Monday, February 28, 2011

Drilling A Hole In My Loom

It's so hard for me to desecrate the "pristine" wood of my very expensive AVL loom, but sometimes I have to. It is just a tool, after all, and needs to do what I want it to do. Thank goodness I have enough experience in the wood shop to know how to do this modification and get good results.

I'm having trouble with the plied threads catching on each other, but it's impossible for me to diagnose and fix without being able to lock the shed open and walk around the loom. It was a simple fix. I just needed to make a peg on a chain and drill a hole in the dobby box so that the new peg will lock the dobby arm down and hold the shed open.

And here are the tricks to drilling a clean hole: clamped blocks and guide holes. First, I marked the location and drilled a tiny hole in the dobby box.

Then I drilled matching holes in two blocks of wood and used a shibori needle to line up the three holes while I securely clamped the blocks into place. These blocks keep the edge of the hole from splintering when the drill bit enters or exits the wood.

And here's the result: a new peg on a chain and a hole to insert it.

Here's one new view of some thread problems in the shed. It'll really help me to be able to fix them in the best way possible if I can stand up from my bench and get at these things from multiple angles.

Sunday, February 27, 2011

Celestial Geometry and Birdhouses

We've decided to encourage the tree swallows to settle in our valley again, this time in nest boxes high up on slippery poles. That should prevent the snakes from getting the young.

We also decided to increase the number of boxes in the hopes of increasing the number of cheer-bringing birds. 12 seems like a good number, and makes for an interesting construction challenge.

As long as we're making a new feature on the land, I'm taking it upon myself to practice celestial navigation and surveying geometry. I'm planning the 12 new posts to lie evenly on a circle that's oriented to true North. We're doing the orientation and construction using only two pieces of string and our eyes.

We had chosen our center location and a radius that would have the nests far enough away from each other and the places where people travel. Then, we were just waiting for a clear night to find North.

Last night, I woke up at 4:00 AM to find the sky crystal clear. It was a perfect time to go out and sight the center line. I drove a stake lined up with the center stake and Polaris to capture the North line.

This morning, Wispr and I went out with some wool yarn to place the rest of the stakes. The first thing that we noticed was that the radius stake, which we had placed "roughly" in the north, was absolutely dead on. I guess we have a better understanding of North than we had thought.

With the 12:00 stake placed, we used line-of-sight and one of the radius strings to place the 6:00 stake.

With the center, 12:00, and 6:00 stakes in place, we can use two copies of our radius string to place the rest of the even-numbered stakes. We now have a hexagon oriented to North.

The next step is a specific to the medium - fold one of the radius strings in half and tie a knot. Now, when that string is stretched between any pair of even-numbered stakes, a radius from the center that passes over that knot will give us the location of the odd-numbered stake between them.

And, that's it! Now we can dig the post holes and sink our poles to create a near-perfect 12-pointed circle aligned to North.

More Samples

The long-repeat meandering zigzag is definitely a success. It creates visual tension against the straight lines of the individual threads, and helps the cloth to look more organic and complex.

I promise that I'll soon be posting pics of the washed cloth. The pattern really pops out when the cloth shrinks.

Saturday, February 26, 2011

Arctic Fox Samples

It's photo essay blog catch-up day. I'm working my butt off for a March launch of my new line of wedding shawls and cloaks, inspired by the coloration of the Arctic Fox. In the next few days I'll post closeups of the color samples after they've been washed and dried.


Here's another quick photo essay on pulling threads through the heddles. It's tough to convey how finicky this work is, particularly because I don't stop halfway through it to take pictures.

[...and then a lot of work happens...]

Warping Frame

Wow, the weeks are flying by like days right now! The weather is grey and cold so it's tough to get up the energy to get my stuff done. I've been doing it, but haven't had the extra energy to write about it.

Today I wet-finished the sample swatches for the new beam, but I've got some other stuff I want to write about before I show them off. First, is the other new piece of equipment that I had to make in order to warp the beam...

The last time I did this, I made due with a reed hung from the ceiling. It was very slow and finicky, partly because I could not get the cones directly below it.

The solution is to set up a device so that each cone feeds to the section of a screen that lies directly above it. This lets cones feed easily without tipping over.

There is one big problem with this setup, though. I wind from the center outward, alternating sides. Since the pattern is roughly symmetrical, I have ended up reversing the cones for each section. This means unthreading them from the screen, reversing the cones, making any changes to the thread order, and rethreading the screen.

Before the next beam, I've got an idea that would cut out a lot of work: I'm designing a rolling cone rack and warping frame. I'll be able to reverse the cones just by turning the rack around, and roll it out of the way at the end of a days' work. It will also have pegs so I can stack a cone with its replacement cone and it won't fall off.

So, this rack worked well enough and showed me how I can make it even better for next time.

Friday, February 25, 2011

Setting A Price On Craftwork: Task Switching

In my last post on the topic, I mentioned the idea of doing one operation to many pieces. I want to talk a little more about that.

