Finally, finishes for knits that don't look home made

Knits can be trouble. Because they stretch, they want to stretch as they go under your presser foot resulting in puckered, stretched, or lettuce-edged seams and edgings that look horrible at worst and home made at best. Here are some things I've found that help create professional looking finishes.

First of all, if you have a walking foot, use it. My primary machine is a Pfaff 1471 with a built in walking foot, but because this pulls from behind the foot rather that in front, I still had stretching when I sewed. A really good combination is Singer's rubber feed dogs (on older models) with the even feed foot. This has the capacity for extremely precise sewing, especially on difficult fabrics, as long as the machine will tolerate the fabric.

Secondly, make liberal use of laundry starch and let it dry before continuing. This seems to take forever and ironing doesn't speed up the process much. Sometimes spray starch is enough but other times it can be enormously helpful to make up a batch of Argo starch and dip your fabric in it before it is cut. The starch really helps control rolling edges and gives body to unruly lightweights that machines sometimes struggle with. However, my Singer combination described above just would not tolerate the lightweight T shirt knit shown below (this was before I learned about using boiled starch). Fortunately, my primary machine did, but if your machine is struggling there are a few things you can try. The only one I know off the top of my head is to sandwich the fabric between layers of tissue paper. Another thing that might help is to decrease presser foot pressure.

For edgestitching, adjust your needle position so you get as much fabric as possible under the foot. And obviously, you should be using ball point needles. Most knits require a seaming stitch with width, but if a straight stitch is used, it can be advantageous to use the straight stitch foot and throat plate.

Hemlines

I did some internet research for hemming knits and almost every site recommended stitching the hem with a twin needle and woolly nylon in the bobbin. Even Nancy Zieman says to do it this way. The combination of zigzaging and built in stretch of the woolly nylon is supposed to give the seam a lot of stretch. Maybe it's just me, but every time I try to do this, I end up with a tuck between rows of stitching. Even if I manage to stabilize the fabric enough to stay flat during the stitching, as soon as I flex the seam, the tuck appears and won't go away. After lots and lots of experimentation, I came up with a good, flat hem solution for casual garments.

Bottom row is wrong side

This is a feather stitch over the turned up, raw hem accomplished the same way you'd do a cover stitch. The raw edge doesn't need to be finished because knits don't ravel and finishing adds unnecessary bulk. Stabilize the entire hem with spray starch if you didn't dip the fabric before starting your project--from the hem allowance to the stitching line--and press it to the wrong side. Stitch from the right side, placing your line of stitches to catch and cover the raw edge. You can either stitch a fixed width from the edge and cut off any excess later or you can adjust your stitching according to the location of the cut edge if your cutting was precise. You can feel the cut edge under the top layer while you're topstitching so you know where to sew. I figure nobody's going to be looking closely enough at my hem to notice whether the stitching varies by an eight of an inch here and there and it saves me the trouble of doing any trimming later. I was surprised by how good this looks on the finished garment. It lies flat and mimics coverstitching very well.

Before I discovered this, I'd been finishing the edge with serging and zigzagging the hem seam:

You don't need to do this

This just doesn't look as good.

 Another good way to finish an edge when a narrow hem is desired is to do this:

I don't know if you can see this well, but this hem was turned up 5/8" and zigzagged 1/4" from the fold. Then I trimmed off the excess close to the stitching. I expect it to try to roll after it's been cut a while, so I'll come back later and trim again when it does. Again, flat seam that isn't bulky. Update: on this particular fabric, this hem isn't stable and the whole thing rolls.

Here's one I thought was clever at the time:

This is a 3 step ZZ on slippery Under Armour in green thread. I thought the green would look somehow bolder but I'm glad it didn't because the stitching is irregular due to the slipperyness of the fabric. If I'd used white thread it wouldn't have mattered. This whole shirt is kind of a disaster as you'll see below.

If you have a fabric that still won't behave, you can fuse a strip of knit interfacing to the hem allowance before stitching. Steam-a-seam was recommended to me so I went out and bought some. When I used this, the hem was stiff and heavy and had no stretch at all. It made it look like masking tape was stuck to the underside of the fabric. You definitely want to use knit interfacing if you do this.

