Chinese Tinkers

Donald B. Wagner
3 August 2009

The cast iron wok is one of the wonders of traditional Chinese technology – to see how they they were cast, see here. With its use follows another wonder: the itinerant tinker who repairs broken woks with molten iron from a small furnace.

The tinkers were a fascinating sight for foreign visitors in China, and we have some excellent detailed descriptions of their work from the 18th, 19th, and early 20th centuries. I have copied the best of these below.

In Guangzhou in the 18th and 19th centuries several large studios mass-produced souvenir paintings for sale to foreign visitors. The painting reproduced immediately below is a gouache from one of these studios showing a tinker at work.

I was amazed and pleased when my friend Liu Peifeng recently showed me the photograph by Pei Chishan shown further below, which shows a tinker with his equipment in 2006.

Gouache painting by an unknown Chinese artist in Guangzhou, mid-19th century. (The Royal Library, Copenhagen, Denmark, Ny kongl. Saml. 346b, 2°, no. 33).

Photograph by Pei Chishan 裴 池善 in Zezhou 泽 州, Shanxi, April 2006, reproduced with his permission, kindly arranged by Liu Peifeng 刘 培锋.

The earliest description of the traditional technique of wok-repair is by the Dutch diplomat Van Braam, in 1795, given here in English translation. The  original French text can be seen here. The resemblance of his illustration to the Chinese gouache above shows that he had seen an earlier copy of the same picture.

André Everard Van Braam, An Authentic Account of the Embassy of the Dutch East-India Company to the Court of the Emperor of China in the Years 1794 and 1795 . . . , vol. 2, London 1798, pp. 78–79.

17 February 1795

During our short stay this morning in the village of Fan-koun [near Beijing], I had the opportunity of seeing a tinker execute what I believe is unknown in Europe. He mended and soldered frying-pans of cast iron that were cracked and full of holes, and restored them to their primitive state, so that they became as serviceable as ever. He even took so little pains to effect this, and succeeded so speedily as to excite my astonishment. It must appear impossible to any one who has not been witness to the process.

All the apparatus of the workman consists in a little box sixteen inches [40 cm] long, six inches [15 cm] wide, and and eighteen inches [45 cm] in depth, divided into two parts. The upper contains three drawers, with the necessary ingredients; in the lower is a bellows, which, when fire is wanted, is adapted to a furnace eight inches [20 cm] long and four inches [10 cm] wide. The crucibles for melting the small pieces of iron intended to serve as solder are a little larger than the bowl of a common tobacco pipe, and of the same earth of which they are made in Europe; thus the whole business of soldering is executed [in three or four minutes].

The workman receives the melted matter out of the crucible upon a piece of wet paper, approaches it to one of the holes or cracks in the frying-pan, and applies it there, while his assistant smooths it over by scraping the surface, and afterwards rubs it with a bit of wet linen. The number of crucibles which have been deemed necessary are thus successively emptied in order to stop up all the holes with the melted iron, which consolidates and incorporates itself with the broken utensil, and which becomes as good as new.

The furnace which I saw was calculated to contain eight crucibles at a time, and while the fusion was going on was covered with a stone by way of increasing the intensity of the heat.

Van Braam's illustration occurs only in the first French edition (Philadelphia 1797), facing p. 281, not in the later editions, Paris 1798 and London 1798.
Dr Bruce MacLaren of the Peabody Essex Museum in Salem, Massachusetts, has informed me that the original Chinese drawing from which this illustration was taken, originally owned by Van Braam, is now in the collection of the Biblioteca Nazionale Centrale in Florence, Italy.

The next description is by the American Consul in Singapore, Joseph Balestier, in a letter published by the United States Patent Office in 1850.

Report of the Commissioner of Patents for the Year 1850, Part I: Arts and Manufactures
. Washington: Office of Printers to House of Reps., 1851 (31st Congress, 2nd Session, Ex. Doc. No. 32).

Chinese Mode of Repairing Cracked or Broken Vessels of Cast Iron

Perhaps no device can be named more characteristic of oriental ingenuity – of the most mechanical people of the east – than this. It is one that could only have occurred where ages of experience in the treatment of the metals had elapsed. The idea an ordinary artisan fusing iron with a handful of charcoal, and handling the glowing liquid as if it were but melted wax or tallow, would be considered by our founders as belonging rather to romance than to reality. Every year thousands of vessels, large and small, are with us thrown aside – costly sugar pans of planters, and the more capacious vessels of soap boilers and brewers, as well as culinary cauldrons – that might be restored to soundness by this simple method, and at the most trifling charge.

