned the horizontal field and had no action on the vertical field, the value of which is practically proportional to the intensity passing through the tank.

However, all recent studies show the importance of this vertical field. It is responsible, in particular, for a “domed” deformation of the sheet of liquid aluminum, an asymmetrical dome, the top of which is offset towards the downstream head of the tank, and which corresponds to a drop in height which can exceed 4 centimeters per report to the reference plane.

• Forces selon l'axe Ox: f(x) = j_yB_z- j_zB_y
• Forces selon l'axe Oy: f(y) = j_zB_x―j_xB_z;
• B_x, By et B_z étant les trois composantes du champ magnétique B selon les axes Ox, Oy, Oz, j_x, jy et j_z étant les trois composantes de la densité de courant dans le métal.

The forces, called Laplace forces, which develop in metal are the source of the deformation of the bath-metal interface.

• Forces along the Ox axis: f (x) = j _y B _z - j _z B _y
• Forces along the axis Oy: f (y) = j _z B _x ―j _x B _z ;
• B _x , By and B _z being the three components of the magnetic field B along the axes Ox, Oy, Oz, j _x , jy and j _z being the three components of the current density in the metal.

The presentation of the solution which the invention brings to the problem of magnetic effects will be facilitated by an analysis of the different components of these forces.

Let us consider the horizontal section of a long tank, figure 2, at the level of the central point 0 which one divides in four quarters by the axes Ox and Oy, and first of all determine the longitudinal forces according to the parallels to Ox.

car jy est constant en raison de la disposition habituelle des barres cathodiques à sorties transversales; il en est de même de j_z.The set of forces f ₁ (x) on a parallel to Ox (of abscissa y) in the first quarter is:

because jy is constant due to the usual arrangement of cathode bars with transverse outputs; the same is true of j _z .

We will have the same in the second quarter:

et que

Sif ₁ (x) = -F ₂ (x), the forces on each parallel to Ox will be equal and opposite. For this it suffices that:

and

These two conditions are met if the curves B _z and By are asymmetrical with respect to the axis Oy.

Case of the vertical field B _z :

in a long vessel, the curve of B _z on each parallel to Ox is antisymmetrical with respect to its value at its midpoint as seen in FIG. 3. So it suffices to make B _z zero on the axis Oy, so that the set of B _l is antisymmetric with respect to Oy. At the center O of the tank, B _z (o) is then zero by symmetry. B _z is maximum on the parallel to Ox passing through the outer edge of the anode system and if we cancel B _z at point M, the curve of maximum B _z will also be asymmetric.

If B _z (M) and B _z (o) are zero, the value of B _l on the Oy axis does not exceed 2 to 3. 10- ⁴ TESLA, for a tank of 100,000 amperes, which is negligible.

Therefore, the values of B _z at all points located symmetrically with respect to Oy have equal values and of opposite sign, and the curves of B _z on each parallel to Ox will be asymmetric.

Case of the horizontal field By:

we keep the condition By (at the central point O) = 0 already stated in patent FR 1 143 879. We note that, when By (o) = 0, the values of By on the axes parallel to Oy are minimal and very low .

The By curve on each axis is then also asymmetric.

et

• Σ F1 (x) dans le premier quart de la cuve,
• = - Σ F₂ (x) dans le deuxième quart de la cuve.

In total, when the two conditions B _z (M) and By (o) = 0 are fulfilled, we see that on each parallel to Ox:

and

• Σ F1 (x) in the first quarter of the tank,
• = - Σ F ₂ (x) in the second quarter of the tank.

The equality of forces leads to a bath-metal interface in the form of a symmetrical dome with minimal deflection.

FIG. 4 represents, in the case of a conventional tank of 115,000 amperes, and in correspondence with FIG. 3: in solid lines, the asymmetrical dome shape, and large arrow (up to 4 cm) in the case of '' an asymmetrical curve B _z (solid line) and the symmetrical dome shape with a small arrow (about 1 cm) in the case where B _z is asymmetrical with respect to the axis Oy, that is to say after setting in work of the invention.

The asymmetric force comes, in the first case, from the fact that the set of positive forces from R to P, F ₁ (x) is about three times larger than the set of negative forces - F ₂ (x), from Not.

