(in case you’re wondering: sewing curtains on the other machines is fine, but it feels like I’m pushing it too hard when I sew at maximum speed for a certain length of time. The Pfaff 238 must have spoiled me, as it was such a beast. The 262 isn’t as strong, but still has a 1.12 amp motor which will fare better with curtains and denim than the Janome’s 0.7 amp motor.)

Anyhow, try as I might, the 262 does not want to be reasoned with – it only wants to do a zigzag. Since opening her up, I’ve joined a couple of groups of vintage sewing machine enthusiasts and created the video above so that they could weigh in as to what the problem might be (bigger video can be found here).

I apologize to those of you who have no interest in sewing and/or vintage machines – I’ll be back on track soon. In any case, all my knitting is secret at the moment so I wouldn’t have anything to tell you about even if I wasn’t so enthralled with gears.

This is what’s left – let me begin with the oldest…

The Singer 221 (also known as a Singer Featherweight) looks older than it really is. It was given to me by my father in law and used to belong to his mother. Judging from the serial number, she must have purchased it either in the late 50s or the early 60s from ‘Chalifour’ on St-Hubert in Montreal:

It can only do a straight stitch on its own (a zigzagger attachment was available for it), but it is famous for the beauty of its stitch and is prized among quilters. I have fabric on hand to make Marcel a couple of shirts – with this machine and my copy of ‘Shirtmaking’ by David Page Coffin, I’ll be all set. Marcel will love wearing shirts made on his grandmother’s machine.

My grandmother’s sewing machine, a Pfaff 262, is only a little older – I believe she purchased it in the mid 60s. She was a dressmaker and used hers quite a bit more than Nanny did hers, as you can see! Many machines of this era utilized removable stitch cams in order to work decorative stitches – the attachment I mentioned earlier for the featherweight used these to work stitches beyond the zigzag stitch. The Pfaff was different in that its cams were built in – one refers to a ‘stitch wheel’ in order to access these stitches.

It’s classified as a ‘light industrial’ and weighs a ton. Unfortunately, it needs to be serviced – I think the timing is off. When I get a minute, I’ll attempt to clean and adjust it.

The serger was another gift and dates back to 1996. It’s a basic machine but can accomplish all I need it to do: it has differential feed, an adjustable cutting width, either 3 or 4 overlock stitch and can work a rolled hem. Parts are still available and it’s easy to work with , so what’s not to love?

A couple years later in 1998, after much saving, I acquired my first ‘fancy’ machine – a Janome 4800. I love this machine - buttonholes used to take forever before I had it. I worked them manually using a zigzag stitch and a lot of marking was necessary. Fine when sewing something precious, but irritating when sewing pajama tops. No longer – the Janome works identical buttonholes one after the other.

As you’d expect from such a machine, it boasts a fair number of built in stitches, but not too many crazy ones. I mostly like the fact that its straight stitch and buttonholes are very nice. The keyhole buttonhole and eyelet are nice extras, and it’s fun to monogram the yoke of a shirt with the alphabet. The stitches can be combined as well as lengthened without affecting the stitch density. The adjustable speed is useful to new sewers – Oona likes that she can slow it down so that it doesn’t take off on her. I love the needle up/down feature.

The newest member of the family is the coverstitch machine – this is the type of machine used to hem tee shirts. It sews a double or triple line of top stitching on one side and an overlock on the reverse side. A similar look can be achieved with a twin needle on a sewing machine but tunneling is often a problem (tunneling refers to a raised effect between the two rows of stitching) and the reverse side isn’t very attractive (IMO). Before purchasing it, I also considered two Janome models – one was similarly priced but could only sew a double line of top stitching which meant being limited to a 6mm wide cover stitch while the brother can sew either a 3mm wide or 6mm wide coverstitch. The more expensive model from Janome had this option as well as a free arm, but I couldn’t justify the extra expense.

Some sergers can work a coverstitch as well. From what I’ve heard, the serger to coverstitch conversion takes a while. Since I’m sentimental about my serger and I have the room for an extra machine, it disn’t make sense to replace it. But the option is available.