In project management, they talk about "startup cost". In accounting, this refers to the costs associated with starting a business. In project management, it refers to the time that it takes a person to start a task or change from one task to another. It is critical to keep this in mind when doing craftwork for a living. Let me start with an example...

Let's say a jeweller needs to make 100 pairs of earrings. Each one has a square of stone hung from a French hook by a rim of gold. The intuitive way to proceed is to cut and polish two stones, cut the gold, solder, burnish, attach the hanging ring and attach the French hooks. Whew! That's one pair done.

Now let's look at the startup costs.
1. Get out stone, put on protective gear and start up the cutter. 2 minutes. Cut two stones.
2. Put on a coarse grit belt. 3 minutes. Rough out stone.
3. Switch to medium grit belt. 2 minutes. Start polish.
4. Switch to fine grit belt. 2 minutes. Finish polish.
5. Get out gold, metal snips, etc. 3 minutes. Cut gold.
6. Get out soldering iron, etc. 2 minutes. Solder gold.
7. Get out shaping and burnishing tools. 2 minutes. Burnish gold.
8. Get out mounting hardware. 2 minutes. Attach hardware.

Overall, it's probably taken an hour to make this pair of earrings, 18 minutes of which was spent switching tasks. This is a low estimate based on quickly switching tasks and staying focused. Many people don't have a separate area for each task so things need to be moved around a lot when switching tasks. Tools can get misplaced in the process, taking even more time.

If you're anything like me, task switching also feels like a great time for a quick break. 5 minutes here, 10 minutes there and poof! The day is gone.

This is the point where many crafters look backwards at the numbers. "I bring in $25 for a pair of earrings and they only cost $5 in materials and take an hour to make, so that's $20 an hour." Once we take income tax, overhead, and the financial and time cost of making the sales, it's probably more like $8-12/hr.

To reduce the startup costs, this jeweller could do each operation to all 100 pairs and the startup costs would drop to 1% of what they were. That 18 minutes of startup cost per pair of earrings becomes 10 seconds per pair. To make the whole batch, she would have saved 30 hours in repetitive tasks and her wage will have jumped to $12-18/hr.

If she could find a niche or improve the customer experience in some other way, she could probably raise her retail price and earn the $20/hr that she wanted.

In my space-limited studio, I've reduced repetition by performing my tasks in a pipeline. First I warp the beam for 80 yards of cloth, then I weave it all and store the cloth on top of the loom or in mouseproof boxes.

Even the weaving is pipelined. I get out enough weft yarn to finish the cloth I'm making in that color and line it all up on a shelf. I wind all the bobbins I have (40) and keep them organized in a box. Then I sit and weave until they're gone or my body needs a break. The cloth collects on the back beam until 1. the current color is finished or 2. there are 30 yards of cloth on the take-up beam.

And, when it's time to sew, I reconfigure the studio for that task and make lots of whatever product it is that I'm making. It's easier to pull out cloth and make many colors of the same product than it is to refer to my design notes and retrain myself to measure, cut and sew a different product.

So, if you find that you have to charge prices higher than your competition in order to pay yourself a fair wage, look for ways to cut out task startup costs. See if you can make many copies of an item in less time than you could if you were making them one at a time.

Monday, February 21, 2011

Als Ich Kann

For those who've been reading my blog for a while, you may remember a stint last summer when I was obsessively reading every issue of The Craftsman magazine, published by Gustav Stickley from 1901 until 1916. This magazine was at the forefront of the development of the Craftsman movement, best known for mission-style furniture and the bungalow architectural form.

At its heart, this movement was not about style. It was about honesty. The leaders of this movement believed in honest craftsmanship resulting from honest work and the use of quality materials. They also believed in forming a strong relationship with the land, which must be completely honest. (You cannot be dishonest about watering the garden and expect your food to grow!)

Stickley used a phrase as the slogan for his work. It is said to be Flemish in origin, "Als ik kan." In German, which I understand, it would be "Als ich kann" and the meaning is just the same - "In the best way that I can."

The more I reflect on it, the more applicable it becomes to my own work. It's very easy to analyze my work in terms of pure profitability, making decisions that will result in a higher volume of sales at any cost. If I did this, I think that the honesty and joy would drain from my work very quickly and I'd be left slogging through it, no better off than if I worked an unfulfilling job for a boss I didn't like.

I use Stickley's motto to push me in a different direction. "IN THE BEST WAY that I can." The best. That's a moving target, isn't it? Every piece of cloth that I produce teaches me ways to improve my craft. Sometimes I need to build new equipment to create a better result or speed up the process. Sometimes I learn valuable lessons about the limits of my materials or ways to make them perform better. Sometimes I learn that there's a new tool that will improve my craft. I'm wary of that last one, though. It's easy to tell myself that some new tool would make me a better weaver, and rare that it's actually the case. A tool, especially an expensive one, needs to fill a clear and present need before I'll consider it. I'll exhaust all of my possibilities first.