In 1794–5, Van Braam – the second in command of the Dutch embassy to Pekin, and who afterwards settled in the United States, bringing with him the chain pump and other inventions of the Chinese – was exceedingly struck with the operation – its simplicity and efficiency. He appears to have been the first European who observed it.

. . .

Mr. Balestier, United States Consul at Singapore and United States Envoy to South Eastern Asia, during his recent visit to the United States, politely offered to procure any information for the Patent Office – agricultural and mechanical – that might be desired from the East.

Among other subjects of inquiry suggested to him, iron founding was named: an art of remote antiquity and brought to considerable perfection by the Chinese. The excellence and lightness of their hollow ware are proverbial; while thc readiness with which they fuse and handle small quantities of the metal in the reparation of damaged wares excites surprise. The last operation Mr. B. had often witnessed, but without giving special attention to it. He promised, therefore, to minute down the particulars sufficiently in detail to enable any mechanic successfully to perform it. His letter is subjoined, and by it every thing is made clear. Van Braam leads one to suppose the sides of repaired rents are fused together or united as in soldering other metals, while from Mr. B’s account, such is not the fact. The fluid metal in filling the crack is spread over it on both sides of the vessel, and thus forms a species of rivet.

Macao, Feb. 6, 1850.
To Thos. Ewbank, Esq., Commissioner of Patents,
Patent Office, Washington, D. C.

Dear Sir, – According to your desire, I have carefully observed the Chinese manner of re-uniting and joining together cracked or severed cast iron vessels, so as to make them as useful as ever after an accident, which is as follows: referring you to the utensils used, which I have numbered as at foot, and have had put into a box directed to you, and shipped per United States ship St. Mary, Commodore Geisinger, bound home, and which the Commodore will hold subject to your order.

I procured the accompanying cast iron pan, measuring 12 inches [30 cm] in diame­ter, by 4 four inches [10 cm] deep. A crack of 3 inches [7.5 cm] was made in it in the first place, and in the second a piece was entirely broken off: giving rise to two distinct operations.

The operator commenced by breaking the edges of the fractures slightly with a hammer, so as to enlarge the fissures, after which the fractured parts were placed and held in their natural positions by means of wooden braces.

The pan being ready, crucibles made of clay, were laid in charcoal, and ig­nited in a small portable sheet iron furnace, with bellows working horizontally. As soon as the pieces of cast iron with which the crucibles were charged were fused, it was poured on a layer of partly charred husk of rough rice, or paddy, which was previously spread on a thickly doubled cloth, the object of which is to prevent the sudden cooling and hardening of the liquid metal. Whilst in this liquid state it was quickly conveyed with the right hand to the frac­tured part under the vessel, and forced up with a jerk into the enlarged fissure, whilst with the left hand a paper rubber was passed over the obtruding liquid, inside of the vessel, making a strong, substantial and neat operation.

You will thus remark that the art of the Chinese for re-uniting cracked or severed cast iron vessels, of all sizes, consists in cementing them with cast iron, whilst in the liquid state.

I have the honor to be, dear sir,
Your most obed’t serv’t,

J. Balestier

The weight of this pot is 31/4 pounds [1.5 kg]. Except at the centre, where a part two inches [5 cm] over is left thick and flat for a base or foot to rest on, the thickness does not exceed, and in fact scarcely reaches 1/10 of an inch [2.5 mm]. The handles are cast on, but appear to have been first formed and inserted into the mould. This does not seem to have been of sand, as the inner and outer surfaces are smoother, and of a different appearance from iron cast in that material. Of the metal used for repairing this pot, Mr. Balestier has forwarded a lump that was not melted. It is part of an old kettle, and differs but little, if any, from our pot metal.

The crucible, not much larger than a thimble, is made apparently of the same material as our common sand crucibles; except the shape, it could not be distinguished from one of them.

The amount of one fusion seems not to cover more than half an inch of the crack, and hence in the piece inserted, no less than nine distinct applications of the melted metal are seen – resembling in the inside so many ragged wa­fers touching each other, while on the outside, where the metallic plaster was applied, there are the same number of rude protuberances.

Dr. Gale, one of the examiners of the Patent Office, has made, at my request, a chemical examination of a portion of the basin, and finds it a very pure white cast iron, containing scarcely any foreign matter, except a little carbon and silex, ingredients always present in cast iron.

The subjoined figure represents the itinerant artist with his portable forge, at work in the street. The front half of the wooden chest is his Fung Seang [fengxiang 风 箱] or bellows, a description of which may be found in Ewbank’s Hydraulics [pp. 247–249]. Its principle is that of the double acting force pump, and it is constructed wholly of wood, except the valves and packing of the piston, which are paper, and singularly durable. The long coarse file, with a prolonged smooth extremity to slide through a ring, fixed on the chest, is a common accessory to a tinker's budget. By the arrangement, he possesses a tolerably good substitute for a bench and vise, and can increase or diminish the pressure of the file on the object operated on, at pleasure.

