richard harris wrote: > I am running Elson's servo system and > EMC 2.2.6. > > Over the last few days I have run > several hundred parts single point threading a M22x1.5 threads > without a single bad thread. Yesterday I switched to cutting a > M20x2.5 and suddenly on ever other part the lathe will rapid during a > cut. I had a similar issues months back and upgraded EMC to 2.2.6 > and the issue went away. The lathe will go to the start the thread > it will do a brief pause for the index pulse and shoot off at G0 > through the part. I went back to my M22 code and ran 30 parts > without fail, I even reduced the spindle speed to what I am cutting > the M20 at thinking I might get more noise from the VFD at the lower > end of the rpm band and no effect. I also increased the number of > passes on the M22 program to replicate the M20 program and no > failures. > > Switched back to the M22 program and > the second part experienced a rapid on the spring pass. > Any ideas what could be the cause? Is > there a way to set EMC to never attempt a threading pass that is > faster than the G0 velocity? > VERY strange! But, I have seen some behavior like that before. Both old revisions of software and hardware can cause it. A problem in the driver on a released version of EMC from the CNC Workshop in 2007 would go crazy like that if the spindle was allowed to run for enough time to count up 16 million encoder counts, then the CNC program was started. The 24-bit overflow was not handled properly, and the Z axis tried to "Catch up" to the thread which was now many feet away. My experience with this problem indicated it had nothing to do with the thread pitch, but how long, and how fast, the spindle ran between threading passes.
Any version of EMC later than July 2007 or thereabouts should not have this problem in software. My log shows your UPC board has the 5/2/2007 firmware, which is the current version for that rev. level of the board. So, that should be OK, too. I'm pretty sure EMC will NOT exceed G0 velocity. But, the threading code will go up to that speed to sync up to a thread. I'm just wondering if there is any quirk in your M20 program that makes the conditions around the threading pass different than the M22 program. Are these programs descendants of the sample threading.ngc program, or were they created from scratch or by a CAM program? As Chris says, you may need to instrument the situation and then send us a screen shot of the halscope when it fouls up. I use a pretty similar setup on my minimill to do rigid tapping, and it seems to run quite reliably there. It doesn't actually need to do a spindle sync for rigid tapping, but I think it does. I also run demos doing a single-point thread on that machine, using it like a lathe, and making multiple passes. I did indeed have these problems at the 2007 NAMES show, and Chris, Jeff, John and I bashed it into submission at the 2007 CNC Workshop. Looking back at the release history, that should have gone into the complete public distro. about 2.1.7, so you shouldn't have such a problem. Kirk Wallace is also using a UPC board on a lathe, but I don't know if he is doing a lot of threading. Maybe you should send me your threading program and I will try it here on the minimill. Is your machine natively defined in mm units or inch units? It "shouldn't" make a difference, but I'm pulling at straws, here. Jon ------------------------------------------------------------------------------ Apps built with the Adobe(R) Flex(R) framework and Flex Builder(TM) are powering Web 2.0 with engaging, cross-platform capabilities. Quickly and easily build your RIAs with Flex Builder, the Eclipse(TM)based development software that enables intelligent coding and step-through debugging. Download the free 60 day trial. http://p.sf.net/sfu/www-adobe-com _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
