I reread the first post. I think you should scrap the dual uni/bi concept, stick with bi polars.
I was looking through some old files when I was testing discrete components for chopping and current limiting. Below is an idea that started off as a suggestion by Phil at PMinmo, I may have adapted it a bit. I am still looking for others, but wanted to show the concept for using a shift register to set a fixed off time. The clock can come from the processor, and can have an affect on the off time as well by changing the speed. Also, choosing which output on the SR you use changes the off time as well, but setting it to output 8 means you get an 8 clock count for time off.
The inputs to the AND's are from a Prop [noparse][[/noparse]not shown]. The PWM input was from a prop, really just a clock, not pwm.
There are 530 fets shown, I really only use 540's for most applications with a 1/4" aluminum heat sink bolted to the fets. I find that when you are experimenting and designing stepper boards, the fets are the first things to test for when there is a problem, and often a shorted fet was the problem. I noticed that when in breadboard and test modes, the fets need to be replaced a lot until the design is good, so TO220's are a good way to start versus SMT fets.
If my memory is correct, I think I built this design and ran it for some time on a uni motor setup. The uni's were not enough power, I switched this machine to bi polar and it made a big difference.
Comments
I was looking through some old files when I was testing discrete components for chopping and current limiting. Below is an idea that started off as a suggestion by Phil at PMinmo, I may have adapted it a bit. I am still looking for others, but wanted to show the concept for using a shift register to set a fixed off time. The clock can come from the processor, and can have an affect on the off time as well by changing the speed. Also, choosing which output on the SR you use changes the off time as well, but setting it to output 8 means you get an 8 clock count for time off.
The inputs to the AND's are from a Prop [noparse][[/noparse]not shown]. The PWM input was from a prop, really just a clock, not pwm.
There are 530 fets shown, I really only use 540's for most applications with a 1/4" aluminum heat sink bolted to the fets. I find that when you are experimenting and designing stepper boards, the fets are the first things to test for when there is a problem, and often a shorted fet was the problem. I noticed that when in breadboard and test modes, the fets need to be replaced a lot until the design is good, so TO220's are a good way to start versus SMT fets.
If my memory is correct, I think I built this design and ran it for some time on a uni motor setup. The uni's were not enough power, I switched this machine to bi polar and it made a big difference.
Post Edited (TChapman) : 1/19/2008 6:47:55 PM GMT