My tangled web, weaving cloth and trim

I enjoy weaving, both cloth and narrow wares and I enjoy working with natural dyes as well as exploring medieval cookery and garb construction and combining my interests into an item, making a simple object into a broader project
Sometimes I work towards creating a specific item to compete an outfit and other times I create an item just out of curiosity – can I do this? What if I do this… Sometimes this works and othertimes….

A linen tablet woven band, as one discovered in a field in Cambridge


To dye thread and weave a band in the fashion of the “Cambridge Diamonds” artifact using linen dyed with woad as the original object
While I am an experienced tablet weaver I had not used the pack idle technique.
I had worked previously with an indigo vat, but not woad and never on linen fibre.
I worked in steps- starting with a simpler band using the same technique, then working with commercial dyed linen and finally weaving a band dyed with woad grown from seed.

Introduction and discussion on the found object

During excavations in a cemetery at St Johns Cricket Field in Cambridge a small bronze strap end was found in a cinerary urn and was assumed to be from the early Anglo Saxon period. The strap end and the small woven fragment has no firing marks on it.1 The strap end is simple and described as similar to one located in Fakenham and dated to the sixth century. 2

The fragment is unique in several ways- it appears to be of dyed linen in a “woven in” pattern  where the design is created from specific threading through the cards rather than the more complex brocade or twill designs. 3

Image of Cambridge band
Bronze strap end with tablet woven fragment, St Johns Cambridge

It also has a unusual weave structure, created using what Collingwood4 refers to as “pack idling.”  There are very few examples of bands created in this technique.5

This band if often referred to as the Cambridge Diamonds or Anglo-Saxon Diamonds when referenced on websites and weaving forums.

Grace Crowfoot examined the strap end in the early 1950’s along with several other weavings from the Anglo-Saxon era, she places this band as being from this period and feels it is the earliest example of Anglo-Saxon woven in patterned tablet weaving as well as the first instance of a textile belt.6

Gale Owen-Crocker dates the strap end and belt fragment to the 7th century. 7

Penelope Walton Rogers disagrees with this dating feeling that the weaving technique and strap end are early medieval. 8

Other examples created in this pack idling method, two are 13th century brocade pieces with the pack idle as the background and one is from the early medieval period in a chevron pattern and found in Felixstowe.9

The Cambridge fragment is 1.1 cm wide and only 1.5cm long. As the fragment is so small it has not been removed from the strap end so the reverse side has not been analyzed, nor has the fibre been tested for composition or dye. Determination of the use of dye, the fibre,  and the weaving technique has been done by visual study only.10

The fibre is bast, likely linen and appears to be woven in a diamond pattern in two shades of blue, dyed with woad or indigo,  and white. 11

Crowfoots sketch of band
Crowfoots drawing of the Cambridge Band 13

Shelagh Lewins was able to examine the fragment and feels that the the colours are too regular to be caused from staining or oxidation from the strap end as they follow the diamond pattern.12 The outer half diamonds are a greenish blue and the inner diamond is a lighter shade of blue outlined in white or natural.

The weave structure is flatter than normal tablet weaving and lacks the usual twisted cord or v of two cords meeting, instead the warps interlace like a Y creating a very thick braid that almost resembles modern webbing.

Felixstowe Band
Crowfoots sketch of the Felixstowe Band16

Crowfoot recreated the pattern using a technique where a weft is throw from left to right, the odd cards rotated a 1/4 turn, then the weft is thrown from right to left and the even cards are twisted a 1/4 turn- on every pick half the cards sit “idle”. 14This is often called the “Anglo-Saxon” style in tablet weaving discussion despite this being the only known example of this technique from this era.15 I prefer Collingwoods term “pack idling”.

This pack idling technique creates a very thick sturdy band, much narrower than you would normally achieve with the number of cards needed as the threads interlace and form essentially a double band held with a single weft. The reverse of the band is a slightly offset version of the front, making it a great technique for belts or straps due as it is both reversible as well as very strong.

I find it surprising that this band is considered to be from the Anglo-Saxon period when this structure has only been seen in other bands dated to the medieval period. One of those bands, known as the Felixstowe band, was also in hemp or flax and was a simple chevron design.