I’ll now ask you for help: my friend already has a sewing machine but is thinking about a serger as he would like to work with jerseys. Do you have any recommendations? Do you own a combination serger/coverstitch? And do you know of other books besides Mr Coffin’s you’d recommend to a guy? Somehow, I don’t imagine he’ll want to hear all about using a serger to produce lettuce edges…

That little top I spoke about the other day is done and I’m pretty happy with it. I’d read quite a bit about the sewing instructions on sewing blogs and pattern review, which led to to research the RTW version online. As far as I can tell, the original was constructed using a coverstitch machine and the vogue instructions attempted to achieve the same effect using twin needle stitching using a regular sewing machine. Since I happen to have a coverstitch machine, I didn’t use a twin needle for this project (never had great results with it on jersey, anyway).

I also opted against the selvedges on the outside, which I think my jersey is too heavy for. Instead, I serged the raglan seams in addition to a handful of little squares. I then tested various needle/presser foot configurations on the coverstitch on those little squares until I found a pleasing combination: 3mm wide coverstitch stitched from the wrong side using the wider topstitching foot for the brother 2340cv. Using the two left needles allowed the serged seam to center well enough so the fabric was a breeze to guide through. The back received the same treatment – both the RTW and Vogue versions call for satin ribbon on either side of a minute gap between the left and right back pieces, but I didn’t have satin ribbon on hand and, again, I think my jersey is too heavy for this kind of treatment.

I completely undid the neckband – the instructions called for it to be topstitched with the raw edges apparent but I didn’t like the look – first, I should have sewn it with the wrong/purl side out. The edges rolled in and I feel it would be more successful with the edges rolling out. It could be my fault, though – I didn’t cut the band on the true bias. I ended up simply serging the neckband and neckline together right at the cut edge – can’t imagine I took off more than 1/16″ or so. Doing so raised the neckline a tad.

The bottom hem was simply turned up 3/4″, pressed and topstitched – I didn’t trim the miter, so multiple layers are present at each point but I don’t think it’ll be a problem. The sleeve hems received the same treatment but were only turned up 1/2″.

Fit wise, it’s pretty good (sorry, too camera shy to model). Being a little shorter than 5’4″, I removed 2″ from the body (1″ at each mark) and 1″ from the sleeve. The cap and underarm section seems a little long on me, but only by a little. I’ll definitely make it again with a very lightweight black jersey.

Looks kind of sad, doesn’t it? Let me retrace a bit.

Last weekend was filled with minor annoyances – a piece of software saw fit to warn me that my hard drive had ‘integrity errors’ (?), which led me down the rabbit hole of backing up, scanning, verifying, etc… After disk utility failed to find any problem’s with my drive’s integrity, I tried booting from a dvd in order to scan the drive further. No go. All the usual tricks failed and the mac restarted each and every time with its happy little chime.

It wasn’t the first thing that had gone wrong, so I decided some instant gratification was needed. I had a pattern for a cute top and just the right fabric, so set out to carefully cut out the pieces. I figured it would take no more than a couple of hours to serge it together (when I told my friend Ali P. the story, she said ‘uh oh’ at this point).

Uh oh was right. Good old Pfaff 776 was acting funny. It had done that lately, mainly seeming to change tension all by itself *midseam*. Further inspection revealed that the needle bar was loose, but I had no idea how to tighten it.

Went online in search of a service manual – found some (my serger is old – Marcel bought it for me when I was pregnant with Oona), but the first few were either photocopies which had to had to be physically mailed. Patience was a foreign word for me at this point so I kept looking until I finally found a pdf and bought it.

I then waited for 15 minutes, hitting ‘get mail’ countless times. Nothing. Fed up, I gave up and decided to do something I couldn’t possibly fail at: cook something.

The service manual arrived nearly 4 hours later. With its help, the machine was quickly dismantled and cleaned – you should have seen the amount of lint trapped within its housing (I though it was clean). Figured out how to adjust and tighten the needle bar, checked the other settings and put the whole thing back together.

The good news is, it now sews very nicely. Bad news? Adobe Acrobat decided to give me a hard time today and make everything fuzzy. Keep me in your thoughts.

