82 The N.Z. Journal of Science and Technology. [Mar. 
closest approach to the desirable spherical shape and to the still more 
desirable “fool-proof” construction. The Electro - Metals furnace is a 
British two-phase, two-electrode type, and probably has more metallurgical 
and electric furnace experience behind its design than any other type. 
Whatever furnace is chosen should have a removable dome-shaped roof 
of silica bricks at least 10 in. thick, and the hearth should be lined with 
basic material in order to enable sulphur and phosphorus to be removed. 
In modern practice magnesite seems to be preferred over dolomite for 
basic lining, but in any case the lining shpuld be laid on in sections 
and not in one piece. The size of furnace desirable will probably be the 
2-ton size, as this is suitable in foundry practice for an output of 500-800 
tons per annum. Ample power should be arranged for, certainly not under 
600 kilowatts, and 750 would probably be better still for this size furnace 
under New Zealand conditions. The advantage of the electric furnace for 
these conditions is chiefly that it enables a charge of cheap mild-steel 
scrap to be used without any expensive pig additions, and, further, that 
the baneful sulphur (and phosphorus) can be easily removed. A typical 
charge would consist of 60 per cent, mild-steel scrap and 40 per cent, 
foundry-steel scrap. This would be charged cold into the furnace and 
would take under 1,200 units per ton to melt and refine, or if current costs 
fd. per unit at the furnace and the yield of castings is 60 per cent, of the 
molten metal the cost of electric energy per ton of finished castings would 
be about £3. This figure can probably be improved upon in certain 
cases where power is obtained from an electric-power system of large 
capacity and where it may be possible to avoid working the furnace during 
peak load. However, the chief aim is to show that in this industry, as 
in every other, the cost of power is quite a small percentage of the total 
cost of production. The heaviest charges in a steel-foundrv are incurred on 
the moulding-floor, and nothing must be done which will interfere with 
the cheapest and most convenient routine in that department. Electrical 
engineers, in their desire to avoid peak loads, speak of the brilliant savings 
that will be made in the power-costs if the melting is done at night and 
pouring in the early morning, but this is a fallacy where steel castings 
are concerned. Undoubtedly, pouring in the afternoon and leaving the 
castings in the mould all night is the correct procedure to adopt in small 
foundries. Even quite small castings should be allowed to cool gradually 
through nine or more hours before being taken from the sand, and if 
pouring is done in the morning this can only be arranged for by having 
a double stock of boxes and a double area of floor. This arrangement 
will, of course, greatly increase the capital cost of the plant, but will also 
increase the working-costs, as the floor will be unwieldy and the few 
shillings saved in the bill for current will be overbalanced by pounds lost 
on the moulding side. Assuming, then, that current is obtained at the 
high rate of §d. per unit and the cost of power per ton of finished castings 
is £3, molten metal of high and suitable quality will be obtained for £8 
to £9 per ton in the ladle, or almost as cheaply as the poorest and cheapest 
metal from a converter, and distinctly more cheaply than metal of reason¬ 
ably good casting quality can be obtained by any other process than the 
electric under the conditions assumed. The selling-price of the castings 
is, of course, another story, depending on moulding practice and on the 
output obtained. At the outset, and with a small business of about 
200 tons per year, a price of £25 a ton for straightforward work should 
be obtained There will naturally be a desire on the part of the owner 
of such electric furnaces to increase the yield, and as such increase can 
only come gradually from the foundry side it is wise to consider other 
possibilities. A small rolling-mill for bar iron naturally suggests itself, 
