On Dec 16, 2006, at 9:40 AM, John Prentice wrote: > Anders, greetings > > Some comments from a personal point of view - others will differ! Ditto!
There are probably as many ways of doing e-stop as there are machines. ;-) From my viewpoint. 24 V for estop is good (reliable). The entire estop chain is pretty much independent of the rest of the logic and the PC. The estop chain includes a latching relay which drops out on estop and has to be manually reset to get out of estop. Some machines may coast over the e-stop cam and suddenly re-energize the servos...not a pleasant surprise. It takes a lot of switches but: for each axis there are estop switches at either end of the axis; inside that are the machine hard limits and of course just inside that are the soft (pc) limits. If your servo drives have + and - inhibits then these connect to the machine hard limits stopping in that direction but allowing one to back off the limit. So for an axis that is out of control the soft limits hit first followed by the machine limits which should halt the axis followed by the estop which may or may not dynamic (resistance) brake the servo motors. Some people have suggested zeroing the servo input and then removing the AC input to the servo power supply. Note that this is backwards of the way emc expects things to work where it controls the estop relay. What ever you do the machine must stop on e-stop independent of the pc. In a manner similar to the servos I would expect that an inhibit or zero drive to the VFD would halt the spindle rather quickly. Dave > > > >> start at the HAL pin E-stop-out. It will be high whenever EMC thinks >> it's OK to go out of E-stop. >> >> E-stop-out drives the coil of a NO(normally open) relay. This relay >> connects +12 V to the beginning of the E-stop chain only if E-stop- >> out >> is high. >> >> On the +12V wire, a number of NC(normally closed) switches follow: >> Red >> E-stop buttons (2), Limit switches(6), NC relays on the servo amps >> (3). >> >> At the end of the chain we are going to have either +12V if >> everything >> is OK, or 0V if we are in E-stop. > > Yes - basic series architecture is right. Not having the PC logic > ground > wandering around the machine amnd using 12 (or even 24 volts) is > also good > to avoid noise pickup. > > I would not want the servo faults to E-stop the whole system and > would be in > two minds about the limits. On a modest size machine it is probably > safe for > EMC to implement the limits. In particular if the spindle stops > before axis > movement nasty things happen to tool and workpiece. If spindle is > running > the tool is likely to be safe. On a big machine you might have to have > limits to protect against servo runaway (e.g. from failed encoder) > which > could smash up the iron. > > The other issues are: > > (a) That having separate limit switches on inputs allows you to use > them as > Home switches too; I always like to reference my machines even when > touching-off to get work offsets set. > > (b) You have to be able to jog to get off the limits. Handles on a > decent > sized machine are generally useless and hurt when they hit you so > you need a > limit by-pass switch in your scheme and it would have to be monemtary > contact for safety. >> >> Use the +12V to power the coils of three NO relays (coils in >> parallel): >> 1. one relay connects +5V to E-stop-in telling EMC everything is OK >> 2. one relay grounds the Poweron/ input of a Pico Systems power >> switch - >> that will give power to the servo drives. >> 3. one relay connects 230 VAC power to a VFD which powers the spindle >> motor > > You might not need 3 relays if you have sufficient isolation between > contacts on a single relay. > > Instead of 3, has your VFD got an Enable input? As with the Pico > board it > would be nice for the VFD to use its dynamic braking capabilities > rather > than having the motor coast to a halt. Also in some circumstances > (e.g. DC > drives) the motor/spindle KE can overvolt the drive when the supply is > disconnected. The DC fuse blowing on a big stepper motor power > supply can > cause real problems like this. >> >>> >> Some questions I have: >> -wiring three relays in parallel at the end of the chain will >> result in >> three times the current needed to pull one coil of the relays. Is >> this >> going to be a problem ? (other than that the +12V supply must cope) >> -is it a good idea to just cut the AC power to the VFD in an E-stop >> situation ? are there any other better ways to safely stop the >> spindle > > See above and VFD manual > >> -If I connect the common terminal of the limit switches to the >> 'upstream' end of the E-stop chain, can I wire the unused NO >> output of >> the limit switch to another HAL pin ? This way an E-stop due to a >> limit >> switch could be diagnosed by EMC as coming from a specific limit >> (+/-) >> and axis (XYZ). > > Yes but you would have to put switches before the first 12v relay > and risk > getting noise onto PC ground. > > It is not IMO necessary for EMC to know which direction (+/-) has > tripped - > you can tell by eye - sometimes very useful to know if it is X or Y > though. > > > Hope this helps > > John Prentice > > > > > ---------------------------------------------------------------------- > --- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to > share your > opinions on IT & business topics through brief surveys - and earn cash > http://www.techsay.com/default.php? > page=join.php&p=sourceforge&CID=DEVDEV > _______________________________________________ > Emc-users mailing list > Emc-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/emc-users ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