Now determine the transverse forces, according to the parallels to Oy with the same conventions as for F (x) we have:

• j_x, dans une cuve bien construite est nul,
• j_z est constant

These transverse forces are much weaker than the longitudinal forces F (x) because they are exerted on shorter lengths (width of the tank). Gold:

• j _x , in a well constructed tank is zero,
• j _z is constant

et que Σ F₁ (y) sur le premier quart de la cuve = - ΣF4 (y) sur le quatrième quart de la cuve.In a long vessel, the conductors of which are usually arranged symmetrically with respect to the x Oz plane, Bx is asymmetric with respect to Ox. It follows that:

and that Σ F ₁ (y) on the first quarter of the tank = - ΣF4 (y) on the fourth quarter of the tank.

• By(o)etB_z(M)=0
  • 1/ On diminue la valeur des champs maximaux qui, pour B_z et By sont situés à la périphérie de la cuve.
  • 2/ Les forces de Laplace seront minimales, égales et opposées par rapport aux axes Ox et Oy.
  • 3/ Il en résultera une surface de l'interface électrolyte-nappe d'aluminium liquide, stable et pratiquement horizontale.

We come to the conclusion that, if we build a tank in which:

• By (o) and B _z (M) = 0
  • 1 / We decrease the value of the maximum fields which, for B _z and By are located at the periphery of the tank.
  • 2 / Laplace's forces will be minimal, equal and opposite with respect to the Ox and Oy axes.
  • 3 / This will result in a surface of the electrolyte-liquid aluminum sheet interface, stable and practically horizontal.

It may also be advantageous to obtain the fulfillment of an additional condition, relating to the component By in the center, that is:

Although this gradient is generally quite low, we can try to make it as close as possible to zero as far as the different conditions are compatible with each other, which effectively amounts to canceling By throughout the thickness of the sheet of liquid metal, which is weak and which varies only a few centimeters from its average level.

Statement of the invention

caractérisé par une disposition particulière des conducteurs de liaison entre les différentes cuves d'une série dans laquelle elles sont disposées en long, tenant compte en outre, de l'influence du champ magnétique d'une file voisine lorsque les deux files sont situées à une distance suffisamment réduite pour qu'on ne puisse plus négliger cette influence. L'invention s'applique aux cuves en long alimentées soit par les deux têtes, soit par la tête amont et au moins une montée latérale sur chaque côté.The problem to be solved being posed, the object of the invention is a method for realizing the two conditions By (o) and B _z (M) = 0 and possibly the third

characterized by a particular arrangement of the connecting conductors between the different tanks of a series in which they are arranged lengthwise, taking account also of the influence of the magnetic field of a neighboring line when the two lines are located at a distance sufficiently small so that this influence can no longer be overlooked. The invention applies to long tanks fed either by the two heads or by the upstream head and at least one lateral rise on each side.

In general, the negative conductors (connecting conductors between the tanks) are arranged symmetrically with respect to the median plane xOz. Referring to Figure 1, denote by Y and Z the coordinates of these conductors in the xOz plane.

The method is then characterized in that the negative conductors are arranged parallel to the axis Ox and pass substantially through points whose coordinates Y and Z satisfying a first system of equation, make it possible to achieve the two conditions By (o) = 0 and B _z (M) = 0, or, which amounts to the same B _z (M) antisymmetric with respect to the axis Oy; it is also characterized in that one endeavors that the coordinates Y and Z satisfy, at least approximately with a third equation, allowing to realize the additional condition -

ou tout au moins, de s'en approcher autant que possible dans la mesure ou les solutions sont compatibles entre elles.

or at least to get as close as possible to the extent that the solutions are compatible with each other.

Finally, it is characterized in that, in addition to the preceding conditions, the field of the neighboring queue is compensated by means of an auxiliary conductor disposed along each queue in which a direct current is circulated in the opposite direction to that of the circulating current in line, and whose intensity is provided by the resolution of a system of equations taking into account the various magnetic influences on each tank.

We will successively examine the case of a series comprising two rows of tanks far enough apart not to introduce an effect of neighboring queue, then the case where there is an effect of neighboring queue, and in both cases, we will distinguish the tanks classics supplied by the two heads and the tanks supplied by the upstream head and by central risers, as described in French patent No. 2,378,107, published on August 18, 1978, in the name of the applicant and whose the structure is recalled in FIG. 5, in the case of a central climb on each side.

We agree to call x the current fraction supplying the upstream head A and (1 -α) the current fraction supplying the downstream head B or the side risers, as appropriate (Figures 5 and 6).

