SPECIAL PURPOSE MACHINE FOR FIELD IMPROVEMENT OF ELECTROLYZERS’ BOTTOM PLATE
György VELEZDI *Summary
The smaller and larger topological damages have been developed long service of duty and caused by electrochemical corrosion on the surface of electrolyzer cells’ bottom plate can result in breakdowns in the continuous work-cycle of electrolyzer plants and increase the maintenance cost considerably. Widely used solution to this problem is the replacement of bottom plates, but it is very expensive and there is a time limit as well, since it can execute only during the yearly general overhaul.
A new process has developed and patented that make field improvement of bottom plates possible while the whole electrolyzer plant could produce free from notable loss of capacity. The essential element of this new process is a special twin-spindle CNC milling machine. The subject of this paper is the designing aspect of this special purpose machine tool.
| _______________________ |
| * György VELEZDI, Dr. Univ. |
| Assistant Professor |
| 5, Károly str. |
| H-3532 MISKOLC |
| Received: May 31, 1999 |
1. Introduction
In 1978, an electrolyzer plant started to operate at the BorsodChem Chemical Works in course of the PVC-III Investigation Project. In the beginning, it contained forty electrolyzer cells but in 1989, further four new cells were added increasing the chlorine production capacity of the plant. The cells are built up by De Nora system and work on 275000-ampere nominal load. They are suitable to turn out Cl2, NaOH and H2 by electrolysis of sodium-chloride solution using mercury cathode and titanium anode.
For twenty years of duty, the surfaces of electrolyzer cells’ bottom plates that contact with the flowing mercury are considerably corroded. The results of corrosion are the crater and channel shape cavities on the surface of bottom plate that is made from conventional steel material. As a result of these topological deviations, the laminating flow of mercury could become turbulent, moreover some places the layer of mercury could break. On those locations where the bottom plate is not covered by mercury, the sodium-chloride solution directly touches the plate so the corrosion is more intensive and may develop hydrogen as well, owing to the ferric cathode. The hydrogen comes into the gas space and may form explosive mixture mixing with chloride. To reduce the risk of evaluation of these situations there is a need for keeping a wider gap between the anode and cathode. Avoiding catastrophic short circuits is another reason to use larger distance between the electrodes. So finally, the cells must work on higher voltage and must be supplied with larger volume of mercury, which means the rise of production and maintenance costs.
To avoid the above-mentioned problems the electrolyzer plants replace the bottom plates for a new one the top surface of which is strongly corroded. The replacement is very expensive and takes long meanwhile full shut down of voltage is needed since the cells are connected in series electrically. It means halt of the total technological line as well if the electrolyzer plant is an integrated part of the continuous work cycle system such as PVC production. Another well known attempt for reducing the break downs caused by corrosion wear is the local repair of the bottom plate using welding technology in those areas where the cavities are too deep. However, in this manner, the topological quality of renewed plate really should not be acceptable.
Within the scope of a research project, a new process has been developed and patented [6] for the field improvement of bottom plates on the base of following novelties:
| The 3-5 mm deep injured layer of bottom plates can be removed on the spot by cutting technology and this way it should be put back into the original state of working surface. The removed layer of material does not jeopardise the normal work of the renewed cell either from an electric or static point of view. | |
| Each damaged plate contains a pair of intact surfaces that can serve as a basis for supporting and moving a suitable constructed milling machine capable of removing the injured layer of the bottom plate. | |
| The cutting process can be done without halting production of the whole electrolyzer line. Only one cell is out of service that is being renewed, but this one is also electrically connected in series with the working others. | |
| The cost of the bottom plate renewal is one order lower then the plate replacement. Thanks to the productivity of the developed special milling machine, one cell might be completely renewed pro month and can work for a newer decade in good condition. |
2. Geometric system of the machine
The machining task is unusual. Generally, the work-piece is located in the workroom of the machine tool but now the work-piece is the bottom plate. It is too heavy (17 tons) and too large (13 meters long, and 2.4 meters wide) to be set on the machine. If we wish to machine such large and heavy work like a bottom plate, the cutting machine must be located on it.
Of course, before the renewing process, all four walls and other units of the electrolyzer cell that are located over the plane of bottom plate must be removed, namely, the cutting tool must be able to reach the whole surface of the plate and the main supporting elements of the machine can be placed out of this area.