The Felixstowe band was attached to a bronze buckle dated to the 14th or 15th century. The band is a simple chevron pattern in white and brown but created with this same pack idle technique.17

My project with test bands, dyeing the fibre and final tablet woven band

As I had never worked in the pack idle technique and had limited experience working with linen threads I started with the simpler appearing Felxstowe Band before moving on to the Diamond pattern.

Felixstowe Band as a test of the pack idle technique

Using a 2 wet spun 2 ply linen in lace weight (2/14) in natural unbleached and white I threaded my cards as per the pattern  in Crowfoots paper A Medieval Tablet Woven Braid from a Buckle Found at Felixstowe.

image of crowfoots pattern of the Felixstowe band
Crowfoots formula to reconstruct the Felixstowe band19

I did my weaving on an inkle loom as it is more comfortable for me to work on than a more period upright or oseberg style loom that may have been used by weavers of the period.

I used wooden cards on the loom and a polished wooden shuttle to beat my weft.

Separating the cards into two packs, odds forward and evens slightly back allowed me to easily keep track of which pack of cards to turn.

Band on loom
Separating packs while weaving Felixstowe band

After wet finishing the band was 9mm wide and has a hand that appears unique to linen, both firm yet drapey. The band is roughly twice as thick as a typical tablet woven band of the same fibre as well as much narrower than one made from the same number of cards.

Weaving the diamond pattern in commercial dyed linen yarn

For this attempt I used a pattern located on the web page of Shelagh Lewins. Her pattern is based off the one initially written up by Crowfoot but it contains a slightly better diagram of the card threading.

Using this threading diagram and two shades of commercially dyed blue linen and a natural unbleached linen I again used my inkle loom and wooden cards.

After wet finish my band is 11mm wide, the same width of the original.

As the turning direction for this band is all forward using an inkle loom means twist will build up on the warp so  I had to reverse the pattern somehow to be able to use all my warp threads. 

If you flip all the cards, reversing their threading (S becomes Z, Z becomes S) and change your turning from odd, then even to even then odd you can then proceed weaving and the twist will undo.  This handy hint was from Shelagh Lewins.

It is worth nothing that Crowfoot drew her reconstruction with 19 cards, but her pattern was only 18 cards. I could have added the 19th card but chose to leave it to match her threading diagram.

Dyeing the yarn and weaving the final band

For my final version of the band I wanted to use linen yarn dyed with woad as opposed to commercially dyed yarn.

I had planted woad in late spring and was able to harvest just under 100 grams of leaves. 

leaves from woad plant
Woad plants ready for harvest

Woad and indigo vats typically are fermented to remove oxygen from them. This can be done with vegetable matter or urine or a chemical vat. I opted to use a chemical vat as I was running out of time to complete my project and as an apartment dweller and pet owner, leaving a vat to ferment for 10 days did not seem ideal.

Using the technique for extraction and vat found on, I tore the leaves and steeped them in warm water, then squeezed out the excess liquid and discarded my leaves.

I then added soda ash to make the liquid caustic. I should have used litmus paper to test the ph, instead I used my fingers to determine if the water was alkaline by how “slippery” my finger tips felt after dipping them.

The liquid then needs to be aerated,  I used a hand mixer to whisk the vat until blue foam appeared on the top. 

Adding dye remover and gently heating  the woad liquor causes the oxygen to deplete from the vat, turning the colour of the vat to a lighter yellow green and causing a bronze sheen on the surface. 

My vat was still a darker colour  but had the bronze sheen. I decide to proceed with dipping my linen.

I had wound off 3 skeins of unbleached  linen yarn and scoured then with a solution of soda ash and a little dish liquid and simmered it for just over an hour – the liquid in the pot turned a hazy yellow showing that any spinning oil or finish had been removed from the linen. I was now ready to dye.

I submerged all three skeins for 5 minutes, and then removed them from the vat.

When removing the yarn it appears a sickly yellow, which changes to green and then blue as the active ingredient in the dye oxidizes.

In an attempt to get two different shades of blue I did a second dip with two of the skeins for an additional 5 minutes.

I allowed all the yarn to dry for 24 hours then rinsed it with neutral soap and a little vinegar to balance the PH of the yarn.

For this final band I used the slightly darker linen for the the outer half diamonds, natural for the outlines and the slightly paler blue for the inner diamonds to match the colours described by Crowfoot.

Again it was woven on an inkle loom with wooden cards.

