Thursday, 23 April 2015

Merewen's Pentathlon Documentation - Part 5 of 5 (Weaving)

Silk Tablet Woven Band With Double-Face and Multicoloured Double-Sided Brocade

Weaving (Group V) Beginner

Merewen de Sweynesheie
Photo by Master Eirik Andersen.
Goal:  To create my own patterns, to weave with silk for the first time, to try brocade for the first time, to learn how to do 3/1 broken double-faced twill, and later, to learn how to do regular double-faced weave.  This is my third weaving project - I had only done threaded-in 4F/4B projects until now.

1.      Study of Historical Examples
2.      Choices and Materials
3.      Considerations and Troubleshooting
4.      Resources

1. Study of Historical Examples

Width - Of the brocaded bands studied, widths ranged from 1.3 to 18 cm.  70% were <2.5 cm, and 90% <5.1 cm, so 20% were in same width range as mine. (Ecclesiastical Pomp and Aristocratic Circumstance, which I will refer to as EPAC.)

Length - Lengths ranged from 0 to 191 inches (EPAC).

Weft - Although metal weft was typical, silk brocade weft was equally period.
In the 13th century, extant examples show silk wefts in plum, yellow, blue, green, dark grey, beige, red, brick red, orange, green, marine blue, light blue, yellowish beige, brown, and white.
In the 14th century, extant examples show silk wefts in red, pink, rose, yellow, yellowish green, green, blue, purple, white, beige, and violet
In the 15th century, extant examples show silk wefts in red, rose, salmon, golden yellow, green, light green, blue, light blue, white, black, and ivory  (EPAC).

Warp - Silk was one of the most common types of warp material for brocaded bands.  Popular colours for silk warps in the:
- 13th century were red, reddish purple, green, dark blue, rosy beige, brown, brownish black, rose pink, pink, dull red, reddish brown, rusty yellow-orange, yellow, yellow-gold, green, blue, bluish dark brown, buff, straw, natural, blackish brown, and black.
- 14th century were red, yellow, green, greenish blue, greenish-grey, blue, dark blue, bluish grey, bluish purple, brownish red, mid-brown, brownish beige, beige, reddish brown, old pink, yellow-brown, purple, white, dark golden brown, grey, and possibly black.
- 15th century were red, salmon, yellow, green, white, buff, and dark brown.
It was also popular to use linen on the back of the warp to save money by using cheaper materials where they wouldn't show (EPAC).

Photo by Master Eirik Andersen.
Material Type - Finer silk threads were generally used.

Threading - 80% alternated S and Z threading, 17% were threaded in the same direction with edges and borders different (at least two border cards per side).

Weaving technique - generally forward as far as twist buildup will allow and then backwards as long as twist buildup will allow, switching back and forth.

Tablets - Wood, bone, leather, antler, ivory, or bronze, tablets would have been used historically.

Loom - An upright loom like the Oseberg loom could have been used, or a warp-weighted loom.  Generally the people performing this work made very few mistakes, so they are believed to have been very experienced.  Given that, I believe they probably would have had quality tools.

2. Choices and Materials

Loom and cards - Inkle loom (not period for this work), and cardboard cards.
This was what I had available to me, so it was what I used.

Materials - Stranded silk threads and silk yarns.  White silk yarn was three ply and blue, gold, and green were 2-ply.
This was what I had available to me, and I could not afford to buy new materials.  I only had four colours of silk thread (untwisted), I didn't feel the colours I have had enough variety to be easily visible against one another (dark grey, silver, medium grey blue, dark blue), and I couldn't buy more.

Weft Type - Silk
Band weft:  Two strands of dark grey-blue stranded silk thread (1).
Brocade Wefts:  Silk brocade wefts. On the front, one length of blue silk yarn untwisted into its two component parts to make the weft wider (6).  On the back, green (7), blue (6) and yellow (4) yarns were treated similarly to the blue on the front.  Silver (5) was made up of 12 strands of stranded silk (eight are shown on the example card).

Warp - Silk
White (3), blue (2), and golden yellow (4) silk warp yarns.
I would have used linen on the back of the brocade if I had any, as that was another typical use to save money, but as I didn't have the materials do this, I decided to make the band double-sided instead.

Threading - S/Z alternating threading in center 17 cards, opposite threading to what was immediately beside it on the 2 border cards immediately beside this, alternate threading to THAT on the 4 center border cards, and opposite threading to that on the other two (outside) border cards.  So:
I felt that this would fit with historical precedent and also create stability on the borders.

Method of hiding the brocade weft - bringing the brocade weft to the opposite side of the band, near the edges.  For the front brocade weft, I passed it to the back within the blue border so that the carry on the back side would be invisible.  For the back brocade weft, I brought it to the front at the border between the white and blue backgrounds.  I also removed the back brocade wefts when not in use.

