SHRINKAGE & STRENGTH EXPERIMENTS

The experimentation with the process of Cuir Bouilli, and the eventual creation of near-authentic period armour, was all undertaken on my four-element electric kitchen stove.

Water was placed in the biggest enamelled-steel roasting dish I could get onto the back right element (which meant I had to �prop up� the edge of the pan where it reached beyond the element). The water was brought up to a �rolling boil�, and then the leather added to the water.

When the leather was removed from the water it was placed under a heavy wooden chopping block (weighted with a big rock) while it dried so that the piece laid flat.

1. Originally 4 cm x 4 cm bark-tanned leather (3 mm [7-8 oz] "russet")

2. Originally 2 cm x 2 cm bark-tanned leather (3 mm [7-8 oz] "russet")

3. Originally 4 cm x 4 cm chrome-tanned leather (1.2 mm [3 oz] dress leather)

Columns:

First column (A): 20 seconds in boiling water

Second column (B): 40 seconds in boiling water

"Holes": 40 seconds

All bark-tanned pieces reached polymerisation temperature and therefore gained significant hardness. As part of this process each lost a percentage of length and width (i.e., they shrank!), and became significantly thicker. In some cases this resulted in the skin side becoming slightly smaller than the flesh side and pressing with weights was important to prevent curling at the edges.

The next step was to check out the shrinkage calculations on real pieces. I planned to build cuir bouilli Finger Gauntlets on the Wisby model. The pattern used is based on ones for steel gauntlets ² .

The shrinkage (barring the back-of-hand piece) in height (down the hide) ranged from 56 - 92%, with the majority at the minimal shrinkage end of the range.

However, the width measurement (along the hide - from head to tail) actually increased in most cases (one exception was 99% of original size). The increases ranged from 102 - 116%! The only explanation I could find was that, unlike the first test where pieces were simply pressed under weights, each piece was moulded to shape along a broom handle and was stretched in the process. Hardness is at a good level.

The other oddity is the back-of-hand piece which shrank to 66% in height and 51% in width. I have no suggestion to offer on this one.

After a break from experimenting for four months, I decided to check whether larger pieces shrink at the same rate as the smaller pieces in the first shrinkage test (With hindsight, I should have done this before doing the gauntlet testing!).

• There are distortions on the two pieces on the left - these were originally square and the hide was deliberately cut from areas with significant tick holes. These and the vaguely triangular piece and the round were only immersed for 20 seconds.
• The pieces at the top right were cut to assess how the direction of grain affected the �curl� of the leather. The strip on the left was cut down the grain and the strip on the right along the grain (see #4, Lessons, below); both curled down the grain.
• The pieces at the bottom right were briefly soaked in cold water before boiling. What shrinkage there was appeared consistent in all directions. Unfortunately I forgot to measure them first so this part of the test is repeated as �Third Shrinkage Test�!

This test incorporated the lessons learnt in previous tests. This time the test was to compare results between leather placed skin down into boiling water vs skin up in simmering ⁴ water as had formed the impression that skin up in boiling water curled and distorted more.

Those in the left column were all briefly soaked in cold water (as above) and placed in water for 20 seconds.

• top left: boiling - skin down
• middle left: simmering - skin up
• bottom left: boiling - skin down

Conclusion: leather placed skin down in boiling water shrinks more evenly than leather placed skin up in simmering water.

Leather on right was not soaked and so was placed dry in simmering water for 20 seconds. Uneven shrinkage resulted even though only simmering.

Results interesting!

• the 4.5 x 4.5 cm piece shrank and gained hardness in a similar fashion to pieces previously tested.
• the larger piece (10 x 10 cm) only shrank a minor amount after two immersions at 180° for 20 seconds each and an additional one for 40 seconds. Although it hardened it still wasn�t as hard as the smaller pieces.
• The third piece was cut to the pattern of the Finger Gauntlet back-of-hand shape plus 1 cm extra in each direction as this had shrunken so hugely in the previous test. It was also cut so that the curve of the hand was aligned down the hide (see photo). The piece required three immersions - once at 20 seconds, and then twice by eye (judging curl rate). The outcome was almost no shrinkage in any direction with a moderately hard result. (This means the piece is far too big!)
• the cake rack significantly reduced overheating, though some minor marks from the rack itself showed. These don't seem to affect strength or cause distortion.

2. Cold Soak First: Briefly soaking the leather piece in cold water until the bubbles stopped reduced shrinkage and extended time the piece could be in the boiling water. The latter point proved useful when shaping larger pieces where the leather had to be revolved through the water as it couldn�t be all immersed at once.

3. Avoid direct contact with pan bottom: The leather scorched easily on the bottom of the pan, so I placed a metal cooling rack (as used for cakes etc) in the bottom. If the temperature was too high, or the leather sat in one place too long, minor scorching still took place but it was restricted to the fine lines of the rack and didn�t seem to affect the outcome.

4. Maintain consistent direction: Cuir bouilli has a tendency to shrink at a different rate depending on the "grain", i.e., the shrinkage rate down the hide (mid-back to leg) is different than along the hide (head to tail). It pays therefore to align pieces the same way as you can allow for the same amount of shrinkage in each direction.

You can also utilise this property when you want curves, e.g., for the fingers of the gauntlets I set the width in alignment with the greatest shrinkage as I could then use the tendency of the leather to curl in that direction to shape the curve of the fingers.

5. Watch out for �False Economy�: I thought to save money while I learnt so purchased a "second grade" hide. This proved to be false economy as cuts, scrapes, and tick holes in the hide resulted in uneven shrinkage leading to some cuir bouillied pieces resulting in either weakened pieces or distortions. As well, some of the hide cannot be used at all for cuir bouilli (though it is OK for some plain leather work).

6. Finally, my thanks to all the folks who have shared their expertise on the Internet, and Marc Carlson and David Friedman in particular. I�d like to re-iterate Marc�s opening comments on his very useful page as my results bear him out:

FOOTNOTES
¹ Experiments inspired by David Friedman's (Cariadoc) comments, pg 7, on Marc Carlson's Leather page. http://www.personal.utulsa.edu/~marc-carlson/leather/hl.html Return to text

² Adapted from Master Cad's (M Davis) original site, now found at: http://www.mallet-argent.com/images/wisby_gauntlet_pattern.jpg Return to text

³ Averaging shrinkage figures based on the 20 second immersion tests seemed relatively safe, though this would have been less so if the 40 second immersion had been chosen as the range of shrinkage was much wider. Return to text

⁴ Simmering water is water brought up to boil, then heat turned down to �, slightly bubbling, vs boiling water, which is water brought up to rolling boil and then held just below that at 2/3rds mark. Return to text

⁵ "Leatherworking in the Middle Ages"; Compiled and edited by I. Marc Carlson Revision 3 March 2003, http://www.personal.utulsa.edu/~marc-carlson/leather/hl.html (since revised 2004) Return to text