Once I’ve drawn a sketch, I can begin drafting the shape – I use Garment Designer (GD), from Cochenille Studio. The general requirements are selected through menus at the top, and each garment shape can be further refined through toggle points – I highlighted the sleeve’s side seams to show these above.

Sometimes it is necessary to use menus in a roundabout manner – in the sleeve above, you can see that the only toggle points available are situated at the wrist and bicep line. I  would like my sleeve to be shaped differently but as I cannot add points, I’ll select a different silhouette:

There are still only two sets of points, but these now allow for modifiable curves.

Once the sleeve shape was modified to my taste, I made a few more changes – changed the neck group to ‘shawl’, lengthened the body and straightened the hem line. It should be noted that GD can print out full size patterns, which is why the hem was curved in the first place. If this was a sewing pattern, I would have wanted to retain the curved hem.

Now that the schematic draft is complete, it needs to be exported to Stitch Painter for further edits. It isn’t necessary to own Stitch Painter to use GD, but I like to make full body charts. To do so, I need to input the gauge of my knitting into GD. Because there are several steps involved, I recorded the procedure.

(I realize that this is a pretty small file, so please follow this link if you’d like to see a a larger version)

This process is repeated for every shape, and the resulting files are combined in Stitch Painter. There’s more to do here, but I’ll leave the rest for another day.

I’ve been inspired by customers – ever since I started to carry B&L yarns, my most enthusiastic supporters have been from Japan. The most popular varieties are Super and Country Roving, and I wish I could read japanese so that I could follow their blogs and find out what they are making. But it’s a pretty safe bet that they making cowichan style sweaters.

Oona loves everything japanese, so jumped at the chance when I asked her  if she’d like a cowichan style jacket. Funny story: in one of my favorite plays, there is quite an argument between the two characters as to whether a garment is a sweater or a jacket. When I designed the costumes for it so many years ago, those lines brought to mind my father’s old zipped cardigan – my grandmother had knit it using a Mary Maxim pattern and lined it throughout.

Years later, Marcel was cast in the role of Ernest and the costume designer found a pictorial Mary Maxim for him on Ebay. Maybe it’s because I’m now a knitter (I wasn’t for the first go-round of Ernest and Ernestine), but I found the costume choice hilarious.

Back to Oona’s SweaterJacket – real Cowichan sweaters are  knit using thick single-ply handspun yarn (it should also be noted that only authentic garments made by the Cowichan band on Canada’s west coast have a right to be designated as ‘Cowichan Sweaters’. As such, no hand knitting yarn has been commercially produced which can be said to be authentic). Over the years, several companies have offered yarn specifically for this type of knitting – White Buffalo and Briggs & Little among them. Since White Buffalo yarn has been discontinued, many now opt for country roving but, sadly, it isn’t available in a clear, bright yellow – and Oona is all about yellow. So, I’ll be using Super.

The yarns I chose can be seen above and here is my sketch. Next, I’ll need to swatch and draft my pattern pieces.

Here’s one of the projects I haven’t gotten to writing about yet – a new dress for Oona. She needed something to wear for a school dance and we both found the RTW options kind of dreary – too many ruffles and mini skirts. And not a pocket in sight.

So, we dug through the pattern stash and came up with Vogue 1236. Perfect – the pleats at the neckline enliven an otherwise super simple shape and it has side seam pockets (I think that all skirts and dresses should have pockets. Oona agrees). None of the fabric on hand suited princess, but we soon found a nice cotton from Fabric Mart for all of $4 a yard. I ended up buying 2.5 yards and there’s plenty left for a skirt.

I often come across someone who tells me they wish they could sew – if you do and you’re serious, this dress is a great project for you. The only alteration to the pattern was to raise the neckline about 2cm as per Oona’s request. This did shorten the dress overall (I hadn’t bothered with a muslim), but she’s 15 – she ought to show a least a little leg, no? As it stands, the hem is a couple of inches above her knee caps.

More projects coming up!

I finished this cardigan back in December, but ended up wearing it so much around the house that it needed to be blocked once more for photographs. Here are the details at last.