Weaving technique - forward as far as possible and then backward as far as possible, alternating, on 1-2, 7-27, and 32-33.  Double-faced turning pattern on 3-6 and 28-31.
I wanted to create a woven-in pattern.  I was unable to do this on the main part of the band while brocading, so I snuck it into four cards on each border.

Pattern -
Front: Double-faced wave pattern on borders.  Blue center pattern is from 14th century Qaran  illustration, reworked to fit my 17-card center width and a shorter pattern length.  (Page 156, Here Be Wyverns, original below)
Back: Four different motifs evenly spaced.  The yellow and gold motif is based on the center of an extant motif in Here Be Wyverns, page 167.
The shield motif is based on a 14th century Belgian motif found in EPAC, page 128, modified to a chevron pattern as per page 89.

The checkerboard pattern is also based on an extant German design (below) from EPAC,

 modified to fit the 17-card-wide pattern, and the tree is my own design, based on my arms.  The idea of alternating motifs with spaces in between was based on various extant examples found in EPAC, including the German band on page 245:

Time spent - Spent about 8 hours learning about brocade and patterns, spent about 10 hours choosing, designing, and reworking patterns the first time.  Spent about 9 hours winding silk, cutting threads, separating blue silk into its two components, warping, and rewarping the loom.  Spent 8 hours learning about twill by reading and trying and taking out when it didn't work.  First brocaded row took 45 minutes (the row before the tree starts).  Spent another 15 hours reworking patterns.  Spent a week and a half practicing and working out the twill weave patterns.  Spent two days learning how to do regular double-faced tablet weave and designing my patterns.  After that I gave up keeping track of time.  Overall, the project took me well over a month of almost solid work, between research, learning, design, weaving, and finishing.

3. Considerations and Troubleshooting

a)      Width and material use:  I wanted to make the band as narrow as possible to be accurate to more extant sources, so I tried to split the blue warp threads on the borders into their two separate plies.  This resulted in warp threads binding together and creating knots, difficulty in turning cards, fraying, and breakage.  I had to abandon this idea, so I rethreaded the loom's borders with whole blue yarn, and kept the split yarn for brocade wefts and embroidery.
Photo by Merewen de Sweynesheie.
b)      Colour choice:  Although I was following the list of colours that were period for my warp and
      weft threads, I also looked into period dyeing to try to choose the most accurate colours to what could have conceivably been dyed, given the colour options I had available to me, which were minimal.
c)      Threading:  In order to decide whether to alternate S/Z threading or use all the same direction threading, I used bamboo skewers to hold the pattern and wove a little bit in one threading manner.  I then flipped the cards and tried the other method, choosing the one I liked the look of best.
d)      Border technique:  I wanted to try 3/1 broken double-faced twill.  I worked hard at designing a border pattern that would accommodate this technique (since it requires two turns per colour before switching).  It took a while to figure out the proper pattern - these were some of my early attempts.  You can see that the angles of the threads near the ends would alternate being correct and incorrect.
Photo by Merewen de Sweynesheie.
At first, when I switched to the cotton for troubleshooting, I worked with only four cards and I had some issues with the weft thread showing through.  I tightened it up, and once I had taught myself how to do the technique, I tried to teach myself how to design a 3/1 broken double-faced twill pattern, and tried to create two that worked (to decide between).  I managed to create one threaded-in pattern that I thought was correct (A), but I found that I was unhappy with the look of the reverse side.  I was also bothered by the long floats in the red on the sides.  Through consulting with people online and further practice and study, I learned that there was no way the resources available to me knew of for me to effectively reduce the float and keep the tight pattern.  Also, I learned that the back would not be able to be improved.  Since I wanted to create a double-sided band, I had to abandon this technique and switch to double-faced weave.  I switched to a new band and added two border cards to each side to help reduce the long float issue and add stability to the band.  I had never done double-faced technique before either, so I again tried to teach myself how to do the technique.  Working on my own, I did not successfully emulate the technique, but I created two threaded-in patterns that looked appropriate (B and C).   The wave pattern was looking a little cleaner than the other design, so I chose that one.  I then tried both patterns in both s/z threading (B and C) and s threading (C and D) to see which would look better (as an average on both sides).  I chose basic s-threading as the best look, and reversed the pattern for the other side of the band.

