Re: STG2

[Jon Elson <jmelson-at-artsci.wustl.edu>]

JohnDRoc-at-aol.com wrote:

> I certainly have stirred things up, haven't I?  This is where I'm at....
>
> I changed the Input Scale to (-) in the .ini file, and it seems to have
> worked.  I didn't know that it could be done, until I saw a sample file on
> one of the EMC sites.  I guess it makes sense though, because the motor is
> located on the back of the machine, whereas a retrofit would most likely have
> the motor on the front.(?)  That's the good news.

I tried to tell you this, THREE times - but maybe my comments were so
long you missed this tidbit.  I said something about making sure the encoder
reads in the right direction, that an axis MUST be completely consistent with
respect to all signs on the parameters (not that they be the same, but that
all of them agree on what direction is positive).

>
> The bad news is, I've discovered the problems with EMC and the encoders doing
> funky things.  I went back to the X axis, to fine tune it, because it is
> rather jittery when I jog - it bounces instead of running smoothly.  As I was
> trying to adjust the tach and signal on the amp, the readout somehow went to
> -20.0000 and is stuck there, even when I restart.  Whenever I try to jog, it
> will now go to 0 and "freeze" there - it then gives me a following error and
> all of the axes "blip" and that's all she wrote.  I can't get a response out
> of any of the axes, and I end up having to shutdown and restart EMC (a couple
> of times I even had to restart the computer).  I don't know if there is any
> correlation between the -20.000 and the fact that I have +/-10 in the
> Max/MinLimits in the .ini file.  I should be able to override and jog past
> (?).

Yes, this sounds like something to do with limits.  Set your limits to a greater
range, at least for now.  EMC records the current position and the offset between

the absolute machine coordinates and the relative coordinate system established
with the g92 commands every time you shut it down.  You might edit the
file (I forget the name, but it can be edited in a pull down menu from the main
screen.

> Fred and Will are receptive to having me send the STG2 board up there, so I
> might go ahead and do it.  That would definitely allow them to test the
> driver and verify the issues that Jon was talking about.  In addition, does
> anyone know where I can get the Dec release?  It isn't on the NIST site,
> anymore.

Well, I have the version for the 2.0.36 kernel (Red Hat 5.2), but I don't have a
copy
of the 20-Dec-1999 EMC for the Red Hat 6.x kernels.  I'm not completely
sure 20-Dec-1999 has (supposedly) tested STG2 code.

Jon