On Tue, 27 Dec 2016 09:18:12 +0700 TJoseph Powderly <tjt...@gmail.com> wrote:
> Nicklas > are you using theses abbreviation meanings? > .u = umschaltung or Voltage > .i = current > .t = time Yes, .u=voltage, .i=current, .t=time > > re: edm.u.Sparc , is this a flag meaning the spark generator is active, > or what? edm.u.Sparc is to set at which voltage level sparc extidinguising turn off generator, this is done by hardware comparator. > re: edm.t.On & Off , are these integer microseconds or floats? I think float SI-units is a good choice. > re: edm.u.Ionization, is this a flag meaning ionization occured? No it set at which voltage level ionization is detected and edm.t.on start to cunt, maybe this should be skipped? Or a parameter instead? > re: emd.u.Seat, is this the discharge voltage, is this also the servo > target? edm.u.Seat is the measured discharge voltage. > re: edm.u.Open, is this the ignition voltage ( usually less that 3 > amperes, and for example AGie 60L is 100 150 200 270 selectable )? Yes edm.u.Open is the ignition voltage. > or the actual cutting supply no-load voltage (neccesarily > discharge > voltage and often 60-120volts, 115 in my system) Is it necessary to set cutting supply voltage? > re: edm.i , i think this is current, Yes it is measured current. But there is also a need to adjust maximum current which I forgot. > when I talked about ipkval it refered to the maximum (peak) current > available is a dead short I think I forgot this one. > in a normal discharge, the reactance of the spark gap acts sort of > like a resistor, the Actual current will be less > the joules delivered is (ontime-ionizationTime ) * (dischargeVoltage) > * (actualCurrent ) > > re: edm.ionizationCount .sparcCnt these are very fast changing data, > too fast for threads, > but outside hardware (i used PICs) can report running averages > these data can be used to evaluate process quality and can lead to > adjusting the offtime , flush, cut duration of jumping strategy Yes they are to fast for threads and counted in EDM power supply. Then sent periodically number of counts since last time sent I think is a good choice and then they could be averaged inside linuxcnc. > re: old mcodes, they are just models, they can be changed easily, likely > individual codes for individual parameters is better > the M110 can be moved and can be broken into individual set_on & set_off > the M112 sets the supplied peak current, and since my interface was > binary, had special code for that > the M199 was a position limit number, a position that constrained the > Hal motion > ( i did not use linuxcnc gcodes tho i operated inside linuxcnc) > the tool was not allowed to exceed this position. > It is related to M198 which is a timer. > Edm process is not complete as soon as the tool arrives to the > programmed position, > as the surface is not completely cut yet. > So, i wait for the tool to maintain the final position until the gap > voltage rises > THEN wait an additional security time (M198 monitors this) > THEN the system is ready to proceed to next step. > This is similar to linuxcnc exactStopMode extreme! > > in general a wedm user interface is different than sink edm > > wedm is a path, like a mill > the wedm cuts while along a path > wedm uses cutter comp a lot, to change the path from roughing ( far > away from the path ) to finishing ( close to the path ) > sink edm is like a cnc punch press > the sink edm gets into position then cuts with very small motion in a > single location > sink edm moves around each location , first achieving th enet shape > during rough stage then about the location ( orbiting ) while refining > the surface > > so the user interface may be very different > > i will look into the user interface also > you still have the coffeecup edm for testing? No, the coffecup EDM have delivered back to child long and broken into pieces long time ago but I have the real machine. The goal with the parameters is to get something similar as for motion.spindle and hopefully same parameters could be used for wire EDM. I think SI-units is a good choice but then they could be scaled to internal units before sent to power supply. ------------------------------------------------------------------------------ Check out the vibrant tech community on one of the world's most engaging tech sites, SlashDot.org! http://sdm.link/slashdot _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