e)      Border pattern:  I originally wanted to make gold bezants in the border to match my arms, but they did not work effectively with the 3/1 broken double-faced twill design, since the border was only four cards wide and the pattern would have to be at least four turns long, making ovals, rather than circles.  Thus I abandoned that option.  I then considered a number of other options before settling on two.  I created 3/1 twill patterns in both, and then double-faced patterns in both before settling on the wave design (which creates yellow triangles on the reverse side of the band, as its blue wave blends into the blue border cards).
f)       Brocading the yellow wave:  I wanted to try brocading over the yellow wave to make it more pronounced, but in practice the area was not wide enough to effectively brocade, so I removed the thread and abandoned that option.
g)      Brocading behind the alternating motifs:  I wanted to add a muted brocaded background behind the individual motifs to add some continuity, so I tried this in silver thread.  The silver thread blended into the white and was indistinguishable.  It also made the white threads look bunched up, so I removed it.  It was early on in my weaving, and I was still working on getting everything right, so I ended up taking that weaving out for other reasons as well.
h)      Designing the tree motif:  My original tree design was as wide as I wanted the finished tree to be.  After starting to weave the pattern, I realized that the tree was going to be stretched out insanely wide, so I reworked the pattern to make it narrower.  I also added tie-downs to the tree pattern at appropriate intervals to hold down the brocade weft without detracting too much from the visual design.
i)        Brocade thread coverage:  It was suggested after I started that I would get better brocade coverage if I ran the two same-coloured brocade wefts through individually on separate bobbins.  I tried this with the blue brocade weft, but the threads were too weak individually and kept breaking, so I had to go back to putting them on the same bobbin.
j)        Which side up:  I originally meant to do the various motifs on the top of the band and the continuous brocade on the bottom.  In practice, it was far easier to run the continuous brocade on the top, even though the individual motifs often contained two colours.  After a little while, I started using a mirror to help me check my work.
Photo by Merewen de Sweynesheie.
k)     Keeping track using pattern design:  While designing my patterns, I was conscious of the fact that there was a lot going on in this band, and that it would be difficult to keep track of.  In order to make things easier, I settled on a 30-turn continuous blue brocade.  In the center of this brocade, I would place one motif on the back side, centered on the center of the solid brocade pattern.  In order to keep track of the border pattern, I tried to make it 10 or 15 cards long, but it would only repeat cleanly at 20 cards long.  10 turns long would result in some cards being opposite to the place that they started the first time.  So for every two repeats of the continuous blue brocade, I went through the border wave pattern three times.  Finally, I timed my switch from forward to backward turning of the cards in the center and borders to line up with the continuous blue brocade.  The first time through was in one direction, and after that I switched direction after every two repeats of the blue brocade pattern.
l)        Using the warp to test and rework the pattern:  After about 8 times through while practicing and reworking my designs, I noticed that the bamboo skewers were starting to fray my warp threads.  I set up a separate test warp at the back of the loom, using cotton, and worked on the border pattern there until I was happy with the outcome.
m)   Getting the most brocade for my work:  After one time through, I realized that the continuous blue brocade would show slightly more blue if I inverted the pattern.  So for most of the remainder of the band I used the inverted pattern.  In order to make the band symmetrical, I repeated the original (non-inverted) pattern at the far end.  While I was unable to find a direct precedent for this, brocaded bands did often have different ends, and many continuous brocades had sections where the pattern switched colours.
n)      Gappiness:  Part of the "gappy" nature of the brocade comes from the size of threads I had to use for my warp (supply), and the fact that at times there are up to 9 brocade wefts traveling through the center of the band (2 on the all-blue side, two green, blue, or yellow on the reverse side, and five thin silver threads).  The band ended up very tight - most likely because I was beating so hard to try to bring the brocade weft pattern together.
o)      Starting cleanly:  I put a plastic ruler into the shed in the beginning to start my weaving as close to the proper width and tidiness as possible.  Anything thin, hard, and straight would have worked.
p)      Finishing the end:  I found ecclesiastical bands that were finished in knotwork (EPAC Page 24), and I liked the design, so I decided to try it, even though my band is in no way ecclesiastical.  Both ends are identically finished.

 4. Resources

An Introduction to Period Tabletweaving, accessed January 2015.

Beginning Double-Faced Tablet Weaving, accessed January 2015.

Double Faced 3/1 Broken Twill, accessed January 2015.

Double Faced 3/1 Broken Twill, accessed January 2015.

Hendrickson, Linda, Double-Faced Tablet Weaving, accessed January 2015.

Natural Dyeing, accessed January 2015.

Sebolt, Cynthia, Hallstatt Tablet Weaving, accessed January 2015.

Spies, Nancy, Ecclesiastical Pomp and Aristocratic Circumstance, 2000.  Arelate Studio, Jarrettsville, Maryland, USA.  Pages 24, 57-76, 82-83, 89, 93, 96-97, 106-108, 127-130, 167, 198, 245.

Spies, Nancy, Here be Wyverns, 2002.  Arelate Studio, Jarrettsville, Maryland, USA.  Pages 135, 156.

Tablet Weaving Theory, accessed January 2015.

No comments:

Post a Comment