6.  The Production Process part 1

by Michael Mooney

This is an account of how mill rolls were produced.  On the following pages (linked at the bottom of this page) are photographs from a book produced by Perry's which illustrate this process.  These photographs are referred to by their numbers in this account.

The production process can be said to start with the making of the patterns (wooden replicas of the necks and wobblers of the rolls) and sweeps (for grain rolls) (See photographs 1,2 and 3). Sweeps were profiled wooden boards used for moulding rolls, which were cast with their profiles.

The neck and wobbler patterns were placed in moulding boxes and sand rammed hard around them. (See photographs 4 and 5) When all the sand had been added the patterns would be withdrawn leaving a neck and wobbler shaped void into which the metal would be cast when the final mould assembly took place.

The method of making grain rolls using sweeps is described alongside photograph 3.

Once the mould had been made and assembled the melting process could begin. The furnaces were charged with a mixture of lumps of scrap roll (worn out rolls bought back from the customer) (See photograph 8), surplus metal from previous casts, the heads and gates from previous casts, and pig iron.

The weights and chemical composition  of each of these various components of the charge were established and  a great deal of careful calculation done by the metallurgists to ensure that the mix was such as to provide a chemical composition as near as possible to that demanded by the nature of the rolls being cast. (See photograph 31)

Chemical constituents included carbon, silicon, manganese, nickel and chromium. It was necessary to calculate the charge so that the minimum of alloying additives (which were expensive) would need to be added to the furnace to obtain the correct composition.

The furnaces were reverbatory furnaces fired by pulverised coal. [A reverbatory furnace is one in which the fuel is not in direct contact with the charge, the heat generated by the burning fuel bouncing- reverberating- from the furnace walls onto the charge.]

Once the furnace was charged- the furnaces shown had removable tops so that the charge could be loaded by crane – and the tops replaced, the pulverised coal was drawn from the silos (see photograph 9) by a system of pipes to the furnace burners where it was blown into the furnaces as a jet of coal dust and air.

The coal/air jet was lit and the heat produced by the burning jet of coal dust and air heated up the furnace and after a while caused the metal to melt. The metal would only be tapped from the furnace when it had reached the desired temperature for the particular metal composition and roll type. (See photographs 10 and 11)

How long a charge took to melt and attain the desired temperature depended upon a variety of considerations: the weight of the charge; the tapping temperature needed etc., but was certainly several hours. The furnaces would be fired up early in the morning (perhaps at around 6.00am) and it might be 1.00pm or 2.00 pm before the furnace was ready to tap.

The furnace would then be tapped out as shown in photographs 12 13 and 14 and cast as shown in photographs 15 and 16.

After casting, the mould would have to be left for some while (in the case of the large rolls for perhaps 3 days) to allow the casting to cool sufficiently to be removed, partly because of the danger and discomfort to the workers involved and partly because if done too soon the thermal shock of the cold air on a hot casting might cause it to fracture.

The moulding boxes and chills (where used) would be removed by a small gang of men called “Strippers”, the gate broken off the casting  and the casting lifted by crane to one end of the foundry, when adherent sand would be removed using a pneumatic chisel and powered grinders. The man who did this was called the “Dresser”. The casting would then be transferred to the machine shop for machining. (See photographs 17 to 30)

Some rolls would have to be annealed before machining. This involved placing them in gas fired annealing furnaces and heating them slowly to a temperature well below the melting point and then allowing them to cool in a controlled manner over days. The purpose of annealing was to alter the metal structure by the heating and cooling process and so make them more resistant to the shocks which they would encounter when in service in the mills.

The following photographs, which appeared in the book which Perry's published in 1967, to mark the modernisation of the plant, give a good idea of how mill rolls were produced.

1. Using the circular saw  to cut timber for patterns.

The pattern maker is Roy Miller.


2.   Shows the pattern maker, Roy Miller, checking the dimensions of a wooden pattern for a roll neck.

3.  Shows Isaiah Wellings beginning to mould a section of a grain roll.

A grain roll was a relatively soft sand cast roll, cast with its profile.  Grain rolls were used in the mills for initial rough forming of steel. The mould was made in sections using a sweep- a wooden board cut to the profile and size necessary. A moulding box was placed on a metal base plate on the foundry floor and the sweep bolted to a rotating central spindle [these are clearly seen in the photo] Iron bricks (called denseners) were built up and sand packed between them, the sweep being rotated to ensure that the correct profile was being maintained. When all the denseners had been placed a coat of loam (always pronounced “loom”) was applied by hand. This was a thick paste of sand and various binders (special clays etc). [In the old days horse manure was added to the loam mix to give added binding power but I cannot recall this being done during my time]

The sweep would be rotated again as necessary until the moulder was satisfied that the correct dimensions and profile had been produced. The mould section would then be removed and dried in coke fired ovens [always called stoves] for perhaps two or three nights. Once dry, the various sections would be brought to the casting pits and assembled to form the finished mould.

4.  Shows the moulder, Tommy Morris, checking that the pattern for a “wobbler” is correctly set in the mould before beginning to mould the sand around it.

The wobbler was essentially the means by which the roll was connected to the mill drive when installed in the mill. [If I remember right not all rolls had wobblers- some had spade ends (see photo on page 55 for illustration of spade ends) I think that it was probably the strip and plate mill rolls which had spade ends rather than wobblers.]

5.  Shows the beginning of the moulding process for the neck of a roll.

The moulder, Walter Jayes, is checking that the wooden pattern is level and at the correct height.


6.  Shows a large chill being sprayed with a carbon black solution. This was to prevent the molten metal adhering to the chill during casting.

