Re: fickle y-axis

Thanks Jeff,

Very clearly explained.

A1-B-A2-B-A3-B-A4-B-A5-B
------X--------------

A1-B-A2-B-A3-B-A4-B-A5-B
-----------X----------

A1-B-A2-B-A3-B-A4-B-A5-B
----------------X------

Momentum causes rotor to bypass B and fly all the way to the next A
coil. Just a little bit more for complete understanding. Starting
from A2, How would starting a cycle cause the rotor X to go to A1
instead?

Thanks

Peter









--- In CAD_CAM_EDM_DRO@yahoogroups.com, "jeffalanp" <xylotex@h...>
wrote:

> Hi Peter,
> Why a stepper can turn with only one phase hooked up.
> Have you ever seen those old cartoons where two guys (A & B) are
> on the railroad tracks with one of those hand pump types of cars?
> The two take turns pushing down on the lever, first A pushes down
> (while B's handle goes up), then the B pushes down (while A's

handle

> goes up). The two phase of your stepper motor are kind of like

that

> (in full step mode). Now if one if the people (B for example),

isn't

> there, you'll have half the power, but as long as A's handle wasn't
> all of the way down, he can start some motion by pushing down on

his

> handle. If he pushes enough so that his handle, after reaching the
> bottom, will cycle all of the way to the top, he can start a full
> cycle by pushing down again. As long as each down stroke provides
> enough power so that momentum will carry the handle back to the

top,

> he can continue pushing, and moving the car. Your stepper can move
> similarly, as long as there is enough momentum to get the rotor to
> the next cycle.
>
> The direction your stepper takes with only a single phase attached
> will be pretty much arbitrary though. (If you've ever seen an old
> foot-powered treddle sewing machine, the pricnciple is also
> similar).
>
> I hope that helps a little.
>
> Jeff
> http://www.xylotex.com
>
>
> --- In CAD_CAM_EDM_DRO@yahoogroups.com, "forumtvm" <forumtvm@y...>
> wrote:
> > Hi,
> >
> > Did some detective work - tried swapping motor cables around.
> Turned
> > out that the problem followed the "y-axis" cable. One of the wire
> was
> > loose, in fact, eventually broked loose from the soldered pin.
> > Changing to a new cable solved the problem.
> >
> > However, I am puzzled as to how a bad cable can lead to erratic
> > behaviour, that is when a movement in the positive direction is
> > commanded, sometimes it move in the positive direction, and
> sometimes
> > in the negative. How would a motor behave if, say phase A wires

are

> > OK but phase B connection is bad/broken. Would this cause the

motor

> > to move in the opposite direction to that commanded?
> >
> > Am more interested in understanding the problem than in getting

the

> > problem solved. Hope someone smarter than me will explain.
> >
> > Puzzled.
> >
> >
> >
> > --- In CAD_CAM_EDM_DRO@yahoogroups.com, "forumtvm"

<forumtvm@y...>

> > wrote:
> > > Hi,
> > >
> > > Using DeskWinNC, Camtronics 2A driver, 33VDC power supply, I

was

> > > getting only 16-18 ipm previously.
> > >
> > > Happy to report that after changing to a Pentium 4 and DeskCNC
> > (with
> > > the same old controller board) I managed to squeeze 35 ipm from
> the
> > x-
> > > axis before stalling. For the z-axis, I could get 17 ipm.
> > >
> > > However, the y-axis seem to be acting up. At one stage, I could
> > only
> > > get it to make noise without any movement. After powering off,
> > trying
> > > another day, it was OK again. Now the y-axis can't seem to make
> up
> > > its mind whether to move forward or backward. In MDI
> > > G91 G01 Y0.1 F10
> > > it is supposed to move forward (same direction) each time it is
> > > activated. But what is happening is, sometimes it move forward,
> > other
> > > times it move backward.
> > >
> > > DeskCNC uses a serial cable from PC to controller board, and
> > parallel
> > > cable from controller board to driver. So I try to see if I can
> > > isolate the problem. I switched to TurboCNC which uses a

parallel

> > > cable direct from PC to driver. Same problem.
> > >
> > > What could be the possible cause? Where should I start looking

to

> > > solve the problem?
> > >
> > >
> > > Looks like a lot more work to get a consistent dependable

system

> > > going.
> > > Thanks
> > >
> > > Peter