On Friday 29 May 2020 21:22:09 andrew beck wrote: > thanks guys > > just one question about all this gene > > why do you suggest breaking the ground off the the computer > powersupply cable and connecting a ground from the pc chassis to the > machine main ground? > > does the grounding that goes through the cable now good enough? The computers 3rd wire should remain if its the ONLY ground connection in the whole system. But since we power gear thru EMT tubing, and that also includes a generally better ground, the power cord to the computer should have that 2nd or 3rd ground removed, and the computers ground taken back to this common bolt by a direct from its chassis ground wire, UNLESS the computer is mounted to and mechanically connected to the frame of the machine and this frame is grounded to that common bolt.
Anytime you can lift just one of those wires from that bolt, and still get ground continuity because its grounded someplace else too, is bad kharma, find that "other" ground and disconnect it. Once, done well, is enough. A shielded cable where the shield is grounded at the far end can inject many volts of noise, enough to blow the interface card. Disconnect that far end. That bolt should connect to the power wirings "static" ground, and the only allowable cross connected point to the powerline neutral is in the main entrance box. You don't care if a nearby lightning strike might cause that bolt to be 200,000 volts from ground for a microsecond during that strike, but the +5 volt line will be 200,005 volts, or 5 volts from what it considers its ground might be at that exact instant. You will be damaged ONLY if some other point in the circuit breaks down and some of that lightning bolts energy actually flows thru your system to get to that better ground. Block it with air, the more the merrier. > regards > > Andrew > > On Fri, May 29, 2020 at 11:53 AM Leonardo Marsaglia > <ldmarsag...@gmail.com> > > wrote: > > > Connect all grounds to this bolt, checking that they are not > > > grounded at the far end. By doing this, you are establishing a > > > single point ground that is the zero volt reference for the whole > > > system. Connect this bolt to the building static ground, the bare > > > wire in most power cabling. > > > > > > Connect the machines frame to this bolt. > > > > > > Connect the - rails of all supplies to this bolt. > > > > > > Connect the - terminals of all cards to this bolt. > > > > > > Break the 3rd pin off the computers supply cable and connect the > > > computers chassis to this bolt. Power the computer not from the > > > wall, but from the same power feeding this box. > > > > > > Connect the shielding of all shielded cabling to this bolt. Do not > > > connect the far end of this shielding to anything that is > > > otherwise grounded. > > > > > > By making sure these grounds are not connected anyplace else you > > > are breaking any ground loops which can and will act as antennas > > > to insert noise into your control signals. > > > > > > By having this single point ground, a nearby lightning strike can > > > inject a 100k volt pulse into the system ground as what would be > > > called a "ground bounce", but the system will not see it as noise > > > nor be damaged, because everything is bouncing in unison. > > > > > > Regardless of what that bolt does during the strike, the 5 volt > > > supply remains at 5 volts to this bolt, and the 24 volt supply > > > remains at 24 volts to this bolt. > > > > > > And when the storm is done, your chances of having anything > > > damaged is reduced to the vanishing point. > > > > Thanks for this Gene. It's more than useful to remember how to > > properly ground logic circuits to avoid external noise and dangerous > > voltage spikes. > > > > El jue., 28 may. 2020 a las 20:35, Gene Heskett > > (<ghesk...@shentel.net>) > > > > escribió: > > > On Thursday 28 May 2020 18:36:10 andrew beck wrote: > > > > hey gene > > > > > > > > sorry for the slow reply > > > > > > > > just getting to this grounding thing now. > > > > > > > > currently I have single large copper block with a lot of holes > > > > in it for terminals, this is connected to the earth(ground) wire > > > > that goes back to the whole shed main ground wire rod in the > > > > dirt > > > > > > > > all my servo drives and vfd main grounds go back to this point > > > > as do all motor grounds etc. this is how heidenhain set up the > > > > machine originally and they have a whole grounding schematic > > > > showing how it all works. I have basically copied that. > > > > > > > > I have 2 meanwell 24v powersupplies that have the ground wire on > > > > the input side also connected to ground. the output side (phase > > > > and neutral) are floating relative to ground and not connected > > > > to ground at all I think. as the meanwell switching powersupply > > > > output should be isolated from the input. I think this is > > > > correct > > > > > > > > the computer cord ground wire is currently connected to this > > > > same large copper block. > > > > > > The sheer bulk of that copper bother me a bit because it can > > > function as an antenna. A single bolt, tightened well into the > > > chassis with all the connections stacked up on this bolt will be > > > quieter. > > > > > > > my computer via the 5i25 currently supplies 5v logic power to > > > > the 7i76 > > > > > > Thats fine, but if using a std bob on the other 5i25 port, do NOT > > > enable the jumper