A bit later the great metallurgical writer John Percy asked a medical missionary in Beijing, William Lockhart, for further information on this technique.

Chinese Method of Mending Cracked Cast-iron Vessels.

The Chinese use extensively for culinary purposes, such as boiling rice, etc., circular bowl-like vessels or pans of thin cast-iron. These pans were long ago described and figured by Count Rumford (Essays, Political, Economical, and Philosophical, 1802, pp. 298–299); and he alludes to the ingenious method of mending them by itinerant tinkers. At my request, my friend Dr. Lockhart, the indefatigable and well-known Chinese medical missionary, now directing an hospital at Pekin, procured a specimen of one of the mended pans, and the apparatus employed in the process, which he presented to the Museum in Jermyn Street, where they may now be seen. The pans are difficult to make, on account of their thinness, and yet they are commonly manufactured by the Chinese themselves. I have received the following description
from Dr. Lockhart: –

“These boilers are much used for cooking rice and other vegetables, and are made very thin, their chief excellence in the eyes of the Chinese being their thinness, as, owing to this quality, very little firewood is required to boil water in them. Some years ago a large quantity of boilers were made in Birmingham of the same shape as the Chinese vessels; but the natives would not buy them, as they were too thick, and wasted firewood. The boilers being so thin, are very apt to break and crack, in which case they are handed to an artisan, who, with the articles of his trade slung in baskets across his shoulders, goes about the streets, crying out ‘Boilers to mend; Jars and other vessels repaired!’ One of these men is frequently seen repairing a boiler, which is not only cracked, but has a piece broken out of it, say an inch square [6.5 cm2]. He first cleans the edge of the fracture by scraping it with a chisel, and rubbing it with a piece of brick; after which he inverts the vessel and places it on a low tripod stand, so that he can easily use his hands inside and outside of it. By his side are his box of tools and various other articles of his trade. He takes a little crucible, of the size of a thimble, and places in it a small scrap of cast-iron; the crucible is then put into a small furnace about as large as the lower half of a common tumbler (that in the Museum is somewhat larger); this is filled with charcoal, and, by means of bellows, great heat is produced, and the iron is shortly melted. The workman now drops the melted iron on a piece of felt, covered with some charcoal ashes or dust, which he holds in his left hand, introduces it into the inside of the inverted vessel, and presses it against the part to be repaired, at the same instant striking the melted metal that protrudes through the hole or crack with a small roll of felt, covered also with ashes. This process he repeats till the deficiency in the boiler is filled up. He then breaks off any rough projecting parts of the new surface, rubs it over with a piece of broken tile or brick, so as to make it tolerably smooth; and having tried whether his work is perfect, by putting water into the vessel, hands it to the owner, and charges 3d. or 4d. for the job.”

A detailed account of the process by Mr. Balestier, U. S. Consul at Singapore, was published in the United States in 1850. [See above.] It was described in 1794-5 by Van Braam, the second in command of the Dutch embassy to Pekin, who afterwards settled in the United States. [See above.] In this description the globule of molten cast-iron is stated to be poured upon a piece of wet paper; in which case the Chinese availed themselves of the spheroidal state of water. Mr. Balestier’s account agrees substantially with that of Dr. Lockhart. There is no actual soldering;
but the cracks are united by little rivet-like pieces of cast-iron, produced, of course, without hammering.

. . .

The furnace is a little shallow circular open vessel of sheet-iron, 5 1/2 in. [14 cm] in diameter at the top, and 5 1/2 in. high, lined with fire-clay, and having a few little bars for the grate, below which the blast is introduced, exactly in the same manner as in the furnace known as Faraday’s. [See Michael Faraday, Chemical manipulation, 1831, pp. 92ff.]

Rudolf P. Hommel travelled in widely China in the 1920’s and wrote a marvellous book about the traditional Chinese technologies.  He saw a tinker at work in Zhangshu 樟 树, Jiangxi.
Rudolf P. Hommel, China at work: An Illustrated Record of the Primitive Industries of China’s Masses, whose Life is Toil, and thus an Account of Chinese Civilization, New York 1937, pp. 31–32, 34.

