Jon Elson wrote:
This is, in general, a bad idea to have a rigid tandem axis.
>
Define "rigid". Even a 1-foot steel I beam has some flex...
Depending on the size and material there is always some flex in the system.
I have a dual ball srews on both sides of my X-axis, some quick details on
the system:
each sides has a 6.4 Nm stepper motor with a 16mm ball screw with 5mm pitch.
Each sides has 2 x 25mm linear guide carriages mounted 30 cm apart, these
are mounted on a 15mm aluminium plate (6082 T651), 70 cm tall with on the
top 2 x 80x80mm 'Bosch' style aluminum beams (heavy duty type).
Both axes are mounted 1.2 m apart from each other.
If I remove the second ball screw there is enough flex between both sides
that the other side will move about 2 cm in both directions without having
to apply too much force.
Using one motor with belt drive could have been possible but then there is
still some elasticity in the belt drive and when building the machine I had
my doubts if one motor could deliver enough power.
I have combines home/limit switches on one side and with some standard
combinatorial logic I make sure that the motor is stopped as soon as the
home is reached (and the axis is moving towards the home/limit switch.
Only when both axes are homed, the home signal will be sent to EMC2.
As soon as both axes are homed, EMC will back off from the home switch and
the direction signal allows both motors to move again.
In this way both X-axes are homed on a 90 degrees angle from the Y-axis. No
steps are lost so there is no way the axes will become miss-aligned. Only
reason for miss-alignment could be during start-up of the motor drivers
where both axes may move to a different full-step location. So in theory
there is a max. 0.03 mm difference in the position of both sides of the X
axis during startup.
There is always enough flex in the system to allow for a small difference in
position between both sides before becoming a problem.
Ray Henry wrote:
> I'm thinking that it might be possible to measure the power to each of
> the two motors and work from that as a sort of "squareness indicator."
>
Yes, this may work for homing. but as soon as you start milling, the force
on both motors differs depending of the location of the mill between the two
axes.
Or as John suggests reduce the gain to the second motor low enough so
> that it is moved when the first is homed, assume that it is also home
> and increase it's power to match the primary motor. After all, if they
> are both rigidly connected to the beam, what good does homing a second
> motor do.
>
If the system was really rigid you'd only need one motor (see above).
If servo motors are used then yes, there should be a way to prevent both
axes from moving at a too different movement.
I guess that using servo motors the position of the motor should be acurate
at any given time, so with EMC driving the position of the servo motors both
axes should be at the same location (+/- some accuracy of the system)
So I think that the system I described above for stepper motors could also
work with a servo drive - but I never tested this myself.
John Kasunich wrote:
One approach is to consider the power-up state as "acceptable" meaning
> "since the motors were free-wheeling a minute ago, there can't be too
> much stress in the machine".
That is what I tend to think. But I am not sure what a servo system (with
relative encoders) might do. If there is the possibility for a servo drive
to move at random during startup then this could harm the system.
But this should, IMHO, not happen because this could also mean an axis could
crash into a limit switch (we are talking about random movement that is not
related to any EMC2 request - so there is no way limit switches are
checked).
Jon Elson wrote:
> I was hoping to find a solution that could easily be implemented
> in a few HAL components.
>
For a stepper drive this is definitely possible, unfortunately I have no HAL
experience so I did this using a few standard 74lsxx logic circuits which
was the fastest way for me to get the system running.
Lie, is your system using stepper drivers or 'real' servo drives with
encoders?
For stepper drives I could draw up the schematics for this (using standard
logic circuits) quite easily.
Regards,
Rob
-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users
