Re: [CAD_CAM_EDM_DRO] Out of Estop.. BUT

> From: imranjawaid@...
> I ran EMC , X axis keeps ticking in EMC

Jon Elson reported the same thing with the newer releases of the EMC. Here is
what he said:

> First, and I noticed this with the 5-May-2000 version, also, is that the
> servo system produces strange humming sounds, only when moving,
> and that the sounds change with velocity. It sounds VERY much like
> a stepper system, with mixing of the system timer with the step pulse
> rate. If you hit just the right rate, it causes some pretty large

amplitude

> vibrations at frequencies as low as about 20-30 Hz, but most of the time
> is closer to 100 Hz, and just audible, not noticable as vibration.
> I wonder if this has something to do with the changes for stepper
> pulse generation and freqmod?
>
> Second, and more serious, is that the 'clicking' is back. I hadn't heard
> this in many months, and was not happy to see it back. It seemed to
> be VERY bad at first, then went mostly away after a while. Then,
> while testing a small program to move the machine around, it became
> very large, and caused 1/4 to 1/2" moves, causing an axis following
> error. Then, as soon as the following error dialog box appeared, the
> DACS started producing random outputs, and the machine went
> crazy, and I had to hit e-stop. There were still random voltages
> coming out of the DAC, I could see them on the LED bar graphs
> on my servo amps showing velocity command in. When I clicked
> OK on the following error dialog box, the DACs all went to zero
> as they should!
>
> Finally, although I have zero as the backlash parameter in the .ini
> file right now, I observed that all moves, both continuous jog and
> moves generated with MDI commands showed going past the
> point and then coming back. This was the behavior I saw with
> the first attempt at backlash compensation, which was later corrected.
> Has the older version gotten back into the distribution code?
>
> I am attaching my .ini file and the run script. Note that the 20-Dec-1999
> EMC works fine, and I have run it quite a bit with absolutely no
> problems, often for 8-10 hours straight. The Mar-2000 and now the
> 7-Jun-2000 versions have these hangup problems within a few minutes.

It's been a while since I've had a servo machine running the EMC, so
obviously something has gone wrong since then with the servo code. I'll check
into this next week, but for now you ought to go back to the 20-Dec version
to see if that fixes your problems. If it does, it's the EMC. If not...

Matt

P.S. I can't find 20-Dec on NIST, or linuxcnc.org, or my own hard drive.
Maybe Jon could send it to you if you don't have it.