![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
mapsedgeThe machine is coming along, but progress has slowed greatly. I keep finding things that need to be done differently or additionally, and that pushes completion time way out. I shall record some of them here to they're on record somewhere:
1. On the first full-scale test, the fabric all bunched in the middle. I thought this was because the driver was bending under the weight of the wet fabric, but the rollers are rigid. However, they fit inside the foam covers with quite a lot of slop. The foam is only secured at the ends, allowing it to sag in the middle until it meets the core and stops.
SOLUTION: make the core bigger around so that the foam cover fits tightly, without any slop. Bubble wrap and duct tape might do it.
2. The fabric is still going to wrinkle. Thinking there has to be some mechanism to keep the fabric spread flat, I looked online to see if I could see how manufacturers do it. Another Indian (dot, not feather) video showed me the light. A company showing off their embosser gave a full video tour of the device and there it was: a roller with a screw profile, radiating outward from the center.
There are several possible ways to make this. So far, I've considered: 1. a bead of hot glue around a roller; 2. one of the roller cores with a hard plastic spiral forming the screw; 3. create a screw in wood using the router as a lathe. I tried option #2 when I got home from work, using a scrap of 3/4" pvc pipe as the core, and another scrap of the same cut into a slinky. It worked, though the slinky part would need some real softening. I don't want those edges dragging across my fabric.
I finally figured out how gear ratios work. I've been trying for a couple weeks now to understand them, knowing that I want the Dye Machine driver to run about 15rpm. Driving to physical therapy this morning doing calculations in my head and suddenly it clicked in a way it never had before. For my own benefit, here's how it worked in my head:
Starting RPM: 750. Target RPM: 10. That's 75:1 ratio (750/10 reduced to 75/1, expressed as 75 to 1 or 75:1. None of that is really important.)
The smallest gear one can reasonably make is a 10 tooth gear (10t.) So I could accomplish what I want just by having my driver be 10t and my driven be 750t. Done. That's a huge damn gear though, so we need to break it down.
Gears multiply. If you have a compound gear setup like so: 10:40/10:40 that's (reduced and expressed as a math problem) 1/4 * 1/4 = 1/16. So my two 10t:40t is giving me a ratio of 16:1. Compound it again with another 10:40 and you get 64:1.
Funny. I just solved the 75:1 problem. The final reduction is 750 / 64 = ~12. One little tweek: if one of those gears is a 47t, we get 75.2:1. Pretty damn close. You could I presume keep tweaking and get it exactly right, but that's way too much effort.
Anyway, that's how it works.
1. On the first full-scale test, the fabric all bunched in the middle. I thought this was because the driver was bending under the weight of the wet fabric, but the rollers are rigid. However, they fit inside the foam covers with quite a lot of slop. The foam is only secured at the ends, allowing it to sag in the middle until it meets the core and stops.
SOLUTION: make the core bigger around so that the foam cover fits tightly, without any slop. Bubble wrap and duct tape might do it.
2. The fabric is still going to wrinkle. Thinking there has to be some mechanism to keep the fabric spread flat, I looked online to see if I could see how manufacturers do it. Another Indian (dot, not feather) video showed me the light. A company showing off their embosser gave a full video tour of the device and there it was: a roller with a screw profile, radiating outward from the center.
There are several possible ways to make this. So far, I've considered: 1. a bead of hot glue around a roller; 2. one of the roller cores with a hard plastic spiral forming the screw; 3. create a screw in wood using the router as a lathe. I tried option #2 when I got home from work, using a scrap of 3/4" pvc pipe as the core, and another scrap of the same cut into a slinky. It worked, though the slinky part would need some real softening. I don't want those edges dragging across my fabric.
I finally figured out how gear ratios work. I've been trying for a couple weeks now to understand them, knowing that I want the Dye Machine driver to run about 15rpm. Driving to physical therapy this morning doing calculations in my head and suddenly it clicked in a way it never had before. For my own benefit, here's how it worked in my head:
Starting RPM: 750. Target RPM: 10. That's 75:1 ratio (750/10 reduced to 75/1, expressed as 75 to 1 or 75:1. None of that is really important.)
The smallest gear one can reasonably make is a 10 tooth gear (10t.) So I could accomplish what I want just by having my driver be 10t and my driven be 750t. Done. That's a huge damn gear though, so we need to break it down.
Gears multiply. If you have a compound gear setup like so: 10:40/10:40 that's (reduced and expressed as a math problem) 1/4 * 1/4 = 1/16. So my two 10t:40t is giving me a ratio of 16:1. Compound it again with another 10:40 and you get 64:1.
Funny. I just solved the 75:1 problem. The final reduction is 750 / 64 = ~12. One little tweek: if one of those gears is a 47t, we get 75.2:1. Pretty damn close. You could I presume keep tweaking and get it exactly right, but that's way too much effort.
Anyway, that's how it works.