The first step for creation of the basic machine structure is to determine the dimension of workspace using the naked bottom plate drawing (see Figure 1).
|
| Figure 1.: Cross section of the bottom plate |
As the bottom plate is almost 13 meters long, the whole workspace must be shared for cutting zones. Otherwise, manufacturing and transporting of the machine would be difficult, moreover; the stiffness and accuracy of machine would not be high enough. The result of optimisation process is a machine with a workspace size that provides the injured layer removal in eight longitudinal set-ups (see figure 2). A short stroke lifting and a travelling motion are needed for passing on the machine between the cutting zones.
![]() |
Figure 2.: Cutting zones division of the bottom plate |
Finally the necessary feed motions of the machine for walk around the cutting zone are 1800 mm longitudinal stroke along X-axis and 2048 mm cross stroke along Y-axis. These two planar motions must be equipped with machined feed, but the third vertical stroke would be manually controlled since this motion is rarely actuated. The purpose of this 125-mm motion along Z-axis is the setting the depth of cut and providing enough room for manual tool change. For the complete improvement of a bottom plate two kinds of cutting sequences are necessary:
| three longitudinal slot millings on both sides and the middle of the top surface of the plate using 16 mm end mill (the purpose of these slots is to provide stretched mercury layer on the bottom plate skin of the cell), | |
| surface milling overall surface of the plate using 100 mm face milling head (it means about 625 kg removed material from a 26 m2 area). |
It is evident that the slot milling can be realised only with one kind of tool path, using longitudinal milling along the X-axis 3 times. However the face milling can be executed in two different ways (see figure 3).
|
Figure 3.: Variants of face milling tool paths |
There is no significant difference between the two strategies either from a cutter load or cutting time point of view. Nevertheless, it is easy to observe that the direction of longer stroke of longitudinal tool path is the same as the direction of slotting and only a short 80-mm stroke must be done in cross direction. Furthermore, the direction of cutting scratches is parallel with flow of mercury. For this reason we can claim, the longitudinal milling strategy is better then the cross one.
3. Practicable basic structures of the machine
Systematically searching for practicable basic structures of the machine, we can find two reasonable versions altogether; versions A and B (see on figure 4). The reason is very simple; the large planar motion can be done in only two ways using a crossrail (moves along X- or Y-axis) and a spindle stock (moves along Y- or X-axis). The third necessary short stroke along Z-axis must be done with the smallest element (quill or small slide) only.
Basic version A |
Basic version B |
![]() |
|
Figure 4.: Basic structure variants of the machine |
|
Further two pairs of double-spindle subversions can be created on the base of A and B. The two spindle stocks can be arranged serial or parallel way (see on figure 5).
| Subversion A-1 | Subversion A-2 |
![]() |
|
![]() |
|
| Subversion B-1 | Subversion B-2 |
Figure 5.: Double-spindle subversions of practicable basic structures |
|
The optimal machine structure must comply with the following four main requirements:
| productive cheep removal | |
| reasonable tool path for the working area | |
| good quality of the machined surface | |
| proper structure for the economical machine production. |
Needless to say the twin-spindle versions are twice as productive as the single spindle ones but the costs of machine building and operation are much less then as double. It has been mentioned previously, the longitudinal milling is better then as the cross one. It is advantageous that is why the much longer stroke can be done with a relatively small size and well guided spindle stock and only a short cross stroke can be done with a large and heavy moving crossrail that is inclined to get wedge in. Better surface quality and smoother operation are the results of it in case of both slotting and facing. After that, we need to choose between subversions B-1 and B-2 only. It is very easy to realise a cutting anomaly at B-1 thanks to the serial arrangement of the two spindle stocks. In the beginning of cutting, the second in row spindle can start to work only plunge cut way, but it must not do either a face milling head or a strait end mill unless damage their cutting edges. It is worth saying that subversion B-2 is the most compact in floor space and requires the shortest main frame elements and because of the shortest strokes the guideways and lead screws length would be minimised as well. At the end of the morphological study it becomes clear that variant B-2 is the most suitable for detail design.
