[spalm]

Progress? Hmmm. Lots of hours spent. I have learned a lot.

I’m still trying to do a cheap upgradeable design. Progress is just so slow, so many things to worry about. I did my first (and second and third) toner transfer PCB today. I took Phil’s advice and ended up using a glossy advertisement from a mattress store as the print medium. It worked so much better than the other papers I was trying. The ad had color pictures all over it, but that made no matter. It was nice and thin, so the removal was much easier also.

I changed the design from post #1. It seems to me that stepper controllers have 5 parts: the Logic (processor), the Current sense/monitor, the Chopper, the Drivers, and the FETs. I moved all of the current and chopper out of the processor and into hardware (just like the original PIC design). After a lot of thinking about it, it just seemed more predictable (and I was not too impressed with the comparators and resistor ladder in the PIC). Simple IC’s are extremely cheap, so the cost comes out as a wash. Problem is that the chip count went up, so it is fairly packed on a one sided board that is etched at home. To packed for me, over 250 holes and 45 jumpers. But this bipolar design from Mouser is still under $15 with pretty good quality 60 volt / 20 amp FETs and $2 in sense resistors.

I came up with a 4 bit resistor network D2A that lets me have 51 different voltage levels for the steps in a sine wave. By pulling up combinations of resistors, and pulling down others, and letting others float. I attached a spread sheet of my playing around with this. It was kind of fun. It also has what a hex value waveform and a ten step waveform look like for comparison.

I have to come up with a better way to route or shrink this thing, or just give up.

Would like some critisism.
Steve

Edit: Just a note: I only see this as a 3 or 4 amp, and about a 45 volt controller. I am not trying to enter into the high drive stuff.