Conclusion and Observations

My final band would have looked better had I over dyed the darker linen with weld or otherwise had more contrast.

While the skeins I dipped twice did take on a darker colour it is slight, I suspect there was not enough pigment in my vat. The band has also faded from when I first wove the piece.

I had started with unbleached yarn, had I used bleached white yarn I would have achieved a more clear blue. As the dye took fairly well from a small amount of home grown leaves I feel this was a successful experiment in processing woad and linen. Hopefully future cultivations of woad will produce more leaves.

The structure of the band is very much as the historic example and both my bands are of a very similar width to the period finds

The structure is both strong as well as soft, and appears ideal for belts or straps as they do not tend to fold as a belt woven in a more traditional tablet weaving method might. With a pattern on both sides, it is attractive from more angles.


  1.  Crowfoot, Grace M. Textiles of the Saxon Period in the Museum of Archaeology and Ethnology, Proceedings of the Cambridge Antiquarian Society 44, pp. 30 1951.
  2. Ibid. p. 30
  3. Ibid. p. 29
  4. Collingwood, Peter. The Techniques of Tablet Weaving Echo Points Books and Media 1982  p.121
  5. Crowfoot, Grace M. , A Medieval Tablet Woven Braid form a Buckle Found at Felixstowe, Suffolk Institute, Volume XXV, Part 2, 1950 p. 203
  6. Crowfoot, 1951, p. 30
  7. Owen-Crocker, Gale R. Dress in Anglo-Saxon England. The Boydell Press, 2004. p 152
  8. Walton Rogers, Penelope. Cloth and Clothing in Early Anglo-Saxon England (AD 450-700) (CBA Research Report 145). York: CBA, 2007.
  9. Crowfoot, 1951, p. 30
  10. Crowfoot, 1951, p. 28
  11. Crowfoot, 1951, p. 30
  12.  Lewins, S. (n.d.). The Cambridge Diamonds. Shelagh’s Website.
  13. Crowfoot, 1951, p. 28
  14. Crowfoot, 1951, p. 29
  15. Crowfoot, 1950, p. 203
  16. Crowfoot, 1950, p. 202
  17. Crowfoot, 1950, p. 201
  18. Crowfoot, 1950, p. 203
  19. Crowfoot, 1950, p. 203
  20. Crowfoot, 1951, p. 29
  21. Lewins, S. (n.d.). The Cambridge Diamonds. Shelagh’s Website.
  22. Crowfoot, 1951, plate 6
  23. Crowfoot, 1951, p. 29

A gallery of other tablet weaving projects


Crowfoot, Grace M. Textiles of the Saxon Period in the Museum of Archaeology and Ethnology, Proceedings of the Cambridge Antiquarian Society 44, pp. 26-32. 1951.

Crowfoot, Grace M. , A Medieval Tablet Woven Braid form a Buckle Found at Felixstowe, Suffolk Institute, Volume XXV, Part 2, 1950 pages 201-204

Owen-Crocker, Gale R. Dress in Anglo-Saxon England. The Boydell Press, 2004.

Walton Rogers, Penelope. Cloth and Clothing in Early Anglo-Saxon England (AD 450-700) (CBA Research Report 145). York: CBA, 2007.

Collingwood Peter, The Techniques of Tablet Weaving (New York: Watson-Guptill, 1982)

Karisto Maikki, Pasanen Mervi, Applesies and Foznoses, Finnish Tabletwoven Bands, SAlakirjat, 2013

Barret, Gina, Benns, Elizabeth,Tablet Weaving; Narrow Girdles, Soper Lane, 2001

Web pages

Have loom, can’t travel, will learn; learning to weave cloth

A number of years ago I acquired a small table loom -it sat in my sewing room, unused for several years while I leaned tablet weaving and focused on making garb and doing SCA service.
Some renovations to my building caused me to clear out my sewing room and I challenged myself to learn to use the loom with the intent to weave cloth yardage for making garb.
It took a year of small steps to lean to use my loom and weave a fine wool twill, which I hand sewed into a hood.
After that project I continued to learn, buying a larger loom, joining a modern weaving guild and applying my new skills to creating more textiles that were appropriate for SCA use.

