I played with AccelStepper and for some reason it is very inefficient and the top limit is 4000steps/sec
"The fastest motor speed that can be reliably supported is about 4000 steps per second at a clock frequency of 16 MHz"
I wrote my own little stepper speed tester and I calculate the delay between steps for any given requested speed, and I can easily issue steps faster than the stepper can handle, so not sure why AccelStepper is so slow.
I have done quite a bit of testing with speeds for a Nema17 from Kysan, and it seems the top speed, regardless of microsteps, is 600RPM. with an unloaded stepper. To get that I had to drive it at the maximum current of 1.5 Amps/winding.
I also discovered from the specs for the A4988 that when using full stepping, that it only delivers 70% of the max current to the stepper (there is a nice table towards the end of the spec). So if you want to actually deliver the full 1.5Amps to the stepper you need to bump up the Vref a bit.
There is a faster stepper available here
http://www.phidgets.com/products.php?category=23&product_id=3304_0 with a top speed of 1250 RPM, but at a significantly reduced torque.
On Saturday, November 10, 2012 11:10:29 PM UTC-8, Tim Driedger wrote: