RE: Parameter PERIOD, Freqmod now running on rtlinux 2.0

["Eding, Bert" <b.eding-at-Nyquist.com>]

I have got freqmod and the rest running on RTLINUX 2.0,
The problem I had with MACHINE not comming out of ESTOP
had to do with the brigeport task and io.
I thought it had todo with the P-Gain as Jon thought because
there was also a following error, so I started with increasing that
value, It did not help.
When I switched to minimill TASK and IO, it worked.

Still I have some questions about IO handling,
I am planning to connect a stepper motor card this week.

The Clock and direction pulses are on D0..D5 of the
parallel port.

the ini file defines serveral other outputs, where the
INDEX is given, if this index starts at 0 for D0,
the default inifiles use:

SPINDLE_FORWARD_INDEX     =	1
SPINDLE_REVERSE_INDEX     =	0
MIST_COOLANT_INDEX        =	6
FLOOD_COOLANT_INDEX       =	7
SPINDLE_DECREASE_INDEX    =	8
SPINDLE_INCREASE_INDEX    =	9
ESTOP_WRITE_INDEX         =	10
SPINDLE_BRAKE_INDEX       =	11

This means that SPINDLE_FORWARD_INDEX is the same output
as X-Direction and SPINDLE_REVERSE_INDEX is X-Clock, isn't it?
Or are these spinbdle outputs in the ini file but not used?

Bert


-----Original Message-----
From: Jon Elson [jmelson-at-artsci.wustl.edu]
Sent: vrijdag 14 april 2000 23:12
To: Multiple recipients of list
Subject: Re: Parameter PERIOD





"Eding, Bert" wrote:

> Thanks Jon I will try that in the weekend.
> What I would like to know is,
> what is the idea behind simulating a servo motor,
> when you want to control a stepper without feedback?

You'd have to ask Fred Proctor (or, possibly Will
Shackleford) about that.  I suspect it was a way to
reuse almost all the code from the simulation version
of EMC, and tap into the routine that simulates the encoder
counts coming in to make the step pulses.

Jon