Thanks guys. I need to understand a lot more about the LCNC code base to understand what Andy finds very interesting on that link! What are the key features that help THC?
Its great to hear other people are looking at PID control for THC and I'm keen to see what you come up with. John, I must confess that a code snippet Andy posted a couple of months ago got me thinking hard about PID based THC. So here are some discussion points to think about:
1. My thinking was PID based THC was suited to a velocity controlled axis (and even comments the PID code suggests the same). Position is meaningless for THC, we only care about volts.
2. Is the servo thread the right thread for THC when published research worked on a 40ms cycle time?
3. If the THC operated on its own slower thread, another component in the servo thread could calculate average, min, max and moving average data that could be presented to the THC thread. (there would be say 40 readings per data set)
4. Can distance travelled on the xy plane be measured easily? If so, we could extend the averaging technique so we can sample torch voltage for the first say 1/2" of travel to automatically set torch height voltage?
5. Can the mods add a dedicated plasma section? There seems to be a bit of interest right now and the forum nuggets are fragmented and hard to find.
Sorry for the brain dump but RickG did say experimental