A chill is a cast iron cylinder of suitable diameter, used to make the mould of the body of certain kinds of rolls [known as chill rolls] The purpose of the chill was, as its name rather suggests, to cause very rapid cooling of the molten metal during casting, producing a very dense metal structure at the cast surface and imparting a hard-wearing surface to the body of the roll.

I cannot remember the name of the worker.

7.  Shows a small chill being dressed. A slurry of sand and binders was applied with a brush. Again the purpose was to prevent the molten metal adhering to the chill during casting.

The chill dresser is Ron Smith.

8.  Shows the scrap gantry to the rear of the foundry. This gantry was installed as part of the mid-60s modernisation.

Not seen in the photograph is the breaking pen. This was a circular enclosure of heavy steel plate (lined I think with baulks of timber). Scrap rolls would be lifted by the crane into this breaking hole and then a heavy steel ball dropped from the crane onto the rolls, thus breaking them into pieces suitable for charging to the furnaces.

Perry’s prided itself on having very low production scrap rates so virtually all of the rolls seen in the scrap gantry would have been worn out ones purchased from the customer for remelting to make new rolls.

9.  Shows the silos used to store the pulverised fuel which fired the furnaces after the modernisation.

The coal (ready pulverised) was delivered by tanker from James Durran of Penistone, Yorks and pumped into the silos. From the silos it was transferred by pipes to the furnace burners as needed.

10.  Shows the furnace chargehand, Harold Hall, looking into a furnace through a piece of blue glass.

Furnacemen of long standing- like Harold - claimed to be able to judge the temperature of the molten metal in the furnace nearly as accurately as could be measured by an instrument, simply by looking at the colour of the metal and applying their experience.

11.  Shows the temperature of the molten metal (the “bath”) in a furnace being taken.

Holding the pyrometer is Bill Farrell, Chief Metallurgist. Standing next to the furnace is Bill Mullinder, furnaceman.

The long pole like object is a pyrometer, an instrument specifically used for measuring molten metal temperatures. The cable seen trailing from the rear goes to a recorder which traces the temperatures on a paper chart.

12.  Shows one of the new furnaces (i.e. post the 1960s modernisation) tapping out into a ladle.

Each furnace had a hole at the bottom of the base from which the molten metal would run.

Before melting began this hole would be plugged with sand. When the melt was of the right temperature and composition the crane-held ladle was dropped into position and the sand plug removed by using a long pointed steel bar [This was tapping out]. The molten metal was then free to run from the furnace into the ladle. 
The man shown wielding the hammer is driving down the steel gates in the runner channel to hold back much of the slag which emerges at the end of the tap-out, while still allowing the remaining metal to flow into the ladle.  [The slag is lighter than the molten iron so floats on top of the iron in the furnace- as the last of the molten metal runs out so does the slag.]
13.  Another view of a furnace tapping out into a ladle.

14.  Another view of a furnace tapping out into a ladle.

15.  Shows a roll being cast.

The rolls were cast standing on end in casting pits. The box into which the metal is being poured was called the receiver. From there the molten flowed vertically down through a part of the mould called the gate [always pronounced “git”] and entered the main body of the casting from the bottom.

Although it cannot be seen from the photos the git was moulded in such a way that the metal entered the bottom of the roll mould at a tangent. This caused the metal to spin as it rose up the mould, sweeping any slag and debris into the centre and rising to the top. This was so that debris etc would not cause defects on the face or in the body of the casting.

16.  Shows the final stage of the cast, topping up the head. The ladlemen here are Billy Elbro (on the left) and Arthur Billingsley.

The head was at the very top of the mould. Its purpose was to provide a reservoir of metal which would remain molten for longer than the main body of the roll and so feed metal down as cooling and contraction took place, thereby eliminating voids in the finished casting which would have had a very deleterious effect.
Exothermic compounds were shovelled onto the head to help the metal in it stay molten longer.

Casting was one of the most skilful parts of the roll founding process. There was no question of delivering a weighed amount of metal to the mould; all was done by judgement and experience. Clearly the mould had to be filled to the top, i.e. the head, but equally clearly it was highly undesirable to put so much metal in that the head overflowed sending quantities of molten iron over into the casting pit and presenting a real risk of burns to anyone in the vicinity.

Each cast was controlled by the Foundry Superintendent or Foreman who (while the metal was being poured into the receiver) stood on the metal plate shown in front of the mould (protected from splashes and sparks by a hand held wooden shield) and watched the metal rise up the mould. When he judged that there was sufficient metal to fill the mould including the head (and taking into account the quantity of molten metal still  in the receiver) he would call “Up” to the ladlemen and they would then swing the ladle back upright, so stopping any further metal entering. Notwithstanding all the experience of the Superintendent or Foreman, sometimes he called “Up” a little too soon and then the head had to be topped up. This is what seems to be happening in this picture. A large part of the skill in correctly calling the ”Up” was not simply judging how much metal had already gone in but how quickly the ladlemen could swing the ladle upright. This would depend upon the size of the ladle (a big ladle would move much more slowly than a small one) and even on the characteristics and strength of the individual ladlemen.

17.  Shows a beam roll being machined in the machine shop. The machinist is Norman Willis(?)

A beam roll was used to impart the H or I section to steel in the mill. A beam roll would be set vertically in the mill to produce this section.

18.  Shows a strip mill or plate mill roll being machined. I cannot remember the name of the machinist. 

19.  Shows a grain roll being machined. The machinist is Norman Willis(?).

20.  The same, taken from a different angle.

21 A small chill cast roll being machined. The operation here seems to be machining wobblers.


22.  A small chill cast roll being machined.


23.  What seems to be a plate or strip mill roll being machined.

24.  A small chill cast roll being machined. I think the machinist is Cyril Darby.


Return to Later
Company History
  Return to
the contents
  Proceed to the
Production Process 2