I am controlling the speed of a single phase, 120vac, centrifugal blower with a particle photon microcontroller, optocoupler, triac, and snubber circuit. I'm currently using phase control, but intend to try integral cycle/burst fire control next.
I am able to control the speed, and am measuring the output with a hot wire anemometer. Full on, the fan is blowing 28 cfm. I can barely reduce this down to 27, and then there is this large jump at the next microcontroller output level down to 12 cfm, from which I can step it accurately down 1 cfm at a time to 0. This meets my purposes, as I need around 27, 10, 5, 3, and 1 cfm at different times in my project.
However, the output level resulting in 12 cfm will, if left running for 5 minutes, eventually speed the fan up to the higher step of 27 cfm.
What is happening? Why is there this big step from 12 to 27, and why if slowed down to the lower step would the fan eventually ramp up to the higher step when run long enough?
I'm currently running a temperature test to make sure I'm not overheating the fan motor with speed control, but I don't have full results in yet.
My second question is about phase control vs. integral cycle control. I've seen competing information on both being better for motor speed control. Any advice on which will give better results?