On Friday, 29 April 2022 09:01:51 EDT Thaddeus Waldner wrote: > Gene, can you share a link to the motor you are using? > > Thanks > Sure:
<https://www.ebay.com/itm/ 172690305077hash=item283524f035:g:jEkAAOSw-3FZI9X1> But he's getting fat on the shipping. Also you can see the src of my excedrin headache numbered somewhere in 7 digits, by looking at the table silk screened on the driver. It was swiped from a 2 phase std stepper motor, so the pulses/rev column needs multiplied by 1.5 as its NOT a 2 phase stepper with a 1.8 degree full step, but a 3 phase with a 1.2 degree full step. Not 200 steps per rev, but 300. So if that printed table was re-composed for a 3 phase motor, it would look like this for the top half of that table. [pulses/rev][ SW1 ][ SW2 ][ SW3 ][ SW4 ][ DIV ] [ 300 ][ off ][ off ][ off ][ off ][ 1 ] [ 600 ][ on ][ off ][ off ][ off ][ 2 ] [ 1200 ][ off ][ on ][ off ][ off ][ 4 ] [ 2400 ][ on ][ on ][ off ][ off ][ 8 ] [ 4800 ][ off ][ off ][ on ][ off ][ 16 ] [ 9600 ][ on ][ off ][ on ][ off ][ 32 ] [ 19200 ][ off ][ on ][ on ][ off ][ 64 ] [ 38400 ][ on ][ on ][ on ][ off ][128 ] where DIV is micro-steps There is another worrysome item, the green miniature motor connector isn't really adequate to reliably carry the current when running continuously near its top speed. I needed to extend the cable a bit and bought some more of that green connector, but they run hot. Hot enough the plug can fall apart with only the weight of another foot of cable pulling on it, so that joint now has teeny grey wire nuts on it. But there's another twin glued to the alu side plate of the new driver box I built to contain bigger PSU's, so its cooled by conduction. I can feel a slight heating but its not falling apart. These green things come in 3 sizes, and based on my experience, would not recomend the smallest version for motor power. For Logic they are fine. One other comment, compared to 2 phase steppers, these are like casper the ghost. I have 2 of the higher powered 2NM and 3NM on my Sheldon lathe, and its so silent you can't hear it moving if its below 20 ipm. Part of the secret to these 3 phase motors is that the motor has an encoder, feeding back only to this driver, and the error modulates the driver current, if its not working hard, the motor runs at ambient temps. So you don't have to shut them off to slow the disk in the power meter. This driver also has an alarm shutdown that can be wired into the e-stop so if it hits something and loses a step, F2 is toggled off & LCNC stops dead in its tracks. > > On Apr 29, 2022, at 3:46 AM, gene heskett <ghesk...@shentel.net> > > wrote:> > > On Thursday, 28 April 2022 21:17:20 EDT gene heskett wrote: > >>> On Thursday, 28 April 2022 19:35:44 EDT andy pugh wrote: > >>> On Fri, 29 Apr 2022 at 00:20, gene heskett <ghesk...@shentel.net> > >> > >> wrote: > >>>>> Surely you already know that? It's 360 x 100 x SCALE ? > >>>> > >>>> Sure Andy. But what is the scale? > >>> > >>> 200 * 16 * 40 * (60/53) / 360 > >> > >> But these motors are not 200, but 300, they are 3 phase, 1.2 degrees > >> a > >> full step. 300 steps per rev for full step. > >> So by purely mechanical means, that would then be: > >> 300 * 16 * 40 * [60/53] / 360, or 603.7735849 according to my TI-36X > >> Pro. But thats about 5x what seems to be pretty close @ 125.xxxxx. > >> Something, someplace is lying like a rug. But where? You got the hal > >> file now, is it wrong? > > > > Two problems to fix before its right. > > > > 1. These drivers have two microstep modes, digital and apparently > > powers of 10. Only one switch is on, sw3, which claims it makes a > > full turn of the motor in 6400 steps. Just one problem. A 300 full > > step per turn motor cannot be made to equal 6400 with any integer > > multiplier. At /16, its not 6400, but 4800 steps per turn. A value > > that's off either by .75, or 1.333333333333 depending on > > interchanging the numbers. If it is truely actually 6400 > > microsteps/turn, what the hell kind of math is it useing? Obviously > > NOT a power of 2. I think these little magic boxes are miss- marked. > > Obviously whoever drew up that silk screen was thinking in terms of > > a 200 full steps per turn motor, and the 3 phase models are 300, not > > 200. So that means 300*16 is 4800 steps per turn of the motor shaft. > > No way in hell can I make the math work using thier silk screened > > figures. > > > > So that's problem #1. And the correct answer can only be found if #2 > > is a 1/1 answer. > > > > [edited] > > > >> I just found the stepgen drawings in the docs and I think my hal > >> file > >> is wrong, the values presented by the position.fb pins are not in > >> the > >> same units as .count's. position.fb has been scaled. > > > > 2. The $64k question then, since the step itself is not available to > > hal for counting, is the stepgen a direct translator?, issueing 1 > > full step per count it reports on the .count pin? > > (yes/no) > > If no, what is the ratio? > > > >> And this needs counts. I'll fix that when I get it back together. > >> Then > >> maybe it will make sense. > >> currently stepgen3.counts is -173640, > >> while stepgen3.position.fb is 1380.7 > > > > Fixed already, but not yet exersized for truthfullness. A sneakernet > > mistake. :o( Cheers, Gene Heskett. -- "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author, 1940) If we desire respect for the law, we must first make the law respectable. - Louis D. Brandeis _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
