stepper motor linear actuator - test results
The important things I wanted to test with this simple setup were:
1) do these things work at all?
2) how strong are they (roughly)
3) how straight are the rods going to be?
4) how precisely will the "nut" move along the threads?
5) how fast can these things go?
I did get one of the actuators working, and it works great. It's strong enough to move whatever I'm going to give, which is great for only 5v. The rods were a little off-center, but I was able to gently bend them back to "straight enough." Whatever the nut gets clamped to will hold the rod firmly in place, so I'm not worried about that. The precision is tied to the speed. These are almost too precise for me. In fact, I'm going to go out on a limb and say they're most definitely more than I'll need. These are accurate enough to place grains of fine sand next to each other. According to the spec sheet, one step = a linear movement of .00012" (.003048mm), which is incredible.
In the stamp code, I left a variable delay between steps. If I set this to 0, the motor would sing, meaning it wouldn't move at all, but the shaft would vibrate in place fast enough to emit a high pitch tone. If I set it to 5, it moved slowly, but wasn't very strong. If I put in a delay of 10ms, I got what seemed to be the optimum movement, but here's the bad part. With a 10ms delay between steps, it took the nut over 256 seconds to move across its entire threaded rod, which is only about 6" long. That's over 4 1/4 minutes!
Hopefully I'll get better results with the PIC microcontroller. Both the stamp and PIC are 20MHz, but it's been my observation that the PIC does things far more quickly. There might be more to the stamp's timing that I'm not understanding, which is very likely with me, but honestly, I don't hold out hope for making these things very much faster. I'm considering upping the voltage a bit, even though I was using 5v here, and they're rated for 5v. If I keep it stepping at a higher rate, perhaps the PWM effect of a higher frequency of shorter bursts will allow me to move them more quickly without killing them. I'm wondering if I can ramp up voltage with stepping speed, such that for more intricate areas, I use less, and and 5v when I come to a stop, but maybe when it's cruising at top speed, it's using more like 9v. Does that make any sense?
One future problem will be if I decide to engrave or sculpt anything, or carve out my own printed circuit boards (a small dream I've had since high school), I'll need something like a Dremel multitool, which is loud. With this thing going so slowly, it will have to run for hours, perhaps even a day or two to finish up. I live in an apartment, and can't subject my neighbors to that kind of punishment. I've considered options like building a sound-proof box, which has much of its own appeal, but who knows what I'll eventually decide. Either way, I don't think I really want to leave a Dremel running overnight.
stepper motor linear actuator - test results
The important things I wanted to test with this simple setup were:
1) do these things work at all?
2) how strong are they (roughly)
3) how straight are the rods going to be?
4) how precisely will the "nut" move along the threads?
5) how fast can these things go?
I did get one of the actuators working, and it works great. It's strong enough to move whatever I'm going to give, which is great for only 5v. The rods were a little off-center, but I was able to gently bend them back to "straight enough." Whatever the nut gets clamped to will hold the rod firmly in place, so I'm not worried about that. The precision is tied to the speed. These are almost too precise for me. In fact, I'm going to go out on a limb and say they're most definitely more than I'll need. These are accurate enough to place grains of fine sand next to each other. According to the spec sheet, one step = a linear movement of .00012" (.003048mm), which is incredible.
In the stamp code, I left a variable delay between steps. If I set this to 0, the motor would sing, meaning it wouldn't move at all, but the shaft would vibrate in place fast enough to emit a high pitch tone. If I set it to 5, it moved slowly, but wasn't very strong. If I put in a delay of 10ms, I got what seemed to be the optimum movement, but here's the bad part. With a 10ms delay between steps, it took the nut over 256 seconds to move across its entire threaded rod, which is only about 6" long. That's over 4 1/4 minutes!
Hopefully I'll get better results with the PIC microcontroller. Both the stamp and PIC are 20MHz, but it's been my observation that the PIC does things far more quickly. There might be more to the stamp's timing that I'm not understanding, which is very likely with me, but honestly, I don't hold out hope for making these things very much faster. I'm considering upping the voltage a bit, even though I was using 5v here, and they're rated for 5v. If I keep it stepping at a higher rate, perhaps the PWM effect of a higher frequency of shorter bursts will allow me to move them more quickly without killing them. I'm wondering if I can ramp up voltage with stepping speed, such that for more intricate areas, I use less, and and 5v when I come to a stop, but maybe when it's cruising at top speed, it's using more like 9v. Does that make any sense?
One future problem will be if I decide to engrave or sculpt anything, or carve out my own printed circuit boards (a small dream I've had since high school), I'll need something like a Dremel multitool, which is loud. With this thing going so slowly, it will have to run for hours, perhaps even a day or two to finish up. I live in an apartment, and can't subject my neighbors to that kind of punishment. I've considered options like building a sound-proof box, which has much of its own appeal, but who knows what I'll eventually decide. Either way, I don't think I really want to leave a Dremel running overnight.