EP0324266B1

EP0324266B1 - Method for setting electrodes in aluminium electrolysis cells

Info

Publication number: EP0324266B1
Application number: EP19880312199
Authority: EP
Inventors: Odd Skaar; Kurt Nilsson
Original assignee: Norsk Hydro ASA
Current assignee: Norsk Hydro ASA
Priority date: 1987-12-30
Filing date: 1988-12-22
Publication date: 1992-09-30
Also published as: NO875479D0; AU2760688A; BR8806985A; NO162975C; DE3875099T2; NO162975B; CA1336701C; DE3875099D1; NO875479L; EP0324266A1; AU615975B2

Description

The present invention relates to a method of setting electrodes in electrolysis cells, especially setting of carbonaceous anodes in cells producing aluminium by electrolysis according to the Hall-Heroult process, as mentioned in the initial part of the present claim 1.
Aluminium is mostly produced by electolysis of aluminium oxide dissolved in a cryolite bath. The electrolysis cells allowing this, consist of a carbon cathode disposed in a steel shell which on the inside is isolated with refactory materials. Above the carbon cathode is provided a carbon anode or a number of rechangeable carbon anodes which are partly submerged in the cryolite bath and which are gradually reduced by the oxygen originating from the decomposing of the aluminium oxide.
Electric current is led from the top to the bottom of the cells, and the cryolite is kept melted by means of the Joule-effect at a temperature close to the solidification temperature. The most common operating temperatures for these cells lies between 930 and 980 C. The aluminium produced is therefore in a liquid state and deposits due to gravity on the cathode.
The carbon anodes are fixedly attached to so-called anode hangers which again are securely held to an anode bar for mechanical and electrical connection. As the carbon anodes are consumed and metal is charged from the cells (the metal represents the actual cathode), the anode bar is lowered to keep a constant distance between the cathode and the carbon anodes.
An electolysis cell of common size is usually provided with approximately 20 carbon anodes, and since the anodes are consumed gradually, each anode has to be replaced by a new one after 20-24 days. Thus, in each cell a used anode is replaced by a new one every day.
According to the conventional setting method, the new anodes are set or positioned so that the distance from the bottom side of these to the cathodes is the same as the distance for the old ones being exchanged. The exchange of anodes is carried out in different ways. The most common way of doing it, is by providing the old (used) anode, or rather the anode hanger with a chalk mark referring to a reference point on the anode bar, usually the bottom side of the anode bar. The used anode is then placed alongside a new anode on the floor to transfer the measurement marked with the chalk mark on the old anode onto the new one, and the new anode is thereafter inserted in the cell.
The here described manual method for setting the anodes is, however, liable to error, caused by the width of the chalk mark, errors of parallaxe during the tranference of the measurements from the old anode to the new anode, irregularities of the surface on which they are stood etc.
The errors and irregularities result in that the anodes are not positioned at the correct level in the cells, and this again will result in unwanted operation disturbances (uneven current absorption, carbon slipping etc.) causing economical losses.
A mechanical device which is also based on the conventional method, is described in GB patent application No. 2.018.291. The device comprises a crane which is employed to exchange old anodes with new ones. Thus, the old anode is pulled out until, after passing through a certain travel, the surface facing the cathode has reached a predetermined horizontal plane. The distance travelled through until then is stored. The new anode is positioned with the surface facing the cathode in a second horizontal plane and is lowered towards the cathode in accordance with measurement of the stored level, the distance between the two horizontal planes and possibly with regard to different saggings of the crane caused by the different weights of the new and old anode.
US Patent No. 454074 discloses a method of setting carbon anodes in electrolysis cells in which the anodes are maintained in a fixed relationship to the cell cathode. Each anode is fixed in an anode rod, the rods being mounted to an anode bar. Each anode rod has a reference mark positioned thereon a set distance from the bottom of the anode. A new anode is set against an old anode by measuring a set distance for the reference mark corresponding to the estimated amount that the old anode has burnt off and projecting a light beam from that position on the old anode rod onto the new anode rod and setting the new anode rod accordingly.
Even though these devices have eliminated some of the subjective measuring errors, the devices are encumbered with objective measuring errors which have influence on the positioning of the anodes. Besides, the above-mentioned devices are expensive to produce.
As previously mentioned incorrect setting of the anodes will give economical operational losses due to disturbances under the electrolysis process. A further disadvantage with the convential setting method is the increase in anode consumption, cfr. later section.
With the present invention it has been a main object to provide a method for seeing (positioning) the anodes in electrolysis cells by which the above disadvantages are eliminated i.e. where:

- a more even current absorption is achieved, whereby the anode slipping is reduced and repositioning of the anodes is avoided,
- the control level is raised due to the fact that a systematic source regarding variation in current absorption is eliminated,
- there is achieved a greater chance to reveal problems connected to the anode-exchange routine, such as anode carbon remainders, mud etc. being present under the anodes,
- the anode consumption is reduced as the anode endurance principally is governed by the "smallest" critical anode butts",
- it is possible to increase the size of the electrolysis cells without having to use individual anode regulating means.

The invention is characterized by the features of the attached claim 1,
Advantageous embodiments of the invention are defined by the independent claims 2-4.
The invention will now be further described by means of examples and with reference to the drawings in which:

Fig. 1 shows a simplified longitudinal section of an electrolysis cell in which the conventional setting method for the anodes is used,
Fig. 2 shows a simplified longitudinal section of an electrolysis cell where the setting method according to the invention is used, and
Fig. 3 shows schematically the horizontal positions for "n" anodes in an elecrolysis cell.

As mentioned initially and as shown in Fig. 1 and 2, a Hall-Herould electrolysis cell 1 for producing aluminium principally consists of a cathode 2 and one or more carbon anodes 3 provided above the cathode. The cathode which contains the cryolite bath is made of carbon blocks 4 placed in an internally isolated steel shell. The carbon blocks are connected to current leads by means of steel bars streching all the way through the cathode (not shown).
The carbon anodes are cast or in some other way fixedly connected to anode hangers 8, which in turn are releasably connected to an anode bar 7 by means of connectors (not shown). Electric current is supplied to the anode bar via flexibles 10, and the anode bar is lowered and lifted in a regulation zone 12 by means of jacks 11.
The electric current is lead, as will be appearant above, from the top to the bottom of the cells. On the bottom side of the anodes, the aluminium oxide dissolved in the bath 13 is decomposed to aluminium metal and oxygen. The aluminium is, due to the gravitational forces, deposited on the cathode, while the oxygen immediately reacts with the carbon of the anode to carbon dioxide. To maintain the constant distance to the cathode, the anodes are lowered. This is done by lowering the anode bar by means of said jacks 11. When the anode bar with the carbon anodes has reached the lowermost position, the anode bar has to be lifted - so called "cross lifting" - while the anode hangers are intermediately mechanically fixed to an assisting bar which is called raiser.
As previously mentioned there are about 20 carbon anodes in an electrolysis cell, and since the anodes are gradually consumed, each anode has to be exchanged after approx. 20-24 days. In each cell there is thus about one anode exchange every day.
In Fig. 1 the anodes are positioned according to the conventional setting method. Since this setting method is previously described, only the disadvantages will now be mentioned.
The main principle with the conventional setting method resides in that the new anodes should be positioned at the same level h above the cathode as the old anodes. In practice it is shown, however, that in connection with the anode exchange several errors occur which result in relatively large deviations in the setting height for the anodes. These setting deviations cause increased anode consumption simultaneously as they lead to operational disturbances due to the fact that the new anodes either draw too much or too little electric current. The relations are as follows:

If the anodes are positioned too low (short interpolar distance between the anode and the cathode) the current consumption is increased and accordingly the anode consumption increases. As opposed hereto, if the anodes are positioned too high, the current and thus the anode consumption is reduced.

