Belly helve

A helve is a heavy cast-iron beam, pivoted at one end, that a turning ring of cams lifts and lets fall on a ball of hot iron. The belly helve is the form that takes its lift partway along the beam. The ball comes from the puddling furnace; what the game does with it is in the iron grades guide. The game's own belly helve is planned: its castings exist, the machine does not yet, and today the vanilla helve still works the bar. This page is about the real machine and how the game's will follow it.

A belly helve still in place at Wortley Top Forge. The long beam is the helve; the head is at its right-hand end. Beyond it, through the arch, is the water wheel that drove the forge.
A belly helve still in place at Wortley Top Forge. The long beam is the helve; the head is at its right-hand end. Beyond it, through the arch, is the water wheel that drove the forge.
The same forge from further back. The helve sits low among heavy timbers, chains and hoists: the framing and the lifting gear were as much a part of the forge as the beam.
The same forge from further back. The helve sits low among heavy timbers, chains and hoists: the framing and the lifting gear were as much a part of the forge as the beam.
The forge hammer at Dowlais, labelled on Truran's Plate 62, Fig. 392: a medium hammer with over 40 tons of castings, nearly 19 of them in motion (Truran p. 220). This is the other common form, with the cams at the nose. The ring turns, a wiper catches the point of the helve and lifts it, passes on, and the helve falls on the ball. Everything stands on one cast bed-plate over 1,000 to 1,500 cubic feet of oak (Truran p. 218).
The forge hammer at Dowlais, labelled on Truran's Plate 62, Fig. 392: a medium hammer with over 40 tons of castings, nearly 19 of them in motion (Truran p. 220). This is the other common form, with the cams at the nose. The ring turns, a wiper catches the point of the helve and lifts it, passes on, and the helve falls on the ball. Everything stands on one cast bed-plate over 1,000 to 1,500 cubic feet of oak (Truran p. 218).
The game's belly helve in its planned drawing: the beam, T-shaped at the tail, with its journals in two capped standards; the lifter on its belly over the ring of three cams on the main; the head over the cast anvil, and the prop that holds the helve clear., iso
The game's belly helve in its planned drawing: the beam, T-shaped at the tail, with its journals in two capped standards; the lifter on its belly over the ring of three cams on the main; the head over the cast anvil, and the prop that holds the helve clear.

How the real one worked, and why it was built so

A puddled ball leaves the furnace as a spongy mass of iron full of liquid cinder. Hammering it while it is still hot consolidates the iron and drives the cinder out, and shapes it to go to the rolls (Truran p. 218, 231). Each ball took 15 to 25 blows in 18 to 30 seconds to become a bloom (Truran p. 219).

Three ways to lift a beam, after Bauerman's classes (Bauerman p. 288, Percy pp. 693-695). The lever is the same in all three; what changes is where the cam pushes. Pushing nearer the axis needs a stronger push and a shorter travel for the same lift at the head, and it clears the space round the anvil.
Three ways to lift a beam, after Bauerman's classes (Bauerman p. 288, Percy pp. 693-695). The lever is the same in all three; what changes is where the cam pushes. Pushing nearer the axis needs a stronger push and a shorter travel for the same lift at the head, and it clears the space round the anvil.

The cam and the head on the same side of the axis make a helve; the cam on the opposite side makes a tilt hammer. Tilt hammers were kept small and fast, for drawing bars and finishing work, with heads of about 5 cwt at most (Bauerman p. 288). Their helve was usually of wood, hooped with iron (Bauerman p. 288, Overman p. 336).

The tilt hammer, the old forge machine (Overman p. 335, Fig. 95). The cast head a is wedged on a helve b of hickory or white oak. The cams l, set in the wheel m, press down the short tail at the far side of the axis, so the head rises; a shaft k carries them and a water wheel drives it. Heads of 300 to 400 lb served blooms of 60 to 100 lb (Overman p. 334).
The tilt hammer, the old forge machine (Overman p. 335, Fig. 95). The cast head a is wedged on a helve b of hickory or white oak. The cams l, set in the wheel m, press down the short tail at the far side of the axis, so the head rises; a shaft k carries them and a water wheel drives it. Heads of 300 to 400 lb served blooms of 60 to 100 lb (Overman p. 334).

