> On Dec 2, 2018, at 6:13 AM, andy pugh <[email protected]> wrote: > > On Sun, 2 Dec 2018 at 02:13, Thaddeus Waldner <[email protected]> wrote: > >> >> Now what I really want to know is, in the case of multiple DC power >> supplies, is it advisable to connect the negative terminals of each, and >> are there any pitfalls in doing that? For example, can I connect the >> negative terminals of the 5vDC supply and the 24vDC supply to the 7i96 >> isolated input common terminal? I believe I incorrectly called the negative >> terminal the ground earlier. >> > > I think it is better, in principle, to keep the logic and power DC supplies > separated, though you need to keep careful track of how they interface > where they meet. If it is at opto-isolators then the two supplies can be > referenced entirely separately. > However, in practice, there is nearly always some component in the system > that ties them together, a drive or break-out board or something. And in > that case you probably want to do it explicitly rather than have an unknown > connection in an unknown place potentially passing quite high currents. >
The two DC rails that I need to have a common ground for are the 5v that powers the MESA card and the 24v that I am using as supply voltage for switches and relays. Some of my switches need to run on 5v, which necessitates connecting the return for each to the isolated input common terminal. Relays are perhaps the biggest source of noise on the 24v supply, which isn't an issue. > Now, is there a “home command” signal that I can connect to the enable >> signals of my ClearPath motors in order to sequence the homing routine? > > > I think you would need to configure LinuxCNC for immediate homing > (HOME_SEARCH_VELOCITY = 0) and have the servo drives control the homing HAL > pin. > This is the reverse of how you are probably imagining it. Rather than > LinuxCNC telling the motors to home, the servo drive (and some HAL logic) > needs to tell LinuxCNC "this axis is now at home" > > Do the servos output any signal to say that they have completed homing? > > It's slightly annoying that the HAL pin to home an axis is in the > (userspace layer) halui component, but this is because homing is triggered > by a queued command. > http://linuxcnc.org/docs/2.7/html/man/man1/halui.1.html > My understanding of the internal homing sequence is as follows. The motor holds position until enable is high When enable is switched high, the motor homes and ignores step/dir pulses Any time the motor completes a move it, switches a feedback high; this is intended for use as a “home completed” signal. After homing is complete, the motor will begin to respond to input pulses, during which time the feedback will keep switching; low for moving to position, high for move completed. My thinking is that I can use the existing homing settings as they are, and as such, the UI will even give me a graphical feedback of how the homing is happening, provided I set the homing velocity in the Clearpath motor equal to the LinuxCNC homing velocity. I still need to somehow control the enable signals for each axis. I need it to finish homing z before it begins to home x and y. Since the motor output is not a “homing done” signal, but rather a “move completed” signal, I need to somehow latch enable high until the system is de-homed. > Home-switches and home-to-index should be just as good as the hard-stop > homing. But probably no easier to configure. > -- > atp > "A motorcycle is a bicycle with a pandemonium attachment and is designed > for the especial use of mechanical geniuses, daredevils and lunatics." > — George Fitch, Atlanta Constitution Newspaper, 1916 > > _______________________________________________ > Emc-users mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/emc-users _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