Pattern: Corcovado by Carol Feller

Yarn: Briggs & Little Softspun (black)/Briggs & Little Regal (Dark Grey/Fir Green)

Modifications: Altered length for height and so as to add 2×2 ribbing on body and sleeves. Changed back to Softspun and 2×2 rib instead of working back neck extension. Also worked additional rows along centre front to fasten cardigan closed with a shawl pin (which I couldn’t find… used a stitch holder instead).

I’m really running behind on updating this weblog with FO details. More to come.

I’ve also posted a close up on my flickr stream. Should I put the kid to work so that we have a knitting ninja calendar in 2013?

But, back to the subject at hand…
It’s been a while since I promised to write about grading, but it’s a lengthy subject and requires an uninterrupted block of time.

Before we can start about grading, let’s take a quick look at patterns and where they come from. When I speak about patterns, I don’t mean stitch patterns but rather the garment pieces garments are comprised of – what you see when you look at schematics. These are drafted using one of two methods: draping or the flat pattern method. Draping is done by pinning and positioning fabric on a dress form to develop the desired garment. I usually drape an a half size dress form, as can be seen in the photograph at the top of this post.

Flat patternmaking begins with what is known as a block or sloper – these are shells which fit the body with only wearing ease added. Sewing enthusiasts can purchase a pattern for one from one the four big pattern companies; in theory, any changes that must be made to it for fit would also be made to any pattern from the same company whether it is for length, width, etc. An example of a fitting shell can be viewed here.

Once a block has been perfected, it can be used again and again to create new styles by drawing new style lines. One could use the bodice alone to create the pattern for a vest –  the neckline could be redrawn to a deep V, with length added below the waist and more wearing ease  added throughout if needed. This is a very basic example, but I think you get the idea.

Once the pattern pieces have been perfected for the sample size, grading can begin. Let’s take a look at a front bodice to begin:

As you can see, the piece above represents only half of the front bodice. There’s no need to show both pieces as they are identical mirror images.

Succinctly, grading (or sizing) refers to extending the range of sizes garments are available in. The patternmaker takes the pattern drafted to fit the sample size and makes the adjustments needed so that it will fit smaller and larger sizes. It isn’t simply a question of blowing up the pattern a certain percentage – doing so would result in parts of the garment being ill-fitting as the shoulders and neck do not follow the same growth curve as the hip and bust do. The adjustments must been evenly distributed in a way similar to this:

The particular pattern piece shows what the grading would be for a difference of 1 1/2″ – the overall size grade can be seen by adding the three numbers at the very bottom and quadrupling them, since we can only see a quarter of the entire bodice. As you can see, the bust and waist are made bigger by the full 1 1/2″ while the shoulder width has increase only by 3/8″ across the entire width. The neck is slightly  wider, gaining only 1/8″ across the width, but deepens a little as well by another 1/8″. The armhole is a little deeper, as is the length from underarm to waist.

I won’t bother showing a plain sleeve – I think you’ve all seen one of those before. Let’s see one being graded instead.

In the sleeve above, which is also graded for a 1 1/2″ difference in overall bust size, we see a few interesting things. First, the forearm and upper arm have different grading rules applied: the former gained only 1/4″ while the latter gained a full 1/2″. These are the only width changes – all the others confine themselves to the length, which needs to be added at the cap as well as at to the forearm and upper arm(note that the length of the sleeve was split in two both to allow for differing grades for the forearm and upper arm and to properly position an elbow dart, if needed). Which brings me to a little aside…

Over the years, I have heard many comments regarding length in knitting schematics. Often, knitters are under the impression that the only changes grading should bring to a pattern are to the width with perhaps some added length to the cap and armhole – the logic being that since patterns are graded to fit the same stature, they shouldn’t have added length through the body and sleeve. Sorry, but that’s wrong.

Girth does indeed add length to a pattern piece. Let’s do a little math – I know how much you all like math.

