So, with the initial spaghetti-breadboard proof-of-concept phase behind me, it's time for Phase 2. I always like to get my projects into another person's hands as soon as possible. I mean, I think the Twister is good and that I'm on the right path, but is this actually true or am I just being optimistic? I have the perfect opportunity to get it into a professional dyer and peer's hands this weekend, at the Vogue Knitting Live event in Pasadena. Brooke Sinnes of Sincere Sheep is one of those rare people that won't fixate on the current prototype-y form factor, but will give excellent feedback about the general function and her particular use case.
There's nothing like an external deadline for motivation, so I've been busy taking the breadboard version, and soldering it into more stable & permanent perfboards. I love these little 1" square guys from Sparkfun, I try to always have a handful of both the single and double-sided ones on hand. Super useful.
The SkeinTwister electronics have gone from the spaghetti mess you see in the background here:
To the slightly less spaghetti-mess cable harness on the bench:
To installed in an enclosure:
The two connector inputs at the top are 12V from the power supply brick, and a 1/4" audio plug from the foot pedal. The foot pedal is momentary, currently configured so that one tap starts the motor, and another tap stops it. The green board in the upper right is the motor current feedback amplifier (the SOT-23-6 chip on the green adapter board), and the motor power control FET. In the lower left is the processor chip and a 12V to 5V linear regulator. I've kept the processor (an Atmel ATTiny85) in a carrier to make software updates easy. I can simply send Brooke a chip with new code if I have to, and she can easily replace it.
The top of the enclosure shows the twist-firmness knob, and the ever-present-on-a-prototype blinking LED:
The knob sets the automatic cutoff on twist torsion. The dyer can set it high for large skeins that need lots of torque to bundle up nicely, and lower for mini skeins that need less torque. When the foot pedal is pressed once (momentary action), the motor will turn on and the hook will start twisting the skein. When the proper amount of torque/current is sensed, the processor turns the motor off automatically. Or, if the setting is too high for the skein you have and your fingers are about to have the life squeezed out of them, you can press the foot pedal again and it'll stop. Ask me how I figured out the need for that feature.
The wires exiting the right side go to the motor, which for the moment, is simply clamped directly to a table, like the first photo above. I've connectorized it, and I'll actually give Brooke a couple of different motors to try. Some are faster, some are slower, some are torquier, some are less torquey. I know which I like best, we'll see if she concurs. Mechanical design is next on the development list, for the hook, enclosure, and clamping mechanism.
This is obviously a rough prototype, but I believe strongly in getting user feedback early, before I'm too far down one path. It makes changes easier and less expensive. I also believe in getting feedback on pre-production prototypes - what I define as my first attempt to make a real production unit. Those will have custom hook, enclosure, and clamping mechanisms, as well as actual PCBs. I'll make roughly a dozen of them, and have a beta test group run them through their paces. Then I'll have representative feedback from a variety of different use cases and conditions, and be able to make adjustments before committing to a production run.
I'm excited for the first unit to fly the nest! I might even be able to demo it for a few more dyers at the show this weekend, the more feedback the better. Hopefully I'll get a few videos to post too!