Lifting the full weight of a helve at its nose is the simplest arrangement, and so the commonest (Percy p. 695). The belly helve moves the lift back along the beam, near the middle (Bauerman p. 288). The cam shaft, the heaviest part of the machine, then lies behind the anvil instead of beside it, and the workers get three sides of the anvil to stand at (Percy p. 695, Bauerman p. 288). That is the whole reason for the form; everything else is the same as on a head-lift helve.

A head-lift helve, Bauerman's Fig. 28 (Bauerman p. 290): a 70 cwt helve on water power. The cam ring lifts five times a turn and raises the head 16 in, and a 25 horsepower water wheel drives it (Bauerman p. 290). The beam, framing and timber bedding are the type; a belly helve differs in where the cam ring stands.
A head-lift helve, Bauerman's Fig. 28 (Bauerman p. 290): a 70 cwt helve on water power. The cam ring lifts five times a turn and raises the head 16 in, and a 25 horsepower water wheel drives it (Bauerman p. 290). The beam, framing and timber bedding are the type; a belly helve differs in where the cam ring stands.

A helve was cast in one piece and is T-shaped in plan, the cross arm making the pivot (Truran p. 218, Bauerman p. 289, Percy p. 694). The long cross arm is there to keep the mass from shifting in its bearings, which are open (Bauerman p. 289), and to resist the sideways push of the cams (Percy p. 694). The face is wrought iron, keyed or dovetailed into a seat so it can be changed (Bauerman p. 289, Percy p. 694), and it needed changing: a face in constant work lasted from three to seven days (Percy p. 695). The anvil is a separate block of cast iron, several times the weight of the hammer, standing alone so that its shaking does not reach the bearings (Bauerman pp. 289-290).

Dowlais helve from above, Truran's Plate 63, Fig. 404, with the head in Fig. 405: the T plan, with the cross arm and its two journals at the right and the head at the left, its round socket for the hammer face and a cast lug beside it. Fig. 405 is the head from the side.
Dowlais helve from above, Truran's Plate 63, Fig. 404, with the head in Fig. 405: the T plan, with the cross arm and its two journals at the right and the head at the left, its round socket for the hammer face and a cast lug beside it. Fig. 405 is the head from the side.

How it was really worked

When not at work, the helve was propped up clear of the turning ring on an iron bar under a projection cast for the purpose (Truran p. 218). Bauerman's stop or gag does the same job, between head and anvil, for a reason that matters: the hammer must never fall directly on the anvil (Bauerman p. 289).

The start is a small piece of drill. With the ball on the anvil, a boy held a small iron block under the point, which raised the helve into reach of the wipers; the prop came out, and the helve fell on the ball (Truran pp. 218-219). During the hammering the bloom was set on end for a couple of blows to upset it (Truran p. 218). A helve cannot be coaxed: a ball had to be of tolerably good iron to stand its dead, heavy blow, and a ball that would crumble under it could be worked sound under Nasmyth's hammer, whose blows could be regulated (Percy p. 695).

The helve lost its place from two sides. In the majority of English forges the steam hammer had to a considerable extent replaced it by Percy's day (Percy p. 695), and in Truran's forges the squeezer had almost entirely supplanted the hammer, at under half the first cost (Truran p. 219). Bar iron still wanted the same force at every blow, and for that the mill hammer of the same pattern held on over the steam hammer (Truran p. 231).

