Re: E-Stop switch

Hi PJ,
Mach1 is a general purpose control software package that does not
really know anything about the hardware that it is driving (very
loosely coupled system). With such a package, it is best to have the
E-STOP switch initiate power-down without relying on software. At a
minimum, the E-STOP should cut DC power to the motors(i.e. drives)
(whether directly through the E-STOP switch, or via the use of a relay
(contactor)). Lower DC voltage/amp systems can probably be done
directly through the E-STOP switch (assuming the switch is rated for
the DC current). Larger Voltages/amps may need to have the power
disconnected by way of a relay(contactor). You want to disconnect at
least the DC, because if you just disconnect the AC, your power
supply may be able to continue to drive the motors after the AC has
been cut due to charges in caps. Or the E-STOP switch could be used
as part of a circuit that energizes a relay for normal run mode, and
upon entering E-STOP state, the relay circuit is un-energized, thus
opening up the relay(contactor) path of DC power to the drive. It is
also good to let the software know this has happened so it can stop
issuing pulses. If looser safety requirements are OK, the E-STOP can
just signal Mach1, which then would trigger an output that would
cause the above mentioned relay to open (and cause it to stop issuing
pulses, though not a recommended method).

With all of that said, I have read in this news group that E-STOP
should ALWAYS cut power to the drives(motors) IMMEDIATELY. This is
not ALWAYS the case. Especially in BIG machines (which aren't really
the focus of this group, but the idea should be understood
regardles.) In a BIG machine with THK type linear ways, and
ballscrews drives, AND tightly coupled software/hardware you want to
delay cutting power to the motors for a short time, yet signal the
controller that an E-STOP situation has occured. Here's why:
Imagine the above mentioned BIG machine's table loaded with 500
pounds (or more) of mass. Add to that the weight of the table
itself. Now start that table moving at a high rapid rate. Now hit
the E-STOP switch where you cut motor power immediately and what you
get it a free-wheeling heavy mass. It will not stop quickly because
of the low friction of the linear way, and little backdrive of the
screws and motors, and is going to slam right into whatever it is
that is in the way, be that the milling head, or your hand, etc.
Instead, in a tightly coupled software/hardware system, the control
will get the E-STOP signal and immediately place the system into a
rapid decel/stop. It can stay in this mode for upto half a second (by
which time the table has been stopped are greatly slowed down),
after which point, it will then release power to the drives and let
the axes 'free-wheel'. Such a system will also have 'fail-safe/watch-
dog' hardware built in such that if the software insn't responding
correctly it will shut down power to the system. In this worst case
you would get free-wheeling as mentioned above, but these tightly
coupled sytems are generally not running under Windows/DOS or PC
hardware, and are alot more robust than PCs and Windows/DOS so this
situation isn't generally going to occur. This isn't hypothetical,
this IS how some BIG machines operate.

Jeff
http://www.xylotex.com

--- In CAD_CAM_EDM_DRO@yahoogroups.com, "P. J. Hicks"
<hickspj467@a...> wrote:

> Hi,
>
> When the proper input line (pin) is triggered Mach1 initiates an E-

Stop condition.

>
> a. Is this (software triggered) the norm? I.e. shouldn't an E-Stop

be a physical/mechanical/electrical sort of thing?

>
> b. I have read mention of removing all power when E-Stop is

activated and seen pictures of the Big Red Switch on several
controller cabinets. Does this mean complete removal of all power to
the system both AC primary and DC motor power or one or the other?

>
> Progress is slow... and sometimes in the opposite direction!
> PJH
>
> [Non-text portions of this message have been removed]