Let’s take two ovals – both are 10″ high but one is 5″ wide while the second is a perfect circle at 10″ wide. The circumference of the ellipse is 24.84 (no need to pull out your calculator, unless you want to – there’s an online calculator right here), while the circumference of the circle is 31.4. Even taking into account that the pattern piece only needs half of the the circumference to reach the top, we still have a difference of over 3 1/4″ in length. By the way, I’d like to apologize for not dressing any balls as proof. I’m too much of a slacker, I guess.

Anyhow. Spread it around, will you? I’m really tired of seeing patterns out there with the same body and sleeve length *for all sizes*.

Enough for that particular intermission. Back to grading.

I’m sure a few of the knitters will already have looked at the examples above and remarked that they are intended for cut patterns. That’s true, but I have yet to find a resource intended specifically for knitwear so used the logic above and converted to apply to knitwear using percentages and divisions. The full grade (1 1/2″) became my 100% while the other measurements were derived thusly:

Full grade: 100%

Front: 50%; full grade divided by 2

Back: 50%; full grade divided by 2

Shoulder Width: 25%; full grade divided by 4

Neck Width: 8.33% (*); full grade divided by 12

Side Depth: 8.33% (*); full grade divided by 12

Armhole Depth: 12.5% (*); full grade divided by 8

Neck Depth: 8.33% (*); full grade divided by 12

Wrist: 16.67% (*); full grade divided by 6

Bicep: 33.34% (*); full grade divided by 3

Sleeve Length: 16.67% (*); full grade divided by 6

Cap Length: 8.33% (*); full grade divided by 12

When I begin to grade a sweater, I begin by choosing my grading increment - in the case of a stockinette garment, I’d opt for a number divisible by 4 to make things easier. If it was a pullover knit at a gauge of 6 stitches to the inch, I might opt for a grading increment of 20 or 24 stitches. I probably wouldn’t choose an increment of 12 stitches as I’d then need too many sizes to run the gamut of 32″ to 50″ finished bust size – I feel it makes the pattern less legible as locating the correct size between a range of 6 (32″/36″/40″/44″/48/52″”) is easier than locating from a range of 11 or so sizes (32″/34″/36″/38″/40″/42″/44″/46″/48/50″/52″) . The fit can always be fine tuned by aiming for a gauge *slightly* looser or tighter than specified for in the pattern.

If my sweater was patterned, I’d have fewer choices as my stitch pattern would then dictate my grading increment. Whatever the increment ends up being, I convert it to an actual measurement and go on from there. A spreadsheet is invaluable for this, as a set of formulas divide the increment for each pattern piece and size, and the next set of formulas translate these measurements into stitches. Here’s an example:

This is just a portion of a spreadsheet, and it is something that no one else sees usually – it ends up being pretty messy, as I create formulas for any calculations I need along the way. You can see that the formulas suggest numbers impossible for knitwear (such as 143.9 stitches) – a line follows where I adjust the number into a whole one, and the following line indicates what measurement I should use on my schematic.

Where a spreadsheet really shines is at keeping track of small changes – imagine I was grading a sweater with a gauge of 4 stitches to the inch, for instance, and that my stitch pattern was 5 stitches wide. I’d probably have to opt for a grading increment of 5″ if I wanted to add 2 repeats of  my stitch pattern on each front and back. The grading for the  front and back would be easy, as each would need an additional 10 stitches per size. The neck width would be a different story, though, as each subsequent size would require an additional 0.41″, or 1.66 stitches:

With the help of the spreadsheet, I can use a little conjecture to decide where it’s acceptable to add a little extra width to the first, and where it is best to opt for a somewhat smaller opening.

Enough for today. If you’re still reading this (anyone?), here’s a little bibligraphy.

Grading Techniques for Fashion Design Second Edition – Jeanne Price & Bernard Zamkoff (the book I use the most, but OOP, unfortunately)

Pattern Drafting and Grading: Women’s and Misses’ Garment Design – M. Rohr (an old favorite; also OOP but the prices aren’t quite as high)

Concepts of Pattern Grading: Techniques for Manual and Computer Grading Second Edition – Kathy Mullet, Carolyn Moore & Margaret B. Prevatt Young (I haven’t read this one, but may for fun)

As always, ask any questions you may have in the comments!