We will now determine the position of the negative conductors, a function of the parameter α.

A/ Cas d'une serie comportant une seule file ou deux files suffisamment eloignees pour qu'il n'y ait pas d'interaction magnetique entre ellesIn order to simplify the calculations, we seek what must be the constant intensity, in a conductor, which put in the place of the cross on the one hand, and of each negative collector on the other hand, would create the same magnetic field as them. We then find the so-called equivalent intensities appearing in Table 1 below, which are valid only for points located in the median plane yOz.

A / Case of a series comprising a single file or two files sufficiently far apart so that there is no magnetic interaction between them

Case 1 Tanks fed by a head and central risers (fig. 5) 1 ° / Realization of the condition By ° = 0

By°= dû au croisillon:

• formule dans laquelle k₁ est un coefficient expérimental qui tient compte du fait que le croisillon est formé, en pratique, de deux branches et de la discontinuité duejà l'intervalle entre les croisillons de chaque cuve d'une file. k₁ est presque toujours voisin de 0,9 et nous conserverons cette valeur par la suite.

By ° = due to the cross:

• formula in which k ₁ is an experimental coefficient which takes account of the fact that the cross is formed, in practice, of two branches and of the discontinuity duejj to the interval between the crosses of each tank of a file. k ₁ is almost always close to 0.9 and we will keep this value later.

h is the height of the spider above the xOy reference plane.

by due to negative collector 1 = by ° due to negative collector 2 =

d'où l'on tire:

2°) Réalisation de la condition B_z(M) = 0

b_z (M) dû au collecteur 1 :

b_z(M) dû au collecteur 2:

By ° is equal to: byo (cross) + b _{y °} (collector 1) + b _{y °} (collector 2) and must be equal to 0.

where we get:

2 °) Realization of the condition B _z (M) = 0

b _z (M) due to collector 1:

b _z (M) due to collector 2:

ou en effectuant:

The condition B _z (M) = 0 is written: b _z (M) crosspiece + b _z (M) collector 1 + b _z (M) collector 2 = 0, that is to say:

or by performing:

We note that 1 has disappeared, which proves that the solution will be independent of the intensity passing through the tank.

qui permet d'obtenir la valeur de Z que l'on reporte dans l'équation (2) pour obtenir la valeur de Y

Using the value of y ² from equation (2) we obtain the second degree equation in Z

which makes it possible to obtain the value of Z which is carried over into equation (2) to obtain the value of Y

by° dû aux collecteurs négatifs:

We have: By ° due to the cross:

by ° due to negative collectors:

En posant, comme précédemment

et, en la combinant aux équations (2) et (5), on en déduit que

We must have: dby (crosspiece) + 2 dby (collectors) = 0

By posing, as before

and, by combining it with equations (2) and (5), we deduce that

We note that

• 1 - µ and v are a function of the parameter α, defined above. We therefore obtain the solutions, that is to say the values of Z and Y, in the form of a curve which is the geometrical place of the position of the negative conductors, which fulfill the conditions set at the start.
• 2 - Equations (5) and (6) are independent of the intensity, insofar as a and h which occur in "and ν are constant. In fact, »h« height of the spider does not depend on the size of the tanks and »a« half-width of the anode system may not vary if the size of the tanks is increased by simply lengthening the system anodic according to Ox. In practice, "a" varies very little, and is worth, for example, 1.20 meters for a tank of 100,000 amps and 1.50 meters for a tank of 200,000 amps. Beyond that, for technological reasons, we no longer increase "a".

Examples of practical achievements Example 1

The above results were applied to a series of 100,000 amp cuvettes, with central risers, for which h (height of the spider above the xOy plane) = 1.77 m and a (half width of the anode system ) = 1.175 and comprising on each long side, 11 cathode output bars. Under these conditions, α can therefore only vary by fraction of 1/11. However, it is also possible to envisage continuous variations in the value of α by varying the electrical resistance of the connecting conductors. Six values of α were considered: 2/11 (0.182), 3/11 (0.273), 4/11 (0.364), 5/11 (0.455), 6/11 (0.545) and 7/11 (0.636).

The application of formulas 1 to 5 gives the following results:

N.B. The sign ± means that this rating is valid for negative conductors placed on each side of the tank.