4. Detail design of the machine
After the structural analysis of the machine, the next step is the creation of functional variants that are able to realise basic functions of the machine. A refined three-dimensional model can be seen on Figure 6 equipped with a couple of functional devices.
|
Figure 6.: Functional variants of the machine |
The basic functional devices that implement the functional requirements are as follows:
| welded main frame, assembled two longitudinal and two cross beam, equipped with rolling guideways for the moving crossrail | |
| a pair of mechanically synchronised ball screw, driven by DC servo motor for crossrail actuation (the final solution differs; backlash free rack-and-pinion drives on both end of the crossrail instead of ball screws) | |
| welded structure moving crossrail, containing two pairs of guideways for the slides of spindle stocks | |
| two variants of ball screw drives driven by DC servo motors for actuation of the longitudinal slides (the final solution contains two independent drive systems) | |
| spindle stocks with main drives driven by frequency-controlled AC motor and manually operated quill feed mechanisms | |
| hydraulic lifting devices and side guides for transporting the machine on the bottom plate among working zones | |
| four frame stiffeners used in case of crane hoisting of machine between cells. |
The detailed construction of each element and subassembly has been carried out in the final stage of the design process. Just as the conceptual planning, the detailed design was carried out in a three-dimensional way. The 3D-modelling technique provided real opportunity to build up electronic representation of the whole machine (see examples in figure 7, 8, 9,10).
![]() |
Figure 7.: 3D model of the spindle stock; 1: welded spindle house – 2: main drive and quill feed mechanism – 3: completely assembled spindle stock |
![]() |
Figure 8.: 3D model of the crossrail that contains all three DC-servomotor driven feed drives for cross and longitudinal direction |
![]() |
Figure 9.: Complete 3D assembly model of the machine |
![]() |
Figure 10.: Montage from three-dimensional models of the machine |
Based on these models all necessary geometric, kinematic, dynamic and technologic simulations have been performed by using the high-end Pro/ENGINEER CAD/CAM/CAE software. Thanks to this up to date designing method, there were not any significant problems during the production and trial of the machine, and the total time requirement of the project could also be reduced dramatically.
Of course, several difficulties appeared at the chemical plant at the beginning of the test run. Most of them derived from the extreme circumstances; these are the very strong electromagnetic field and relatively high temperature (60°C) of the bottom plate’s surface generated by 275000 ampere working current of the electrolysis, as well as chlorine-polluted air of the plant. The strong electromagnetic field makes difficult the cutting process. The chip removal is not enough intensive at tip of the cutter since all ferromagnetic units (bottom plate, cutter, most of the machine parts and itself the chip as well) become magnetic. The chip sticks to almost everything, flutes of the cutter get filled and after a time, the chip removal becomes impossible. Therefore strong compressed air flow -- sprayed by additional inserted nozzles – blows away the removed material from the cutter tip environment and this air flow cools the cutting tool as well providing long enough cutting edge life. A powerful vacuum apparatus collects the scattering chips providing clean room in and around the workspace of machine and minimising pollution of the cells. The sealed electric cabinet -- including numerical controller device -- of the machine operates under overpressure of clean and cool air avoiding breakdowns caused by aggressive gases present in the electrolyzer plant.
5. Concluding remarks
Over the past six months four bottom plates have been improved by means of the above presented new process and using the developed special milling machine. The technical and economic results of the research-development project have fulfilled the expectations. It has been proved that the new process is able to solve the problem of bottom plate renewal in mercury cathode electrolyzer plants.
The productivity of the bottom plate improvement is suitable for the cell renewal schedule of the electrolyzer plant. The net improving time is about 40 hours for each bottom plate. The whole renewal of a cell requires 14-20 days added to the time of other necessary activities that are needed for completing this job. It means 10-12 cell renewals annually, so the 44 cells of the plant would be renewed for four years, and probably provides satisfactory operation for the whole service life of the plant. The renewal would take more then twenty years in the way of conventional plate replacement. The difference between the cost of bottom plate replacement and field improvement takes about 100 thousand dollars per each cell.
The cell operation under lower voltage provides notable saving of electric energy thanks to the better bottom plate surface smoothness. Less need of maintenance and losses caused by breakdowns decreases production cost of the plant too. Nowadays it is needless to say how important the protection of the environment is, so finally the loss of environment stress as the concomitant result of less mercury consumption must be mentioned.
More than one hundred electrolyzer plants are working all over the word using the same mercury cathode technology. All of them have same problem derived from bottom plate corrosion, so they could be potential customers for this new procedure, therefore a claim was put in to a patent for the new bottom plate improving process in august 1997.
Probably these facts contributed to winning the Innovation Prize of the National Technical Development Committee on the 7th Hungarian Innovation Grand-prize Competition in 1999 [5].
REFERENCES
| Back |