Necklines

Necklines give me the most grief. Almost any kind of treatment (binding, facing, turning under) comes out stretched or otherwise botched. Ripping stitches out of a knit is a nightmare and I've had to rip out many a neck treatment. My first attempt at making a neckline without ribbing was the under armour shirt mentioned above:

I turned the edge under twice and stitched it, not knowing what was going to happen. I ripped this out and put on foldover elastic but cut it the length of the neck opening circumference. This was even worse. I ripped out again and cut the elastic shorter with no improvement. I ripped it out yet again and let it sit for 9 months. By this point, the neckhole had grown considerably from all the stress.

In the meantime, I learned how to make this from a back issue of Threads:

In the May 2003 issue, there is an article, "Neckline Finishes for Knits," that describes various finishes for knit necklines, including this one. The author also has videos on Threads' website that show you how to do it. I don't know what to call this; it isn't bound and it isn't ribbed, although the process is identical to applying ribbing. The secret is the difference between the binding length and the neckhole circumference. Your binding has to draw in the bodice fabric without gathering it so you have a nice, tight finish. The seam is stitched with a narrow 3 thread overlock so it stretches. This was a breakthrough.

Encouraged, I went back to the under armour with my new technique. Now I have success!

I didn't even bother to change the thread in the machines. I just used what was already there. The shirt was already such a mess, it isn't going to matter. What does matter is that I can do this! And since I learned about this, I've noticed how many of my ready to wear clothes have this finish. Some of them have the seam allowance topstitched, but when I tried to do this, I got another stretched out neckline. I just let them float. They're not going to stay where I want them anyway.

A few hints that are in the video but not the article:

1. Sew the binding to the neckhole with the garment lying flat across the machine bed like this:

so you can sew a curved seam and also see when you need to stretch. She's got the foot inside the neckhole with the shoulders on either side.
2. Stretch in the same areas as you did when determining the length of the binding, i.e. more for sharper curves and less for shallow ones.
3. Don't stretch the bodice while sewing if you can help it.
4. You may need to stretch the seam slightly while serging.
5. Use a wider seam allowance when serging in order to cover the basting stitches. This is where I go wrong a lot. Having an adjustable seam guide on your serger is invaluable.

You can sew a piece of ribbon over the seam allowance to stabilize the back of the neck if it wants to stretch too much.

Who would have thought...

...to bondo a sewing machine? I didn't like the results I was getting with my paint layers. They were not going on smooth and many, many casting flaws were still obvious. While it's true that I have worked on a lot of cars, I never worked on the OUTSIDES. I don't know a thing about shiny smooth finishes. I finally asked for advice from a guy in Florida who does amazing finishes on Featherweights professionally. He helped me understand what I've been doing wrong.

To rectify the problem, I bought more oven cleaner and stripped the machine down again. This time proved to be significantly more difficult than the first. Then I primed and sanded it twice and then bondoed it:

That pink rash is all casting flaws. They were showing up as craters in the black paint. The black stuff is where the metal is showing through after sanding off the surplus bondo. The surface is very smooth and another coat of primer will reveal any flaws I missed. I probably don't have another coat of primer left in my spray can. This project is starting to get expensive.

Kool aid dying

Whenever I read about something, I nearly always have to try it. I have no real interest in custom dying yarn, but this sounded like fun. Turns out, it is. The problem is what I decided to do with the dyed yarn. The project I settled on was a fair isle sweater manually patterned on the Bond. The process was slow and tedious and I haven't gotten farther than making one piece. Plus there's my tendency to work on a project till I find something else that catches my attention, and then the first project gets put on hold indefinitely. That's why updates here are far between. Here's what I've got so far:

Looks pretty good, but I don't know about that bright blue. Kool aid is limited in the blues available, so this is the best I could do. I really wanted to include the whole spectrum because it was so heavy on reds. The plan was to have warm stripes alternating with cool stripes with yellow being the unifier. I don't think the blue is quite as loud in person.

The pattern is one I dug up on the internet. Getting a pleasing color sequence turned out to be quite difficult. I tried many combinations before settling on this one. It was accomplished using the two carriage technique. I highly recommend this to anyone who does fair isle on the Bond. It is far preferable to hand knitting every contrast stitch.

The dying is quite easy and there's no science to it. The yarn will absorb all the dye from the bath, leaving the water clear. The quantity of kool aid in the bath determines the intensity. Due to the nature of the process, you can't really stop once you like the color, so you have to start with the right amount of powder rather than try to stop early. It's well documented on the web, so I'm not going to give the particulars. When it's cool, rinse it well, and most of the smell will come out in the rinse. I believe this only works with natural fibers, or it could be only wool. I'm not sure. I used Knit Picks Wool of the Andes. The hot water will cause the wool to felt, so don't agitate it too quickly or it will ball into a prickly mess. Also, if you're trying to match dye lots, make careful notes about the quantity of powder used. Even then, the results are unpredictable. I haven't washed this yet, so I don't know how colorfast it is. I've read that it doesn't fade but don't know for sure.