These values have been plotted on the diagram, FIG. 7 which is, in fact, a transverse half-soup ¹ of a tank in length, passing through the central point O. The lines bordered by hatching indicate the external dimensions of the box of tank.

It is noted that for values of α greater than 0.55, the conductors should be inside the tank. For economic considerations and for space, we can therefore choose α between 0.35 and 0.55.

et voir si les solutions sont compatibles entres elles, tout au moins de façon approchée, et si les trois conditions peuvent être réalisées simultanément:We will now seek to carry out simultaneously:

and see if the solutions are compatible with each other, at least in an approximate way, and if the three conditions can be realized simultaneously:

The condition B _{y °} = 0 implies

soit en résolvant:

By carrying the value of Y ² in (6), it comes:

either by solving:

By introducing this value of Z into (6A), we find Y.

= 0 et By° et d'autre part B_l (M) = 0 et B_y° = 0 et voir si leur intersection donne, pour la position du conducteur, une valeur acceptable.We can therefore draw, on a diagram, the curves (Y, Z) = f (α) which satisfy on the one hand

= 0 and By ° and on the other hand B _l (M) = 0 and B _{y °} = 0 and see if their intersection gives, for the position of the conductor, an acceptable value.

Example 2

We seek to simultaneously realize the three conditions:

We add to table I, to facilitate the drawing of the curve Y, Z = f (x) satisfying B _z (M) = 0 and By ° ≠ 0, a value of α = 0.6 which gives Y = 1.86 and Z = -0.98.

Then we draw the curve Y, Z = f (α), satisfying the two equations

We see that the two curves are suppressed at a point with coordinates Y = 1.96, Z = -1.01.

This point corresponds to a position of the conductor inside d: box, but in practice, we can take a neighboring point, outside; the two corbs only moving apart slowly, the solution is still acceptable.

Example 3

The same results were applied to a 200,000 amp vessel, for which h = 1.77 m, a = 1.50 m also having 11 cathode bars on each long side. The following values are obtained:

On the diagram, figure 9, one notes that the curves of Y, Z = f (x) are very close for 100,000 amps and 200,000 amps, but that the value α = 0,546 leads to a geometric impossibility.

The two curves would moreover coincide if the increase in intensity had been obtained by the elongation of the cathode alone. In reality, the anodes have also been enlarged (factor "a"). Otherwise, a and h being constant, the same would be the case for µ and v and equation (5) being independent of the intensity, the curves Y, Z = f α) would be identical.

Case 2 Conventional tanks (Figure 6) fed by the two heads of the spider. No neighboring queue Head A (upstream) receives an intensity α l, head B (downstream) receives (1 -α) 1

Les équations (10) et (11) sont identiques à (2) et (5), soit:

The calculation is identical, but the equivalent intensities (see Table I) of the spider and the negative collectors being different, the new values of µ 'and v' are as follows:

Equations (10) and (11) are identical to (2) and (5), namely:

Example 4

Based on the above result, the position of the negative conductors was determined for a conventional 100,000 amp vessel, with a cross supplied by the two heads with h = 1.77 m and a = 1.175 and also comprising 11 outlet bars. cathodic on each long side.

These values are plotted on the diagram, figure 10; we note that α = 0.8 leads to an impossibility, and that, for economic reasons, we are led to choose α between 0.65 and 0.75.

B / Case of a series comprising two adjacent lines in the same building

In practice, the economical solution is to install the two rows of tanks in the same building.

car b_z (M) cuve = -b_z (N) cuve, puisque b_l est anisymétrique par rapport à Oy, alors que b_z (M) (file voisine) est sensiblement égal à b_z (N) (file voisine) et de même signe.We then introduce a vertical field due to the neighboring file, fairly uniform in value and of the same sign. If we call b _z (tank) the vertical field of the tank with its file, and b _z (neighboring file) the vertical field induced by the neighboring file, we see, on the diagram in figure 11, that it is impossible simultaneously obtain:

because b _z (M) tank = -b _z (N) tank, since b _l is anisymmetric with respect to Oy, while b _z (M) (neighboring file) is substantially equal to b _z (N) (neighboring file) and the same sign.

In this figure 11, curve 1 represents the variation of b _z (tank) with no neighboring line, along MON, curve 2 represents the variation of b _z (neighboring file) along MON, and curve 3 represents the variation of b _z (tank + neighboring file) along MON

It is therefore necessary to compensate for the effect of the neighboring queue.