In Fig. 2 is shown a similar electrolysis cell as is shown in Fig. 1, but in which the anodes are positioned according to the present invention. The method will be described as follows:

When manufacturing new anodes, or rather when the anode hangers are assembled to the anode carbons, the anode hangers are provided with one or preferably two reference marks 16, 17. The reference marks may be in the form of a readily removeable paint and are painted in a predetermined distance from the bottom side of the carbon anodes which is equal for all the anodes. A fixed rule 18 is further made having markings 19 in vertically, spaced apart relation. The distance between each reference point 19 defines the expected anode consumption per. unit of time. This anode consumption is dependent upon several factors such as carbon quality and current density.

The rule 18 is thereafter fixedly positioned on the anode bar, one for each anode in the cells and in a distance from the metal plane 15 (the surface limit between the bath and metal) which is equal for each of the anodes. The rule 18 may be drawn on paper which is glued onto the anode bar, or it can be painted or drawn directly on the anode bar.
The reference points 19 on the rule 18 are provided with numbers 1,2,3 and so forth in rising order upwards (not shown). The length of the rule is dependent upon the length of the regulation zone 12 for the anode bar and how large the part of the anodes is which can be consumed. Hence, the rules have to be longer than the sum of the length of the regulating zone for the anode bar and the length of the maximum anode consumption.
During testing of the setting method according to the invention it was found that the rule had to be approx. 80 cm long. Further, the anode consumption was calculated to be 1,6 cm/24h. Thus it was found that the rule should contain approx. 50 marks. Instead of using only one reference mark on the anode hanger and a rule being 80 cm long, it was experienced that the anode hanger could be provided with two reference marks with a spacing of about 40 cm to be able to shorten the rule to half the lenght, i.e. 40 cm with 25 marks. The lowermost reference mark 16 on the anode hanger will thus normally be employed when new anodes are positioned in the cell, while the uppermost reference mark 17 on the anode hanger normally will be employed when "crossing" of the anode bar takes place.
As mentioned above, the rules 18 are positioned at a distance from the metal/bath surface limit which is equal for all of the anodes. If thus a line is drawn along the anode bar which toches the upper 21, or lower 22 end of the rules, it will have a shape which to a large extent corresponds to the curved metal plane.
In connection with the anode exchange the new anodes are positioned according to the expected setting height, i.e. the reference mark 16 or 17 on the anode rods 9 is placed in correspondance with the respective reference point (setting point) 19 on the rule 18 for the anode bar.
The calculation of the set point is normally accomplished by means of a calculator which adds one reference point 19 (1,6 cm) for each day. When deviation occurs, for instance if the current consumption increases for every passing day, the calculator will decide to reduce the current consumption and give a signal indicating that the anode should be repositioned at a higher level (one reference mark) above normal setting height. The results are presented each day on a dayly set-list which the carbon exchange operators are using.
With the here described method a considerably more accurate setting/positioning of the anodes is achieved, thereby reducing the anode consumption, due to the fact that several sources of errors are eliminated. Further, a more even current distribution in the cells is achieved giving further reduction to the anode consumption as the distance between each of the anodes, in the cells and the underlaying metal plane is equal.
As to the metal plane, this can be calculated by measurements or theoretically by means of magneto hydrodynamic models. In the following it will be further described how the metal plane preferably can be calculated by means of measurements.
The above-mentioned rules 18 are fixedly positioned on an anode bar in an electrolysis cell in the same horizontal plane, and the anodes are positioned according to the same reference mark, i.e. the bottom sides of the anodes are positioned in the same horizontal plane.
A statistical material is worked out in the form of measured current consumption I for anodes with an operational time of 24h. This is done for each anode position in the horizontal plane. Fig. 3 shows schematically the horizontal positions for n anodes in an elecrolysis cell. I.Lj can for instance represent the arithemtic average of m singular current consumption measurements, I, and which gives the equation:

where E(I) is the expected value of the current consumption I and p(l) is the probability density distribution for the current I.
With a reasonable statistical basis, i.e. more than 100 measurements for each of the n anodes, it is possible to determine the current consumption distribution for the cell.
To be able to calculate the height of the metal plane underneath each anode, it is necessary to find the co-relation between the distance from the bottom side of the anode to the metal plane d, and said current consumption I.
This is done by studying the reaction of the current consumption when the anodes are positioned at an abnormal height, Z. Normally, when the anode is positioned at a point of time k, it is positoned 1,6 cm (one mark) higher than the anode positioned at a point of time k-1 (assuming that the difference of time is 24 hours). The probability value for the difference between the current consumption for the anodes positioned at the point of time k and k-1 is then zero.
This under the assumption that the stipulated anode consumption is 1,6 cm/24 h.
By positioning the anode "abnormal", i.e. not 1,6 cm higher than the previous anode, but for instance 0 cm or 2 x 1,6 cm higher (so-called stop or hop corrective attempts), the probability value will be unequal zero:
81 is the response of one pertubation 6z which is + or + 1,6 cm with regard to what is "normal". Thus, a relation is given between the current consumption and the positioning of the anode, relative to the metal plane:
By gathering a statistical basis of many corrective attempts (hop/stop), one can find an estimator for the probability value for ∂l/∂Z, for instance the arithmetic value which has the correct probability when/if ∂l/∂Z has a normal distribution.
With the estimator ∂l/∂Z it is now possible to get back to the previously mentioned current consumption distribution, and the curvature displacement of the metal plane relative to the average metal hight can be estimated according to the equation:
where DZ_j is the displacement below the anode position j, µ̂_j is the estimated current consumption of the anode position j and p is the average current consumption for all the anode positions, n.

Claims

1. A method of setting electrodes in electrolysis cells, particularly carbonaceous anodes (3) in cells (1) for producing aluminium according to the Hall-Heroult process, where the cathode (2) of the cells contains a bath consisting of aluminium oxide dissolved in melted cryolite, the aluminium metal being deposited on the bottom of the cathode and having an upper surface forming a metal plane (15), each anode having an anode rod (8) provided with at least one reference mark (16 and/or 17) defining a predetermined distance from the bottom surface of the anode, the anodes being connected to an anode bar (7),
characterised in that
the anodes are positioned or set according to a rule (18) having reference points (19) each defining the expected anode consumption per unit of time, whereby the reference mark (16 or 17) on the anode rod (8) should correspond to a reference point (19) on the rule (18) which is in accordance with the predetermined setting height, the rule (18) for each anode being fixed to the anode bar (7) at a predetermined and equal distance from the metal plane (15).

2. Method according to claim 1, characterised in that the metal plane (15) is determined by means of measurements or by means of theoretical calculations based on magneto hydrodynamic models.

3. Method according to claim 1, characterised in that the expected anode consumption is 1,2 - 2,0 cm/24 h.

4. Method according to claim 1, where the anode bar is adjustable in a vertical direction over a regulating zone (12) from an upper point to a lower point where the anode bar is subject to cross lifting, characterised in that the anode rod is provided with two reference marks, an upper reference mark (17) and a lower reference mark (16) whereby the lower reference mark is normally applied when new anodes are being positioned, while the upper reference mark is normally applied when the anode bar is subject to cross lifting.

5. Method according to claim 2 for calculating the metal plane by means of measurements, characterised in that:

(a) the rules (18) are fixedly positioned on the anode bar in the same horizontal plane, and that the anodes are set according to the same reference point, i.e. such that the bottom side of the anodes are situated in the same horizontal plane,

(b) a statistical material is worked out in the form of current consumption per. anode position in the horizontal plane, calculating the arithmetic average of m singular measurements, according to the equation,

where E(I) is the probability value for the current consumption I and p(l) is the probability density distribution for the current I,

(c) that the relation between the distance from the bottom side of the anode to the metal plane (d) and said current consumption I thereafter is found by setting the anodes "abnormal", i.e. that the probability value is unequal to zero,

where I is the response of a pertubation Z which is one reference point higher or lower relative to what is normal, such that there is a relation value between the current consumption and the positioning of the anodes relative to the metal plane, ∂l/∂Z, and that

(d) the curvature of the metal plane underneath the anode position relative to the average metal height is calculated in accordance with the equation

where DZ_j is the deflection of the metal plane underneath the anode position j,

is the assumed current comsumption for the anode position j, p is the average current consumption for all of the anode positions, n, and (∂l/∂Z) is the estimator for the probability value of al/aZ.