To make fabric like this.
I can’t look at it without smiling.
The seeds of this project were sown at SOAR, the Spin-Off Autumn Retreat at Tahoe in October 2012 when I attended Stephenie Gaustad’s ‘Spinning for a replicate or reenactment textile’ aka the Class of Awesome! It truly was a Class of Awesome: I left with far more knowledge and confidence about spinning for weaving as well as a class handout containing descriptions of several historic textiles. ‘Donbæk Check’ – named for the burial site where it was found, near Frederikshavn in Denmark – sounded fascinating. Dating from the Iron Age, the 2/2 twill fabric had a chequerboard pattern produced by the light falling differently on the intersections of singles spun clockwise and counter-clockwise in the warp and weft. I’d heard of/seen pictures of twist-patterning in tablet-woven bands but hadn’t thought about it on fabric. Back in the UK I made a special expedition to the University Library to photocopy the reference and vowed I’d try this one day.
Fast forward to January 2017, when I was preparing to teach my first ‘Spinning for weaving’ workshop. While I enjoy making *any* fabric from my handspun, I wanted to show the students something memorable, a fabric that would be impossible or at least very difficult to weave from commercial yarns… and I remembered the Donbæk Check. Perfect. I had 2 x 100g packs of charcoal grey Wensleydale combed top from Julia Desch/Diamond Fibres: this is not pin-drafted top from giant drum carders, this is real (commercial) combed top. It’s special. It’s lovely. I spun each 100g in a different direction using a short forward (‘worsted’) draw for a dense and shiny singles yarn that would – I hoped – reflect the light effectively.
The fibre: a slightly darker version of the Wensleydale top, together with a washed lock from a Wensleydale fleece. Shiny!
The singles, before and after steaming to set the twist.
I wanted to protect the smooth surface of that yarn and wasn’t sure how much I’d use for the warp, so I used a gelatine size on all of it.
The sized singles yarns drying on the clothes tree. The plastic bin is weighted with just enough water to hold the singles taut, removing the pigtails, but with minimal stretching. Remember that wet wool is weaker than dry wool.
I wound a warp from just over half the yarn by weight, alternating clockwise and counterclockwise yarns every 26 ends to yield a rough 1″ check. Sett at 27epi for twill calculated from wpi it was seriously sleazy. I unwove the first 2″ and re-sleyed at 36epi, an eyeballed guess. This produced what I think is a lovely looking fabric that wove with a delightfully clear shed; although I hadn’t spun the yarn particularly tightly, the fact that it’s a longwool meant there were fewer ends, well locked down, so relatively little fuzz.
I was worried at this point because although I’d dutifully alternated the clockwise and counterclockwise singles in the weft (carrying the unused yarn up the selvedge), I could not see a pattern. I checked the underside… no pattern. Well. I needed to know how the cloth would behave after finishing (simple washing, in this case), and as the class had been moved forward I needed an example of it as soon as possible, so I cut that strip off as a sample, roughly stitched it to bind the ends and create a tiny ‘loom state’ sample, then washed the slightly larger piece.
The size was obscuring the pattern on the loom state fabric. It’s even more subtle than I expected: if the fabric is flat so the light does not play on the surface the pattern is scarcely visible.
I didn’t have time to finish the sample warp before the class, so I took the table loom and that small sample to show them, which was probably better than the finished sample alone, as students who hadn’t woven were able to try their hand.
Then I raced home to weave it off the loom, stitched the ends and threw it into the wool wash cycle of the washing machine. I ironed the damp fabric on the counter top (not the ironing board, which is too soft), using all my weight and the ‘linen’ setting to flatten it: the chequerboard formed hills and hollows in the unironed fabric.
Unsurprisingly the weft stripe is more obvious on one side, the warp stripe on the other, but the chequerboard is visible on either side if the light is right.
It’s even more visible if light passes through the fabric. Isn’t that interesting? I’ll just go and find a hand-lens so I can mark the twist directions of the singles in the sample…
In The Big Book of Handspinning Alden Amos wrote “If warp and weft have the same twist direction, the threads will bed together better during the weaving … If the warp and weft have opposing twists, the individual yarns will be plainer or clearer in outline, the cloth will not be quite as dense.”
First I must consider another factor: the amount of twist I put into the singles when spinning. I almost always spin singles clockwise or ‘S’ – it’s the modern tradition – and this means my hands don’t get as much practice drafting while spinning counter-clockwise. Spinning counter-clockwise felt awkward, and I had to adjust my whorl and treadling to accommodate the slower drafting. When I look at the singles at 20x magnification, I can see that the counter-clockwise singles have a shallower twist angle – less twist – than the clockwise singles.
In that photo the clockwise (S) singles certainly do exactly as Alden suggests: they bed together tightly, forming the densest squares (S/S). Looking more closely at the fabric than I can photograph – you’ll have to take my word for it – I see that the counter-clockwise singles also bed together tightly, but only in small areas of the Z/Z squares: the variation in twist is interfering, and this is why on average the squares look less dense. In the S/Z or Z/S squares, the yarns do seem to form a more regular grid, as though the warp prevents the weft yarns from cuddling together and vice versa. It occurs to me that this sort of interaction would have caused the hills and hollows in the fabric when it came out of the washing machine – and that my hot iron and heavy pressing may have altered the ‘natural’ interactions of the singles.
All food for thought. Further experimentation is required!