Mending of Cast Iron

We had a maltreated cast iron brazier with some pieces broken out, and called the bowl-mender to have it fixed. The day previous I saw him in town going around and collecting bowls to be fixed, so I knew he was about. He came with his outfit and a helper to work his bellows. He put the bellows down on the ground, the round cannon type, about 2 1/2 feet [75 cm] long, and 7 inches [18 cm] in diameter. To fasten it to the ground he produced a piece of rope with a spike at each end. He laid the rope over the bellows and drove the spikes into the ground. The stove, a clay cylinder about 13 inches [33 cm] high and 6 inches [15 cm] in diameter had a hole on the side near the bottom and a metal tube served to carry the draft from the bellows to the stove. The connections were carefully smeared with clay to prevent the air from leaking out. The inner space of the stove had a diameter of about 3 1/2 inches [9 cm]. To start the fire first a few pieces of burnt clay were put into the stove then some lighted straw and on this some small pieces of anthracite coal up to the rim. The bellows were worked from the start of the fire as there was no other way to get air to the bottom of the stove. Soon a brisk fire was going, after poking down the coal the man took an earthenware crucible, warmed it first on top and then put it into the stove and heaped coal all around it. Then he took some small bits of cooking-bowl fragments, put them in the crucible and gradually filled it heaping full with larger pieces. The size of the crucible was about 2 inches [5 cm] in diameter and 2 1/2 inches [6 cm] deep. Over the whole top of the stove he then put a covering of twisted straw smeared with wet clay and over this a fragment of an iron bowl to keep the straw in its place. All the time the helper was pushing the bellows, continuing for about 15 minutes until the iron was melted in the crucible. The man stirred the metal a few times and took off some slag from the top until finally the metal was clean and ready for use. A few bowls were to be mended besides our brazier and to support these pieces three iron pins were driven into the ground and the bowl to be worked on placed upon them. The worker had in front of him a little basket of wood ashes and placed some upon a piece of felt which he held in his left hand. In his right he had a little clay spoon and took from the crucible a spoonful of the liquid iron, placed it upon the felt and pushed it from below into the hole to be mended. At the same time he pressed a short solid cylinder made of cotton cloth rolled tightly together upon the hole from the other side. In this manner the amount of liquid iron was made to completely fill out the hole and the surface became smooth from the pressure exerted from both sides. This is all there is to the mending of a small hole. In our brazier the hole was big and a piece had to be inserted which the mender took from an old bowl. He held it in place with two strips of bamboo as shown in Fig. 49, and then filled the space around it with several applications of the liquid metal until the whole surface was solid again. To my surprise the man mended also a few holes in an enameled wash basin of foreign make. Before the iron in the holes had become cool they were smeared with straw dipped into wet clay. The explanation given was that the iron sticks better in the hole. To smooth the iron in mending cooking bowls sometimes the mender rubs the still red hot iron in the hole with a piece of resin.

The observations were made at Changshu, Kiangsi.

Fig. 49. Mending of cast iron. Sketch showing how a fragment of a cast iron is held in a hole of a cast iron brazier which is to be mended. The outer shaded surface represents the wall of the brazier in which the hole is found. Pieces of bamboo hold the mending piece in place and the space between mending piece and the edge of the hole is filled out by successive applications, one next to the other, of the liquid iron taken from the crucible. As soon as the mending piece is held fast in the hole by a few applications of the liquid cast iron (which cools as soon as applied) the bamboo strips are removed.

This page was last revised 17 August 2009 by Don Wagner.

Addition, 29 May 2012:

My friend Paul Rondelez has pointed out to me that a similar technique was used by itinerant tinkers – perhaps Gypsies – in Ireland in the 19th century. P. W. Joyce, writing in 1905 of a reconstruction of an early Irish type of furnace, ends his article:

I may add that in my young days I have seen, in the County Limerick, furnaces of much the same kind as that described above, used by wandering tinkers, who also practised foundry on a small and simple scale; but they used anthracite coal, not wood-charcoal. They made up with their hands, in a rough-and-ready way, and in a few moments, a small furnace of moist clay, which they placed securely in a wooden frame, and into which they fixed the pipe of their bellows. By means of this rude contrivance they succeeded in melting small fragments of cast iron, with which they mended – very roughly, indeed, but quite effectively – pots and pans, or other cast-iron articles, that had been gapped or cracked. They formed a strong mould of moist clay round the broken part, into which they poured the white molten metal, which firmly adhered on cooling. The women of the several houses always put aside their broken vessels, waiting for the next visit of the tinker company, who never failed to find plenty to do in every hamlet.

P. W. Joyce, ‘The Old Irish Blacksmith’s Furnace’, The Journal of the Royal Society of Antiquaries of Ireland, Fifth Series, Vol. 35, No. 4, [Fifth Series, Vol. 15] (Dec. 31, 1905), pp. 407-408.