Ivy League Vest

I wanted to make a fair isle vest in reds and browns, so I settled on Eunny Jang's Ivy League vest.

It's all nice and tight on the model. Mine came out rather loose:

The colors look unnatural in flash photography, but in real life, they aren't this loud. The yarn is Palette by Knit Picks.

I know I said in an earlier post that I was done with hand knitting, but machine knitting this wasn't going well. I punched a card for it and started knitting, changing colors as needed. The color changes were pretty tedious, and every mistake required ripping out the row and resetting the row counter and punch card. The last straw was when I got yarn wrapped around one of the patterning drums. I was pretty irritated and ripped hard on the yarn, damaging the drum. That's when I resigned myself to hand knitting the project.

The gauges between the hand knit and machine knit products were vastly different:

I can see now that I was knitting much too tightly on the machine. You don't know how well or poorly you're doing until you actually take it off the machine. The swatch I made was nowhere near this tight. Owing to the complex shaping and neckline, I had planned to cut and sew this vest by making rectangles of fabric on the machine and seaming them on the sewing machine. Plus the machine knit sample just looks better because all the stitches are the same size. Mine are all over the place.

My hand knitted gauge wasn't right either and I had to omit rows to get the length right. The stitches per inch were the same as the given gauge, yet the vest is still too loose. It is unfinished in the photo above, but we had one cool day this spring and I knew I wouldn't have another chance to wear it for a long long time. I expect I'll have a go at shrinking it after I've dealt with the strings.

Steeks are pretty neat. This vest was knitted round with steeks placed at the armholes and neck as placeholders to be cut out later. You still shape the armhole and neck as normal, then knit across the steeks to the next section in order to continue knitting round all the way to the shoulder bind offs.

66 Update

Good news on the 66 front! There's going to be a decal source soon which has renewed my interest in the project. I ordered the electroless nickel plating kit from Caswell and now I have just about all I need to get the show on the road. The kit looks extremely complicated.

 

It came with a custom printed manual an inch thick, and two little stockpots. You have to make all these calculations about the area of the piece being plated and how much nickel in the bath is being consumed so you can replenish it as it is depleted. Looks like a job for a weekend.

Quite some time ago, I started painting the head black, but I used a brush because I wanted to build up a thick coat like the original Japanning and then spray on some thinner ones. Considering how much sanding is required to remove the brush strokes, I'll probably be back down to the primer when I'm done. Perhaps it wasn't the best decision.

The next steps are to finish sanding the head and prep the metal parts for plating. Shiny ones like the throat plate will need to have a mirror finish if they are to look anything like they used to. I could just buy new ones, but that conflicts with the original plan of only replacing parts that I can't restore sufficiently, like the etched chrome covers. I especially want to save the slide cover because it has patent dates stamped on it, all before 1910. The goal was to do this cheaply, but I don't know if that's going to be feasible. I figured the cost of the plating kit would be lower than buying all the parts piecemeal.

I showed the handwheel in an earlier post. It still looks the same. Since the plating process is finicky about absolute cleanliness of the metal, I may be able to only plate the outside diameter and leave the spoked part for painting. I don't really want to buy another one, as it has the potential to be very expensive for this budget.

Let's recap the spending so far: we were up to $82 and the plating kit was another 80. That makes $162 and hopefully there won't be much more to buy. I'm using a lot of supplies that I already had, like black paint. However, I did buy supplies to do a hand crank conversion since I used the treadle table on the 306: $45 for a handmade wooden base that will have to be stained, and $12 for a hand crank. I'm basing my budget on machines for sale on ebay, which are often in rough shape and sold with no base or table, yet still have high prices.

Bob Ross style painting

Bob Ross is the ultimate motivator. After years of watching Bob paint hundreds of 30 minute masterpieces, I finally decided to try it. I've never painted before as you can see, but it really is easy once you get a little practice.

I bought his master paint set which comes with instructions and a glossy print of the painting project. The example painting is quite beautiful. Mine wasn't.

Then I tried doing a black canvas and only using 3 colors. The result was almost good:

What am I gonna do with these things?