Miscalculations

Or, “Why you should always do a full-width sample”:

I finally got to weave this morning after B. got up (I don’t weave while he’s asleep because of the noise).  After spending about half an hour weaving a header and troubleshooting various bits and pieces, I started to weave in earnest.  After about three inches, I looked at what I’d woven, and thought to myself, “Hmm, that doesn’t seem right.”

So I took out the sample from the gamp I’d woven previously, and looked at it.  Sure enough, the full-width version was elongated, by about 20%.  (Compare the bottom right with the original sample, on the left-hand-side.)  This made an enormous difference in the finished look, not least because it made the motifs spaced slightly closer crosswise than lengthwise, transforming the slight vertical (flattering) line of the sample to a slight horizontal (“makes-you-look-fat”) line.

I then tried thinning out the weft thread to shorten up the motifs.  I had been using two strands of metallic gold thread as weft, so I cut it to one strand.  That worked (top right), but the motifs didn’t stand out nearly as well, so I don’t like that much either.

Which leaves me two options: shorten the treadling sequence between motifs to restore the vertical line, or re-sley at a wider sett to give the yarn more room to pack in.  I’m going to start by shortening the treadling sequence, as that requires the least work – I can do that on the computer, but to re-sley I’d have to pull out a different reed and spend another 3-4 hours re-sleying.

I’m guessing I will wind up re-sleying, though, as I don’t especially like the look of the elongated motif.

The lesson here is ALWAYS to do a full-width sample, especially if you are weaving at the full width of your loom.  My sample/gamp was 17″ across, which I thought would be wide enough, but when I increased the width by roughly 30%, the picks per inch dropped by 20%, changing the appearance pretty drastically.  I think that to bring it back into balance I should probably widen the sett by 10-15%, meaning 50-54 ends per inch.  Testing that theory will take another 3-4 hours of re-sleying, however.

I don’t actually regret not having done a full width sample first – why go through 15 hours of beaming on, threading, etc. when you can just use the first half-yard for samples? – but it would have made things somewhat more predictable.

More tonight – I gotta get to work!!!