Baby Steps

My objective was to learn to create cloth with wool suitable for use in creating garb, but first I had learn about weaving and how to use the loom I had.
My first loom was a 22” wide 4 harness jack loom – this is a modern craft loom that works by raising the harnesses to create your sheds, either with treadles of leavers. Being a table loom, mine used levers.

Table loom with twill
Table loom warped with 2/2 twill in 2/8 cotton

It is similar in function to the medieval horizontal loom in that it has a reed to separate the warp, an incorporated beater and the warp is worked and tensioned horizontally but uses a slightly different method to open the shed.

The horizontal treadle loom was introduced to England in 1000 CE1  and began to displace the earlier vertical loom.2 Being a more efficient way to produce cloth its introduction lead to the creation of weaving guilds and the industrialization of weaving.3

As my interests lie mostly in the era post 1000 CE this loom seemed a good starting point.

Broken twill
Broken twill in 2/8 cotton

A jack loom is modern creation, possibly from the 1930’s4 which creates its shed in a slightly different manner from counterbalance and countermarch looms but it was the tool I had access to at the time.

Using lessons in “Learning to Weave” by Deborah Chandler, I learned basic plain weave and 2/2 twills and started expanding into broken twills using modern 2/8 weaving cotton as it is easily available, low priced and easy to work with.

To learn the mechanics of weaving, like even beat and shuttle throws I worked on a number of moderns projects with cotton and cotton/linen in a variety of twills. 

Starting with Wool

Inspired by the beautiful work by Drifa in Rauða that I had seen on facebook and her blog as well as the always inspiring Mervi Pasnen, I wanted to work with wool.
For this first attempt I selected Harrisville Shetland as it has an affordable price point and is available in a good range of colours. It is S spun rather than Z spun, which was more commonly used in the medieval period.5  It is also quite a bit heavier than examples located in London or Greenland being “fingerling” weigh with a manufactures recommended set of 12-15 EPI for twill while historical examples from Greenland are typically finer at 7-10 threads per cm6 or 18-25 EPI (ends per inch).

I also chose to use contrasting coloured threads as I liked the aesthetic. 

Using the weaving draft from Drifa’s web site7  I warped my table loom at 12 EPI with a black warp and I used evergreen for the weft.
Due to my yarn choice the result was a large bold broken diamond. While likely not historically accurate in size or contrasting colours I was fairy happy with the end result and used this fabric to handsew a hood in the style of the  Skjoldehamm Hood.

Working with finer wool and recreating a 2/2 twill inspired by one found in London

My laurel had gifted me with 2 large spools of undyed 28/2 wool. With this finer yarn I wanted to try and create yardage similar to textiles found in London.
The wool is a bit old and “snappy” so rather than weaving it in its current state I thought dyeing it to bloom fibre may make it easier to work with.
Objects Nos 38 and 159 from the Baynards Castle excavation in London were woven in 2/2 twill in contrasting tints of madder red and white.8  No thread density is reported in my source book but the cloth is described as “good quality” and “of similar quality” to object No 392 located in the same excavation site9. No 392 is woven of fine yarn with a warp of 17 threads per cm.10  (aprox 43 EPI).

I experimented with the 28/2 yarn in plain weave and twill. Threading at 32 EPI produced an even twill but the plain weave was too tight a set.

I had dyed some of the yarn with madder and found that the processing did seem to increase the tensile strength of the threads so it was used as the warp and the undyed as the weft in my sampling.

After this sample warp I was ready to proceed with creating my yardage 

Using the 28/2 yarn I dyed with madder (see appendix) as the warp I threaded my 4 harness loom at 32 EPI (4 threads per dent in my 8 EPI reed) in a “straight draw” of 12341234.

I then wove my twill trying to maintain an even beat and twill line at 45 degrees to create an even 2:2 twill.

The best technique was to throw the weft, change the shed and then beat on the open shed.

I could see some vertical striping from some variations in the dye on the thread, but over all it looked very nice.

After weaving the wool was soaked in hot tap water with a little orvus paste to loosen any residual dye. I had played wth overdying my sample and I was hoping this might also “mute” the natural undyed weft to a more pleasing colour. This released more dye than I had anticipated and over dyed the natural weft a light orange.

