and then into a new 400 kA cell thermo-electric design [1] by extending the length of the potshell.
Then in 2003, by further extending the length, and this time by also slightly increasing the potshell
width, he presented a new 500 kA cell thermo-electric design [2]. Later in 2005, still extending the
length of the 500 kA cell potshell, he presented a new 740 kA cell thermo-electric design, and
claimed that there is no foreseeable limit to a cell size as far as the thermo-electric cell heat balance
aspect of the cell design is concerned [3].
design of the 500 and 740 kA cells claiming that there seems to be no foreseeable design limit to the
cell size as regards the MHD and potshell mechanical aspects [4-6]. The recent increase in cell
amperage in newly constructed smelters and prototypes, for example in China [7-9], confirms of
this point of view.
the cell lining design, so much so that a cell lining seen as best world practice ten years ago now
appears old-fashioned, if not obsolete. This evolution clearly warrants a second retrofit study in
order to incorporate the new best practice design ideas: using new or improved modelling tools, the
aim is to boost the 500 kA cell design into a 600 kA cell design, while still keeping the same
potshell and busbar.
three different busbar designs for the 500 kA cell. The first one is a "classical" asymmetric busbar
(Fig. 1) that auto-compensates for the return line [10]. The second one is a symmetric busbar (Fig.
2) with an external compensation busbar inspired by Pechiney 1987 patent [6, 11]. The third one is
also a symmetric busbar (Fig. 3), but it uses a different configuration for the compensation busbar
[6].
stability analysis code. This upgraded version takes into account the presence of the open bath
channels, and is hence better at predicting the shape of the bath-metal interface [12]. There is little
change in the calculated vertical component of the magnetic fields (Bz) and in the stability
deformation is now significantly different.
mechanical (TEM) model presented in [13] has been used to calculate what would be the anode
voltage drop of a 1.95 m long by 0.665 m wide anode. This anode has 4 stubs of 175 mm diameter
and incorporates a new type of stub hole design. There are 48 such anodes in the 500 kA cell design.
In Fig. 4, the 1/16 TEM anode stub hole model predicts a voltage drop of 214 mV from the bottom
of the anode carbon block to the top of the stub. The obtained average contact resistance is then fed
to the standard, half anode thermo-electric (TE) model, which in turn predicts 265 mV for the total
anode voltage drop from the clamp connection to the bottom of the anode carbon block (Fig. 5).
slot TEM model presented in [14] has been used to calculate what would be the cathode voltage
drop of a 4.17 m long by 0.665 m wide and 0.58 m high fully graphitized cathode block. That
cathode block has two collector bar slots and uses 220 mm high by 140 mm wide collector bars