On Thu, 2008-04-17 at 19:35 -0400, John Kasunich wrote: ... snip > That seems rather low, but making accurate resistance measurements under > an ohm or two with a regular meter is an exercise in futility anyway. I > don't know your general level of electronics expertise, nor the > equipment that you have handy, so some of this be be out of place. > Anyway, you might want to try a 4-wire resistance measurement. If you > have a bench power supply with constant current mode, put a known > current (an amp or so) through a winding and measure the voltage drop. > Ohms law gives you the resistance, and as long as you make the voltage > measurements right at the motor terminals lead resistance doesn't matter.
I have a meter, scope and bench supply. One problem with the supply is that it indicates amperage on a selector dial and is very inaccurate. Could I put the supply, 1 Ohm resistor, and winding in series? I could measure the resistor voltage to get current and measure the winding for voltage drop. > One other note - tell us if you are measuring from one end to the other, > or from one end to the center tap. I measured from center tap to each winding end. From end to end, I got 1.4 Ohms (with 1.0 for the leads only). > > The 8 Amp figure comes from the driver circuit > > board where the current limit potentiometer indicates an 8 Amp max. Is > > any of this useful? > > Is the pot turned all the way up? The motor might be 4 or 6 or 7A. I don't know where it is set. I didn't want to touch it if I didn't know how to set it, but I suppose measuring the resistance from the wiper to each end would not hurt. > If the motor is indeed a half-ohm or less and can deliver enough torque > at 7 amps or less, a Gecko would probably run it quite well - a lot > smoother and probably faster than the existing drive. > > > I scoped the driver board inputs again and corrected my signal diagram > > on the schematic. The gray traces are the result of the coil input and > > it's inhibit. I confirmed this with scoping the far side of the input > > optocoupler. I played with the axis speed and noticed the inhibit gets > > smaller and goes away around half speed. The trace leads me to believe > > the designers were trying to soften the low speed steps, with a brief > > half step, so that the motor would not overshoot and degrade the surface > > finish. If this is the case and not for some sort of electrical reason, > > I can switch over to EMC2 right away, and not worry about blowing out a > > board. Of course Stepgen doesn't have this inhibit signal available, so > > I may need to find another way, maybe using selective and adjustable > > delays in HAL. If step overshoot is a common stepper problem, is there a > > common way to fix it? > > > > So far we're assuming that the existing drive has a current limiter > (actually a pair of them) on the lines going to the center taps. The > presence of a "current" pot sure points that way. > > The inhibit is something else - it prevents both ends of a winding from > being connected to ground at the same time. I think that happens on purpose, because two windings are on for each of four steps. > That would be bad. Whether > it is actually needed for that reason, and how long it needs to be, is > hard to tell without knowing a lot more about the drive. At very low speed, the inhibit pulse starts at about 25% of the input pulse on time, or, the resulting on pulse is 75% shorter than the raw input pulse. This % decreases, as speed increases, until it disappears at mid speed (50 IPM). I am tending towards ignoring the inhibit and feed straight Stepgen type 8 into the Drive and see how it works. > For example, > the 2N4126 and 2N4124 transistors on your schematic are puny little > small-signal transistors. Probably part of the drive circuits for the > main switches. I just haven't gotten that far yet. The 2N4124/6's drive what appear to be a pair of transistors wired as a Darlington. I should have it on the schematic soon, but each winding end goes to a 25 Amp and 3 Amp diode (four sets). The 3 Amp goes up to the supply for freewheeling (?). The 25 Amp goes to a Darlington pair, then to ground. Each common (center tap, two) goes to a fuse (I should check the Amp rating shouldn't I?) and to a similar Darlington pair and the supply. The Darlington pair transistors are soldered as a module and face to face, so I can't see what they are without a bit of disassembly. I wouldn't want to take one apart unless I need to. It would be nice to get a part number and specs on them though. > It would be more interesting to see what the main power > transistors are. MOSFETs are fast, bipolars aren't. The latter are > more likely to need an inhibit. > > Hint: When reverse engineering any power circuit, start with the power > components. The big parts will define the basic topology of the > circuit, which usually one of a few standard configurations, and is also > most of what you need to know. There are a million ways of doing the > little stuff like transistor drive, every drive is different and for the > most part it doesn't matter, as long as it works. > > If you could trace from the motor terminals to the power transistors and > from the power transistors to ground and/or the main power supply, we > would be able to tell what we're dealing with. (See above) > Good background reading (if you haven't read it already): > http://www.geckodrive.com/photos/Step_motor_basics.pdf > In my opinion, reading and understanding that paper should be step one > for anyone working on a stepper machine. > > Regards, > > John Kasunich I read the Gecko "Basics" and appreciate the resource, but I need allot more information if I am to become dangerous. I also appreciate your and the lists help. So far, I have been nipping at the margins on the Shizuoka, but tonight I ripped out the Bandit. Now I am at the point of no return. -- Kirk Wallace (California, USA http://www.wallacecompany.com/machine_shop/ Hardinge HNC lathe, Bridgeport mill conversion, doing XY now, Zubal lathe conversion pending Craftsman AA 109 restoration Shizuoka ST-N/Bandit CNC) ------------------------------------------------------------------------- This SF.net email is sponsored by the 2008 JavaOne(SM) Conference Don't miss this year's exciting event. There's still time to save $100. Use priority code J8TL2D2. http://ad.doubleclick.net/clk;198757673;13503038;p?http://java.sun.com/javaone _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