While this was more over dye than I had expected the finished result is a lovely soft wool in a pleasing autumn tone. The contrasting twill line is minimized as are some of the vertical colour issues caused by some variation in the dyed wool.

muted red wool twill
my twill cloth after wet finish to remove excess dye and over dye the warp

The end result was 4 meters of cloth, which has a fine hand and good drape. I used this fabric to handsew a button hood (see appendix).

It took just over a year from my first experiments to completing my hood. While there were some flaws in the cloth from snapped warps I was over all satisfied with the cloth density and drape as well as the colour.

Moving on to Counterbalance loom

After my year long odyssey of teaching myself to weave on the table loom I decided to upgrade to a counter balance floor loom. This new loom is wider, uses treadles rather than leavers and opens sheds by lifting one harness while lowering the other – this provides a more even tension on the warp and is more similar to the action on a medieval treadle loom.

I also enrolled in an modern online weaving guild which has improved my technique and overall skill with this loom.
I have woven quite a few lengths of broken diamond twill for norse “apron fronts” as well as plain weave wool wraps using “colour and weave” to create patterning.

Another recent project was 5 m of 24 EPI plain weave from 28/2 wool, this wool will eventually be overdyed and handsewn into a Roman Stola.

Moving forward Id like to weave more with finer wools and create more yardage for garments.

Gallery of Weaving Projects


  1. Pepper, Gwendoline, Time looms over us: Observations from an experimental comparison of medieval English loom-types, Archaeological Textiles Review, No 61, (2019) pg 74
  2. Elisabeth Crowfoot, et all  Textiles and Clothing, c.1150-1450: Finds from Medieval Excavations in London, Museum of London, Boydell Press 2001 pg 22
  3. Pepper, 2019, p 74
  4. Hoogt, M. van der. (2016, January 5). Ask Madelyn: Jack Looms. Handwoven.
  5. Elisabeth Crowfoot, et all, 2001 pg 36
  6. Lilli Fransen, Anna Nargaard and Else Ostergard, Medieval Garments Reconstructed, Norse Clothing Patterns, Aarhus University Press, 2011, pg 36
  8. Elisabeth Crowfoot, et all, 2001 pg 37
  9. Elisabeth Crowfoot, et all, 2001 pg 37
  10. Elisabeth Crowfoot, et all, 2001 pg 38

Appendix 1

Dying the wool and sewing the hood, information on dyeing the yarn for my wool yardage and constructing a buttoned hood 

Preparing and dyeing the yarn

For a warp 21” wide x 4.2 meters long at 32 EPI I would need just over 2800 meters of warp.  I wound off 9 skeins of 400 m and one of 200 meters.  For a total of 3800 meters of yarn to dye.

measuring off wool
Skeins of wool before scouring, mordant and dye

This was 325 grams of wool. I knew from past experience that to get a nice red the best ratio was about 80-100% weight of fibre (WoF) for madder. So 325 grams of wool would require 260-325 grams of madder.

To dye with madder, first you need to scour your wool and treat it with a mordant to allow better absorption of the dye stuff. 1

The wool was scoured with a neural soap (orvus paste) by placing the wool and soap in a large pot of warm water and holding it at a “gentle heat” of about 60 degrees C. I do not use a thermometer, instead relying on measuring the water temp by dropping it on the wrist or other ways of “reading” the pot.

After scouring, the wool was treated with a mordant. I used alum and a little cream of tartar to bring out a nice red. The chemicals are dissolved in hot water and the wool was held at just under a simmer for about an hour, then allowed to cool. It was then allowed to partially dry and then wrapped in damp linen for a few days.

After mordanting it was time to dye- I measured out 250 grams of powdered madder root add placed it in cheese cloth to prevent the ground roots from sticking to the yarn. 

I soaked this bag of madder in a warm pot and added some chalk as hard water creates more vibrant reds and the water here is quite soft.

I then added half the yarn to the pot and held the water around 60 degrees for 2 hours.  You can gauge the water temperature by dripping on the wrist or feeling the heat coming off the pot, if small bubbles form the water is over 70 degrees and this will effect the colour..

After two hours I removed the first batch of yarn, added a little more madder and placed the second set of skeins into the pot.

After drying the yarn appeared fairly consistent in colour.

The pattern and construction of the hood

I drafted the pattern based on the size of both London Hood 2463 and the Greenland hood D106064 including the top seam and “horn”, adjusting the pattern from Medieval Garments Reconstructed, Norse Clothing Patterns5 to fit my somewhat small skull and generous hair.