Period basis, and what is the game's own

  • Period: the belly helve itself, a T-shaped casting with its cross arm in the standards, lifted between the axis and the head, its wrought lifter and face seated by dovetails (Percy p. 694). The game draws this, and makes its lifter and faces separate smithed parts as the Mersey helve had them.
  • Period: the anvil is apart from the hammer and much heavier (Bauerman pp. 289-290). The game's helve has its own cast anvil on a stock, about 4.4 t together under a moving 1.7 t, the stock sunk in the floor on an oak bed.
  • Period: the helve was propped clear of the turning ring and let down by hand (Truran p. 218). The game keeps the prop, a post with a handle, and it holds the helve clear of the cams with the ring still turning.
  • Period: open bearings on the cross arm (Bauerman p. 289, Percy p. 694). The game's caps are its own: see the open question below.
  • Period: the cam shaft was placed below the floor on a belly helve (Bauerman p. 288). The game's ring stands in a cell at floor level, on the main that runs through it.
  • The game's own: nobody stands at the work. A shingler, a boy with the iron block, and a helper bringing the ball are all replaced by placing the work on the pad and knocking the prop out. Vanilla's helve also works by itself.
  • The game's own: the ball is worked only at 750 degrees C or hotter, and a colder ball takes the blows and does not change. Real iron welds and consolidates hot; a single threshold is the game's.
  • The game's own: no ball crumbles under this helve.
  • The game's own: the faces do not wear yet. A period face was changed within the week (Percy p. 695).
  • The game's own: there is no squeezer (Truran p. 219 has it taking most of the work); the game shingles under the helve, and the steam hammer takes the pile forgings.

The numbers

The helve follows period practice with the sizes rounded to whole blocks and voxels at true scale; see the scaling page and scale and units. Every game value here is planned.

In the gamePeriod figureHow it was scaledGame value
Moving mass10 tons at Mersey (Percy p. 694); usual 30 cwt to 5 tons (Bauerman p. 289)Drawn casting at true density; the low end of the usual rangePlanned 1.74 t
Fall of the head16 in average (Truran p. 219); 16 to 20 in (Bauerman p. 289); 18 to 24 in on the 10 ton helve (Percy p. 694)Whole voxels at the head, cam travel by the leverPlanned 15 in (0.375 m)
Where the lift acts"about midway" (Bauerman p. 288)One third along, the lever 3 to 1Planned, 1.0 m of 3.0 m
Cam ring speed18 to 19 turns a minute (Truran p. 219)Set by the main; 2 rad/s is 19 rpmPlanned 19 rpm at the main's usual speed
CamsFour (Truran p. 219); five (Bauerman p. 290)The game's own countPlanned 3
Cam ring size5 ft across (Truran p. 218)Whole voxels in the ring's cellPlanned 2.8 ft (0.84 m) at the tips
Blows a minute72 to 76 (Truran p. 219); 70 to 100 (Bauerman p. 289); up to 60 at Mersey (Percy p. 694)3 cams times the main's speedPlanned 57 at 19 rpm, 105 at the Watt engine's 35 rpm
Energy a blowNone givenThe game's own: the helve's weight through its liftPlanned 3.21 kJ; the game's own, against 1.4 kJ for the vanilla helve
Power25 hp water wheel, 70 cwt helve (Bauerman p. 290)The game's own load from the energy of a blowPlanned 3.07 kW (4.1 hp) at 2 rad/s
Anvil and stock5 to 6 tons under 5 to 7 tons of helve (Truran p. 218); "several times" the hammer (Bauerman p. 289)Drawn castingsPlanned about 4.4 t, 2.5 times the moving mass
Time on a ball15 to 25 blows, 18 to 30 s (Truran p. 219)The game's own count, two steps a blowPlanned 40 blows, 42 s for a one-ball plate; 80 blows, 84 s for a two-ball bar; the counts will follow the planned redraw of the ball

Open, and what you can tell us

The books say a belly helve's bearings were open (Percy p. 694, Bauerman p. 289), and they say the lifter acted about midway. If the beam's centre of weight lies outside the lifter, and in the game's drawing it does, then while the cam lifts the head the journals should rise out of open bearings. The game therefore keeps the caps. The books do not explain how an open bearing held on a belly helve: whether the real lifter stood past the centre of weight, whether the cross arm was simply heavy enough, or something else. The scan of Percy's drawing of the Mersey helve is too damaged in our copy to measure. If you know a measured drawing or a surviving helve, we would like to hear.

Picture credits