A number of solutions have already been proposed for this, for example; those which are the subject of French patents n ° 1 079131, filed on April 7, 1953 (Compagnie PECHINEY), in which it is proposed to create an electric loop around the tank, n ° 1 185 548 filed on 29 Octrobre 1957 ( Electrokemisk AS) in which an asymmetrical feeding of the heads A and B is carried out, as well as in French patent n ° 1,586,867, filed on June 28, 1968 (Vsesojuzny Nauchnoissledovatelsky i Proektny Institut Aluminievoi) or in French patent n ° 2,333,060 deposited on November 28, 1975 (Aluminum Péchiney) where it is proposed to position the negative collectors differentially on each side of the tank.

These last two solutions significantly reduce the effect of neighboring file, but not uniformly over the entire tank. On the other hand, by (tank) and b _z (tank) are no longer antisymmetric compared to Ox. We thus create an asymmetry of the Laplace forces in Fy.

The method which we will now describe uses a compensation conductor traversed by a current -i which flows in the opposite direction to the current 1 of the series, located on the outside of the two rows of tanks, and placed at a minimum distance from the box, compatible with electrical safety.

An analogous method has already been described in American patent US Pat. No. 3,616,317. However, it should be made compatible with the conditions set out above, and in particular with obtaining B _{y °} = 0 and B _z (M) = 0. Figure 12 is a schematic sectional view of an electrolysis room (1) comprising two adjacent rows, of which only the anode systems 2 and 3 have been shown, and in which the cells are arranged lengthwise. The compensation conductors are in 4 and 5. The connection conductors of the tanks are omitted so as not to overload the drawing. The rounded arrows schematize the directions of the magnetic field created by each compensating conductor.

• a: la demi-largeur du système anodique
• d: la distance du conducteur de compensation au bord extérieur de l'anode
• I: la distance entre les bords intérieurs des systèmes anodiques des deux files de cuves
• E: lechamp produit par le conducteur de compensation en M (côté intérieur)
• F: le champ produit par le conducteur de compensation en N (côté extérieur)
• e: le champ de la file voise en M
• f: le champ de la file voisine en N
• m: le champ bz de la cuve sans effet de file voisine

We call:

• a: the half-width of the anode system
• d: the distance from the compensation conductor to the outer edge of the anode
• I: the distance between the interior edges of the anode systems of the two rows of tanks
• E: the field produced by the compensation conductor in M (interior side)
• F: the field produced by the compensation conductor at N (outside side)
• e: the queue field in M
• f: the field of the neighboring queue in N
• m: the bz field of the tank with no neighboring line effect

On en déduit un rapport K

Taking into account, the two compensation conductors:

We deduce a report K

1 and de are values fixed at construction and practically independent of the size of the tanks.

K varies with »a« but rather weakly, as shown by equation (12c). For a series of tanks whose distance between anodes I = 7.40 m and whose distance between the compensation conductor and the outer edge of the anode d = 1.80 m, we find:

For a tank of 100,000 amperes, where a = 1.175, identified with that of Examples 1 and 3: K = 0.52.

For a 200,000 amp vessel, where a = 1.5, identical to that of Example 2, K = 0.47.

We can therefore base ourselves on a value of K = 0.5.

We will now choose a tank diagram for which the value of the vertical field b _z (with no neighboring file) is m at point M, and therefore - m at point N.

• m+e+KF = 0
• -m+f+F = 0

D'où les valeurs:

e et f sont directement proportionnels à i, et les valeurs de m, E et F seront donc également proportionnelles à i. On peut donc définier 3 caractéristiques:

valables pour toutes les cuves disposées en long, en prenant 1, par commodité, exprimé en kilo- ampères.We determine M by the equations:

• m + e + KF = 0
• -m + f + F = 0

Hence the values:

e and f are directly proportional to i, and the values of m, E and F will therefore also be proportional to i. We can therefore define 3 characteristics:

valid for all tanks arranged lengthwise, taking 1, for convenience, expressed in kilo-amperes.