The final cloth is very soft and  light so needs to be lined for support. I used a fine creamy natural linen for the lining.

Considering the method of lining, I needed to support the wool but also needed to ensure the hand woven fabric was not going to fray. I wanted the front opening to be nicely faced to allow for more support for the buttons and  holes. So I opted to line the entire hood minus the liripipe, sewing the hood together and turning it back on itself.

Seaming was done by hand with the same yarn used to weave the wool. This thread is not ideal for sewing, linen thread may have been a better choice but linen sewing thread is somewhat difficult to source.

 A running stitch was used along the front opening as the strength was not required, stab stitches were used to “set” the lining in place when turned.

For the top of the hood I used an overlap with small stitches and overcasting on the raw edge  – this way there is no “open” seam on the top of my head.

I hand sewed button holes equally spaced along the front opening to match the size of the hand cast pewter buttons I had been gifted many years ago. These buttons are consistent with cast lead/tin buttons dated to the later part of the 14th century. 6

Conclusions and observations 

If nothing else this project reinforced the medieval mindset of waste nothing, this was a fairly labour intensive process -from dying to weaving to the relative ease of sewing.

The yarn bundles clearly had some overly tight choke ties as there were some light areas near the ties. It also might have been a better idea to split it into three batches rather than two as I think the dye may have been more even if there was more room in the dye pot. 

Due to some irregularities in the dye bath you can see vertical stripping in the finished cloth. There is also some horizontal stripes due to some irregular beats and from the angle of the beater hitting the fell. My loom is a fairly small table loom, with a larger one this might be less of an issue.

Soaking the cloth after weaving helped to lessen some of this as the release colour overdyed the fabric. I may experiment with this in the future with piece dying fabric that is woven with yarn

The “horn” on the hood does keep the hood more open and creates a visor like effect. This is similar to what I had found when making a skjoldehamn style hood with the “cockscomb”, a step often left out that greatly improves the function of the hood.


  1. Maiwa Guide to Natural Dyes p 7
  2. Maiwa Guide to Natural Dyes p 9
  3. Elisabeth Crowfoot, et all  Textiles and Clothing, c.1150-1450: Finds from Medieval Excavations in London, Museum of London, Boydell Press 2001 pg 190
  4. Else Ostergard, Woven into the Earth, Textiles form Norse Greenland, Aarhus University Press, 2004 p 214
  5. Lilli Fransen, Anna Nargaard and Else Ostergard, Medieval Garments Reconstructed, Norse Clothing Patterns, Aarhus University Press, 2011, pg 123
  6. Geoff Egan and Frances Pritchard , Dress Accessories 1150-1450. Medieval Finds from Excavations In London, The Boydell Press, 1991 pg 275


Pepper, Gwendoline, Time looms over us: Observations from an experimental comparison of medieval English loom-types, Archaeological Textiles Review, No 61, (2019) pg 71-87

Elisabeth Crowfoot, Frances Pritchard & Kay Staniland, Textiles and Clothing, c.1150-1450: Finds from Medieval Excavations in London, Museum of London, Boydell Press 2001

Lilli Fransen, Anna Nargaard and Else Ostergard, Medieval Garments Reconstructed, Norse Clothing Patterns, Aarhus University Press, 2011

Else Ostergard, Woven into the Earth, Textiles form Norse Greenland, Aarhus University Press, 2004 

Geoff Egan and Frances Pritchard , Dress Accessories 1150-1450. Medieval Finds from Excavations In London, The Boydell Press, 1991

Hoogt, M. van der. (2016, January 5). Ask Madelyn: Jack Looms. Handwoven. 

The Maiwa Guide to Natural Dyes. © Maiwa Handprints Ltd. 2020.

Wren Digital Library – O.9.34, Romance of Alexander, f032v

Websites for further reading

One Reply to “My tangled web, weaving cloth and trim”

  1. This is an amazing confluence of history, science, and art. Your observations about how labor-intensive all aspects of a given project were is thought-provoking in the era of fast fashion and buy, not make. Thank you for sharing the breadth and depth of your weaving projects with us!

Leave a Reply

Your email address will not be published. Required fields are marked *

3 × 1 =

Request a One-on-One Conference

Please login or register to book a one-on-one appointment with this Exhibitor!