"In the best way THAT I CAN" This tells me that I shouldn't settle for producing a lower quality of cloth because I think that the customers won't notice. I need to make sure that every piece of cloth is up to the highest standards that I can set. Sure, most customers may not have the skill to recognize the reasons for the high quality of my cloth, but I'll bet that they can sense that quality anyhow. What will nurture me as a craftsman is the admiration of experienced weavers. This will come, not only from the quality of the cloth, but from the low prices I'm able to charge because of my efficiency in producing it.

In winding the current beam, I made some choices that are costing me time, but producing a much better result. I've begun using very thin thread and loosely plying it together with a cone winder. It takes a few days to ply the first set of cones and about 20 minutes a day to keep the winder busy, replacing cones that run out while I wind the beam.

There are a few other time costs associated with this new method. While winding, it takes an extra 15-20 minutes per section to tie a knot in each thread. This is necessary so that I know which tiny threads were plied together in the winding.

I'm also sinking a bit of time into the thread patterning. There's a very noticable feature in this beam - a mottled grey and brown stripe running down the center of the cloth and fading to white at the edges. There are several other subtle things happening as well. There are noticable threads in each section, two shiny champagne threads, 5-15 dark threads, and 1-2 thick/thin white yarns. I am using my lifelong obsession with finding the repeats in manufactured patterns to create micro-interest in this cloth and help it to look more organic. For the shiny threads, I move them around in each section so that the 2" repeat isn't obvious. For the dark threads, I reduce the number as I move out from the center and move them around to hide the repeat. For the thick/thin white, I gently increase the occurence as I move toward the edges. This whole process takes about 5 minutes per section.

There's one last time cost that I'll be mitigating before the next beam. I'm weaving cloth with a color pattern that's roughly symmetrical. This means reversing the order of the cones on the floor for each section and rethreading the spacer. This 20 minute per section cost will disappear when I build a rotating table to hold the cones and the spacer, letting me easily turn them together without rethreading.

Another way to mitigate these time costs will be to buy a bigger beam. As it stands, the time that I invest in setup needs to be paid for by the 80 yards of cloth I can produce. If I produced 200 yards of cloth, the time cost per yard would be 40% of what it is now. Instead of paying for 45 minutes of setup time from each yard of cloth, I'd only be paying for 18 minutes.

So, with a bit more investment in equipment, I'll be producing higher quality cloth with the same time investment per yard that I spent on cloth of a lesser quality. This never-ending pursuit is my application of Stickley's motto, "als ich kann."

Tuesday, February 8, 2011

Homemade Warping Tools

I've been so busy doing the work that I haven't taken time to write about it. I'll try to rectify that in the next few days, writing about what I've been up to and where I'm going.

Now that I need to warp all of my own beams, I've decided to create and modify some tools to make that task easier and produce better cloth.

First, I modified the tension box that I made a couple of years ago. I removed the skinny pegs, replacing them with threaded metal inserts so I can screw wider pegs in instead. This will increase the potential tension in this box and give me better control.

Then, I removed the slot that holds the comb and replaced it with a bolt and wingnut. By making the comb rotate, I have precise control over the width of the threads leaving the box. The wingnut lets me lock it into any rotation.

Next, I needed to be able to slide the tension box in front of the beam and lock it into any position. I needed a track. The loom is designed to accomodate a track like this, so I just used the existing holes to tell me where to drill oak plywood for my own homemade track. The track itself is split down the center so the bolts can pass between the two halves, allowing me to slide it along the full width quickly and easily.

Between the modified tension box and the new track, there's no more futzing with a bulky table and clamps. I don't have to guide the threads into place with my hand anymore, giving me more accurately wound beams.

The final thing that makes winding beams difficult is the counting. It's very hard to keep track of all the threads and count the beam rotations at the same time. For one thing, there's no clear indication of a single rotation. When the beam is turning, it just becomes a jumble of passing pegs. With the way I've been doing it, I need to stop and hold the count in my head if a thread snags, breaks, or a cone runs out or tips over.

One of the last things I learned from Annie was a method for keeping track of beam rotations without counting. I've made a wooden piece that inserts into the end of the beam and holds a threaded rod. From that rod, I hang a brass bell using a thin wire.

For the first section, I mount the rod so that turning will move the bell toward the loom, away from the end of the rod. I place the bell at the very end of the rod and wind that section until it's completely full. I use masking tape to mark where the bell ended up on the rod. Then I move the mechanism to the other end of the beam, ready to turn in the other direction when I wind the next section.

For each subsequent section, I place the bell up against the masking tape and wind until the bell falls from the end of the rod with a loud clang. And, just like magic, I'm freed from counting at all. I can stop and start at will, replace cones, fix snags, and do whatever else I want without ever worrying about how many times the beam has turned. I can stay focused on making sure the threads behave to produce the best possible cloth.

There's just one more tool that I needed to make for winding the beams, but I'll write about that tomorrow.