K being close to 0.5 the equations can be written, in practice:

Practical applications Example 5

• e = 24,4 . 10-⁴ Tesla
• f = 18,9 . 10^-4 Tesla

on a:

D'où i', intensité dans le conducteur de compensation, d'après l'équation (12a) ou (12b) = 226A/1000A soit 22,6% de I.For a series of 100,000 amp cells, identical to those of examples 1 and 2, the values of the vertical field due to the neighboring line in M and N, with 1 = 7.40 m and d = 1.80 m are respectively (measured values):

• e = 24.4. 10- ⁴ Tesla
• f = 18.9. 10 ^-4 Tesla

we have:

Hence i ', intensity in the compensating conductor, according to equation (12a) or (12b) = 226A / 1000A or 22.6% of I.

• B_z (M) = B_z(N) = 0
• En M = - 9,6 . 10^-4 T - 15 . 10^-4 T + 24,4 . 10^-4 T = 0
• En N = +9,6 . 10-4T - 28.4 . 10-4T + 18,9 X 10-⁴T = ₀

We check that at points M and N, we have:

• B _z (M) = B _z (N) = 0
• In M = - 9.6. 10 ^-4 T - 15. 10 ^-4 T + 24.4. 10 ^-4 T = 0
• At N = +9.6. 10-4T - 28.4. 10-4T + 18.9 X 10- ⁴ T = ₀

Example 6

With the same tanks as in Examples 1 and 3, the distance d was reduced to 1.20 m (this reduction is only possible if the arrangement of the tanks allows it). This results in a significant reduction in the intensity i in the compensation conductor.

• K = 0,41
• m' = -0,1184 - 10^-4T/1000A
• F' = -0,307 - 10^-4T/1000A
• E' = -0,126 · 10^-4T/1000A
• i' = 169A/1000A soit 16,9%

We have:

• K = 0.41
• m '= -0.1184 - 10 ^-4 T / 1000A
• F '= -0.307 - 10 ^-4 T / 1000A
• E '= -0.12610 ^-4 T / 1000A
• i '= 169A / 1000A or 16.9%

• En M = - 11,8 - 1 2,6 + 24,4 = 0
• En N = +11,8-30,7+18,9 = 0

We verify that at points M and N, we have B _z (M) = B _z (N) = 0

• In M = - 11.8 - 1 2.6 + 24.4 = 0
• In N = + 11.8-30.7 + 18.9 = 0

In conclusion

on obtiendra un champ total B_z nul en utilisant un conducteur de compensation parcouru par un courant i qui sera fonction de sa distance au bord extérieur de l'anode.If we make a tank whose field m 'by 1000 amps at point M is equal to

a zero total field B _z will be obtained by using a compensation conductor traversed by a current i which will be a function of its distance from the outer edge of the anode.

When transforming an old series, for reasons of electrical safety and space, d will be close to 1.80 m and the intensity i of the compensating conductor will in this case be 22.6% of the current I of series.

In the case of a new series d may be lower because it will be located in an independent channel isolated from the series. For d = 1.20 the compensation current will only be 16.9% of current I.

This method therefore makes it possible to minimize the investment cost, and the consumption of the compensating conductor.

This compensation method will now be combined with the previously described method, aiming to make B _z (m) and By (o) zero.

Case 3 Tanks with central rise (figure 5) feeding the spider through the head A at an intensity α i by the central climbs under an intensity (1 -α) I. We take into account the neighboring file

In this case, the condition B _{y °} = 0 does not change because the neighboring file and the compensation conductor being substantially in the plane of the metal, they have no action on By. We therefore have the same equations as (2) and (5) of case n ° 1.

• On a:
• b_z (M) croisillon + b_z (M) collecteur 1 + b_z (M) collecteur 2 = m En reprenant le même calcul que dans le cas n" 1, on trouve:
  
  En posant:
  
  On a:
  

On the other hand, the condition B _z (M) is modified:

• We have:
• b _z (M) crosspiece + b _z (M) collector 1 + b _z (M) collector 2 = m Using the same calculation as in case n "1, we find:
  
  By asking:
  
  We have:
  

cette équation est identique à l'équation (5) di cas n° 1, sans file voisine, avec un nouveau coefficient ν₁ Elle permet de calculer Z que l'on reporte dans l'équation (14) pour obtenir Y.By drawing the value of Y ² from equation (14), we obtain the equation of the 2nd degree:

this equation is identical to equation (5) in case n ° 1, with no neighboring file, with a new coefficient ν ₁ It makes it possible to calculate Z which is transferred to equation (14) to obtain Y.

Example 7

• 1 = 7,40 m et un conducteur de compensation placé à une distance
• d = 1,80 m comme dans l'exemple 4.

