Hello Guest it is March 29, 2024, 01:42:31 AM

Author Topic: Chasing lost steps, now moves verrrrryyyy slooooowlyyyyyy try4  (Read 6004 times)

0 Members and 1 Guest are viewing this topic.

Offline bradD

*
  •  15 15
    • View Profile
Chasing lost steps, now moves verrrrryyyy slooooowlyyyyyy try4
« on: January 12, 2014, 03:36:27 PM »
Reading through the forum in an effort to fix a problem with lost steps in my X and Y axes and probably my Z axis as well, I came upon the Windows XP Professional System Optimization Guide which I chose to follow. Now the machine moves at a snail's pace. Really a snail's pace. It's that slow. Rapid traverse is minutes per inch, not IPM. Even so, my slave X axis misses a lot of steps and racks the gantry.

When I run Driver Test, I now get a nice straight line at about 40 microseconds between pulses to about the 200 mark, the the plot goes straight up off the scale for the duration of the test. The test message says my system is excellent, which is just not so. Prior to that, I was getting pulse variations at regular intervals.

Before I started changing things, I had a fairly reliable drift to +Y in the Y axis, and a problem in the X axis that would take the gantry out of square.

My machine is a CRP4896 gantry router w/ NEMA 34 motors and CRP drive electronics. The A axis is slaved to the X axis. The computer is a Dell Optiplex 755 1.6 GHz running XP Pro. Logitech wireless keyboard and mouse, and wired Ethernet to a wireless repeater. I recently added the Rosewill PCI parallel board, which was working normally though still losing steps prior to my following the Optimization Guide.

So now my system is all honked up, and I would very much appreciate some help.

I have attached my XML file for your convenience.

Offline Hood

*
  •  25,835 25,835
  • Carnoustie, Scotland
    • View Profile
Re: Chasing lost steps, now moves verrrrryyyy slooooowlyyyyyy try4
« Reply #1 on: January 12, 2014, 05:41:01 PM »
xml seems to run fine here, you have it in step mode so it will only move 0.1inch at a time when jogging.
On the diagnostics page what does the Frequency say? Also Time in Int?
Hood

Offline bradD

*
  •  15 15
    • View Profile
Re: Chasing lost steps, now moves verrrrryyyy slooooowlyyyyyy try4
« Reply #2 on: January 12, 2014, 07:54:41 PM »
Pulse freq stays pretty steady at 96. mostly 95-97, but goes to 1 periodically.

Time in Int was average 8.4 in e-stop, 9.7 with e-stop off, occasionally in the low 10's, but I saw some 13's and a 16.

Offline bradD

*
  •  15 15
    • View Profile
Re: Chasing lost steps, now moves verrrrryyyy slooooowlyyyyyy try4
« Reply #3 on: January 12, 2014, 08:07:26 PM »
I forgot to mention earlier, but I have the "Abnormal Condition" light lit, and it says "One or more axis enabled or disabled."

Offline Hood

*
  •  25,835 25,835
  • Carnoustie, Scotland
    • View Profile
Re: Chasing lost steps, now moves verrrrryyyy slooooowlyyyyyy try4
« Reply #4 on: January 13, 2014, 03:00:04 AM »
Your computer is having serious problems with the driver, the pulse should be very close to the kernel which in your case I think was 25KHz, so it should be 25,000 for the frequency. You only have 95 so that is why things are really slow.
Dont really know what to suggest other than possibly deleting the driver and then installing  it again by running the driver test. You may have to try and reverse some of the things you did in the optimisation steps to see if that helps but as you were having problems before you may just get back to where you were.
 Some computers are just not suitable for Mach via the parallel port and yours may be one, so you may have to try another computer or get an external controller instead of using the parallel port.
Hood

Offline bradD

*
  •  15 15
    • View Profile
Re: Chasing lost steps, now moves verrrrryyyy slooooowlyyyyyy try4
« Reply #5 on: January 13, 2014, 11:51:28 AM »
I really appreciate the input, Hood.

I strongly suspect that the new parallel port board might be at fault, so I'll try that today, too. Being back where I started will be better than what I have now. I'll report back.

Thanks again,
Brad

Offline Hood

*
  •  25,835 25,835
  • Carnoustie, Scotland
    • View Profile
Re: Chasing lost steps, now moves verrrrryyyy slooooowlyyyyyy try4
« Reply #6 on: January 13, 2014, 01:22:00 PM »
It is definitely more than just the port as your frequency is only showing as 95.
It wouldnt even matter if you had no parallel port installed as the driver is software and that is what you are seeing with the frequency, ie how close it is to the selected kernel.
Hood

Offline bradD

*
  •  15 15
    • View Profile
Re: Chasing lost steps, now moves verrrrryyyy slooooowlyyyyyy try4
« Reply #7 on: January 14, 2014, 06:56:28 PM »
Had to RTFM to uninstall/reinstall the driver. Tried that a couple of times to no avail. Finally gave up and pulled the parallel port card, did a Windows re-install and downloaded a fresh copy of Mach3. At least I'm back to where I started.

After all that, I ran the driver test. See attached. This is what I had before. It looks like a picket fence. Is that an ok level of pulse variation? On the good side, the pulse frequency looks much better on the diagnostic page, 23-24kHz.

What do you think, Hood?

Offline ger21

*
  • *
  •  6,295 6,295
    • View Profile
    • The CNC Woodworker
Re: Chasing lost steps, now moves verrrrryyyy slooooowlyyyyyy try4
« Reply #8 on: January 14, 2014, 06:58:55 PM »
Still no good. Try changing the PC from ACPI to Standard PC. You can usually do this in the bios, but it may require re-installing Windows.
Gerry

2010 Screenset
http://www.thecncwoodworker.com/2010.html

JointCAM Dovetail and Box Joint software
http://www.g-forcecnc.com/jointcam.html

Offline bradD

*
  •  15 15
    • View Profile
Re: Chasing lost steps, now moves verrrrryyyy slooooowlyyyyyy try4
« Reply #9 on: January 15, 2014, 01:05:45 PM »
Hi Gerry, thanks for the response. I changed to Standard PC using Step 9 of the XP Optimization Guide abd got back to where I was when I first posted to the forum.

The attached screenshot shows the driver test after. Once the trace goes off the scale, you don't see it again before the test stops.

I'm thinking I should give up on this computer unless you have somethiing else up your sleeve.