Showing posts with label experiments. Show all posts
Showing posts with label experiments. Show all posts

Tuesday, 15 March 2016

Turning lace patterns upside down

This post isn't news to the knitting world, but since there seems to be quite a bit of confusion about it on forums when you google "knit lace pattern upside down" or somesuch, I thought it was worth a post.

Lets say (like me) you want to knit a seamless cardigan from the top down, but you want to include an all-over lace pattern.  Most lace charts are written bottom-up, at least in the stitch-dictionary on my shelf.  Maybe instead you are knitting a lace scarf and don't want to knit two shorter ones with a graft in the middle.  What to do?

Some A lot of posts on forums breezily suggest "just turn the chart upside down and replace decreases by increases!"  Don't listen to them, my friends.  While this may work for some lace patterns, it won't work for all.  And it may well look different from the lace knitted right-way-up, so I wouldn't try to match the two.

It is true you can expend a lot of brainpower and swatching time to devise an upside-down chart that retains most of the design features of the original pattern, but it might well not look the same in some key features.  There's a reason to graft your scarf in the middle!

Here comes an example:



This cute pattern of three holes is called "little flowers" in my stitch dictionary, or perhaps it's really little hearts.  Either way, it's cuter this way up than inverted ("little stack of bowling balls"?)  The chart is thusly:



Knitted bottom-up, you make the middle yarn-over (yo) on row 1, then the two yo on the next knit row 3.

Now turn the chart upside down.  You will make the two yo as your first pattern row - you can either make the decreases outside them, or switch to one centred double decrease in the middle.  By the issue comes when it's time to stack that last yo symmetrically on top of and between the lower pair.  There's a single stitch where we want our yo to go.  Where to put it?


The most obvious is to k2tog before the yo, or else yo then ssk.  Both these options lead to a "fat" bit between two of the holes, one side or the other, depending on which side you "got rid" of that central stitch to.  For clearer diagrams of this effect and how you can use it to make deliberate lace patterns, see here.

Here's my swatch:


Looking at the swatch above, the lowest motif is the original lace pattern worked from bottom up.  The next one up is the chart knitted upside down with a centred double decrease between the two yarn-overs.  Note that the pattern is no longer symmetrical - there's a fat bit between the bottom right hole and the central hole on top, due to the ssk before the last yo on row 3.

How do we get rid of the fat bit?  I tried two options.  The top motif in my swatch is the following chart:


I make one extra stitch on the purl row 2 by purling into the front and back of the central stitch, then I get rid of it again by doing symmetrical decreases around my top yo on row 3.  Symmetry is restored, but the pattern is less delicate - instead of a little yarn twist between the holes, there are fat bits on both sides where stitches have been knitted together leaning in both directions.

The second motif down is a bizarre non-standard variation on this theme, where the top hole is created by making the central stitch on the purl row very loose*, then on row 3 knitting into both sides and the top to pull it open.  I don't recommend it, and it looks much the same as my other symmetrical option which is easier to describe and execute.

* Insert needle into stitch to be knitted, yarn round needle twice, pull through stitch.  Like making a yo part way through the stitch.

The essential difference is that stitches are shifted by 1/2 stitch knitting top-down relative to bottom-up.  Think about the shift in the peaks vs troughs in a wave, or the number of fingers on your hand vs. gaps between them.  So, there's no easy way I can think of to exactly replicate the effect of knitting bottom-up while knitting top down.

In summary, I will be knitting my lace cardigan bottom-up with the original chart.

Tuesday, 14 April 2015

T-shirt decoration technique testing

I'm struggling to find cute reasonably priced t-shirts for toddler, as I'm not a fan of slogans, skulls or Disney/Pixar merchandise.  Yes, I'm fussy, but toddler likes bright t-shirts with animals, dinosaurs and monsters on best.  He's also a big fan of tools, Henry Hoovers, Pocoyo and Postman Pat, but there's a shortage of t-shirts in these interest areas too.

I thought I'd have a go at decorating plain t-shirts for toddler.  I made myself an inspiration board in the dead of night.  I found an old stained t-shirt in the rag bag and I've tested out various bleaches, appliqué and fabric paint.  It's going through every wash in my machine now to test durability.

Bleach


In my cupboard, I could find Domestos Toilet Gel, and Mold and Mildew remover (Dettol, I think).  I tried a freezer paper stencil and sponged the bleach over it, before removing the stencil and washing the shirt in cold water.

My bottle of Domestos is lime scented and green.  The green gel stained the bleached bit of the t-shirt green-ish, so negative points for that.  That said, the gel nature did mean that the bleach didn't leach outside the stencil in the 5+ minutes I left it to do it's business.  The bleach isn't really strong enough for this purpose, and the result is a design that's pretty faint and hard to see.

I know the Mold and Mildew remover can bleach clothes, because my husband accidentally wrecked his new brown dressing gown with it while cleaning the bathroom.  However, it turns out it's not strong enough to do stencilled designs.  You have to leave it on the shirt for a long time before it bleaches the fabric and during this time it leaches under the stencil.  You get a very visible bleached patch in the shape of a generic blob.

Verdict:  I need something stronger than 3.5% sodium hypochlorite, and I have nothing in my cupboards that nasty.  I'm not going out to buy some specially, so it's a no for the bleach.

Applique


My appliqué cotton patch is holding up well on the test shirt, and since the t-shirt fabric doesn't have that much stretch then it doesn't seem to matter that the patch is made of a woven fabric.  I used double-sided fusible to apply it, then zig-zag stitched around the edges.  I think this is a winner, but I might be wary about using a woven for large area designs in case it pulls the fabric out of shape.  Also, pre-wash will be key!

Fabric paint


Alas, I have to actually pay for this version as I have nothing in the cupboards.  I forked out £1.80 for a fruit-of-the-loom t-shirt, and £2 for a tiny pot of DecoArt SoSoft fabric paint.  I used my friend the freezer-paper stencil. and copied a design I found on pinterest (sorry original artist!).

The t-shirt cut is a bit boxy for my liking, but the quality is good and they have a lovely range of bright colours.

The paint on my sample shirt seems to be standing up to washing reasonably well, and it's certainly gone nice and soft.  It feels just like the screen printed bits on commercial shirts.  It's a win!  I've only washed the actual t-shirt once after painting and it's not washed as well as my test sample, but we'll see how it goes with wearing.