We consider 100,000 amps identical to those of examples 1 and 3, placed in two rows per building, with a distance between anode

• 1 = 7.40 m and a compensation conductor placed at a distance
• d = 1.80 m as in Example 4.

We calculate in the same way and with the same conventions for ν

In FIG. 13, the curve I in dotted lines corresponds to the solution of case n ° 1, the curve 2 in solid line, corresponds to the solution taking into account the neighboring file. We note that, for practical and economic reasons, we will choose α between 0.35 and 0.50.

Example 8

Cells of 200,000 amperes are considered, identical to those of Examples 1 and 3, but placed in two rows per building, with a distance between anodes I = 7.40 m and a compensation conductor placed at a distance of 1.80 m as in Example 5.

We obtain in the same way, the values of y and z which are plotted on the diagram, figure 14.

Case 4 Conventional tanks (Figure 6) comprising a spider fed by the upstream head (A) at an intensity αl and by the downstream head (B), at an intensity (1 -α) l

The neighboring queue is taken into account.

The calculation is identical to that of case n ° 1, but the equivalent intensities of the spider and the negative collectors being different, we will have new values for the coefficients µ and v.

Byo n'étant pas influencé par la file voisine, ne par le conducteur de compensation situé dans le plan xOy du métal, on a:

Réalisation de la condition B_z(M): 0
La nouvelle valeur de ν'₁est:

Achievement of the condition By (o) = 0

Byo not being influenced by the neighboring line, not by the compensation conductor located in the xOy plane of the metal, we have:

Realization of condition B _z (M): 0
The new value of ν ' ₁ is:

Equation (19) giving z is the same as (16) with the new values of µ ' ₁ and ν' ₁ :

Example 9

qui sont reportées sur le diagramme, figure 15.Consider conventional tanks (Figure 6) of 100,000 amperes, identical to those of Example 4, placed in two rows, in the same building, with a distance between anodes, 1 = 7.40 m and a compensation conductor placed at a distance d = 1.80 m (as in example 5). We obtain, in the same way, as in the previous cases, the values of Y and Z:

which are plotted on the diagram, figure 15.

Conclusion

It follows from these various examples, taking into account the practical and economic considerations which lead not to pass the conductors at an exaggerated distance from the tank or in direct contact, that for the implementation of the invention one choose the current distribution coefficient α; equal to or less than 0.55 in the case of tanks fed by a head and at least one central rise on each side, and preferably between 0.45 and 0.55, equal to or less than 0.75 in the case of conventional tanks supplied by the two heads, and preferably between 0.75 and 0.65.

Claims (2)

1. A process for reducing magnetic disturbances in series of electrolysis tanks, operating at high current strength, for the production of aluminium, the tanks being disposed lengthwise, each tank being supplied with current from the preceding tank both by way of the upstream end at a fraction α of the current strength, and by way of at least one other point disposed between the upstream end and the downstream end inclusive, with a fraction (1 -α) of the current strength, characterised in that the horizontal transverse component By of the magnetic field at the centre of the tank is nullified and that the vertical component B_z of the magnetic field is made anti-symmetric with respect to the axis Oy by disposing the negative conductors in such a way that they are parallel to the axis Ox and pass through the points of co-ordinates Y and Z satisfying the two relationships:

and that the vertical gradient of the horizontal component of the magnetic field at the centre of the tank

is also nullified, by determining the co-ordinates Y and Z of the negative conductors in such a way that they satisfy as far as possible a third relationship:

the coefficients µ and v being independent of the current strength and dependent on the half-width 'a' of the anodic system, the height'h' of the cross member above the cathodic reference plane xOy, and the fraction α of the current strength which supplies the upstream end of each tank, the coefficient µ being equal to:

when the tanks are supplied by way of the two ends, and being equal to:

when the tanks are supplied by way of the upstream end and by way of at least one central riser input member, on each side, the coefficient v in both cases being equal to:

2. A process for reducing magnetic disturbances in series of electrolysis tanks operating at a high current strength, in accordance with claim 1, characterised in that the current distribution coefficient α is made:

-equal to or lower than 0.55 and preferably between 0.45 and 0.55 in the case of tanks which are supplied by way of the upstream end and at least one central riser input member on each side,

-equal to or lower than 0.75 and preferably between 0.75 and 0.65 in the case of conventional tanks which are supplied by way of the two ends.

Images