for that port as the bobs std grounding will > > > short circuit that, you must supply a separated 5 volts to that > > > bob, or plug in a usb cable to steal it from the pc's usb > > > circuitry, > > > > > > > and one of the 24v meanwell powersupplys supplies the 24v field > > > > io for the 7i76 > > > > > > I have all supplies - outputs tied to ground. It might work 99% > > > of the time without it, till the first nearby lightning strike... > > > With all that grounded, it will probably keep right on working > > > after the strike. > > > > > > > my encoder problems started when I connected up the servo drive > > > > simulated encoder output to the 7i76 encoder counter. I haven't > > > > connected the 5v and 24 v grounds together. they did work for > > > > awhile but don't now > > > > > > Tie the -'s to the ground bolt. Then do a full powerdown on the > > > whole thing, with about a 10 count in the dark, then boot > > > everything back up and test it. And report what you get now... > > > > > > > I have a 7i89 and 7i84 coming soon and will need a external 5v > > > > powersupply anyway i think so will change when they arrive here > > > > from america. > > > > > > > > anyway I read your reply and got a bit confused could you read > > > > my email here and let me know what I need to change and why I > > > > should do it this way just so I understand. > > > > > > Basicly, your lack of a common ground has possibly allowed > > > voltages well above the breakdown voltages of the chips on these > > > cards, between the - connections and a real ground. They all have > > > what can look like an SCR as an isolation tub under each > > > transistor its built from, and these SCR's can upset the whole > > > thing if they get turned on. Thats why the count to 10 or more > > > powerdown is done, giving these SCR's time enough at a low enough > > > voltage to turn them off, at which point the circuit looks more > > > normal and may even work. If not, then something has been damaged > > > on the card and it should be replaced or tested and repaired as > > > needed by Peter. His turnaround time from CA to WV here in the > > > states has been very good. Its several thousand miles. But if I > > > recall correctly, you are not exactly local. > > > > > > > regards > > > > > > > > Andrew > > > > > > > > On Thu, May 21, 2020 at 9:31 PM Gene Heskett > > > > <ghesk...@shentel.net> > > > > > > wrote: > > > > > On Thursday 21 May 2020 01:57:06 andrew beck wrote: > > > > > > hey peter > > > > > > > > > > > > I had a look and the unless there is a shorted out pin in > > > > > > the servo drive connector itself I think the wiring is all > > > > > > good. > > > > > > > > > > > > and I highly doubt the servo drive has problems all I did > > > > > > was connect the 24v back into the servo on pin which is how > > > > > > it has been running for the last ages. > > > > > > > > > > > > anyway how do you recommend commoning the grounds? > > > > > > > > > > > > it is as simple as connecting the field power 0v ground with > > > > > > a wire to the logic power 0v ground? > > > > > > > > > > Yes, but how you do it can be very important. > > > > > > > > > > In your control box, install a longer bolt, #6 or #8-32 to the > > > > > chassis. 4mm if metric. > > > > > > > > > > Connect all grounds to this bolt, checking that they are not > > > > > grounded at the far end. By doing this, you are establishing > > > > > a single point ground that is the zero volt reference for the > > > > > whole system. Connect this bolt to the building static ground, > > > > > the bare wire in most power cabling. > > > > > > > > > > Connect the machines frame to this bolt. > > > > > > > > > > Connect the - rails of all supplies to this bolt. > > > > > > > > > > Connect the - terminals of all cards to this bolt. > > > > > > > > > > Break the 3rd pin off the computers supply cable and connect > > > > > the computers chassis to this bolt. Power the computer not > > > > > from the wall, but from the same power feeding this box. > > > > > > > > > > Connect the shielding of all shielded cabling to this bolt. Do > > > > > not connect the far end of this shielding to anything that is > > > > > otherwise grounded. > > > > > > > > > > By making sure these grounds are not connected anyplace else > > > > > you are breaking any ground loops which can and will act as > > > > > antennas to insert noise into your control signals. > > > > > > > > > > By having this single point ground, a nearby lightning strike > > > > > can inject a 100k volt pulse into the system ground as what > > > > > would be called a "ground bounce", but the system will not > > > > > see it as noise nor be damaged, because everything is bouncing > > > > > in unison. > > > > > > > > > > Regardless of what that bolt does during the strike, the 5 > > > > > volt supply remains at 5 volts to this bolt, and the 24 volt > > > > > supply remains at 24 volts to this bolt. > > > > > > > > > > And when the storm is done, your chances of having anything > > > > > damaged is reduced to the vanishing point. > > > > > > > > > > Stay well Andrew. > > > > > > Cheers, Gene Heskett > > > -- > > > "There are four boxes to be used in defense of liberty: > > > soap, ballot, jury, and ammo. Please use in that order." > > > -Ed Howdershelt (Author) > > > If we desire respect for the law, we must first make the law > > > respectable. - Louis D. Brandeis > > > Genes Web page <http://geneslinuxbox.net:6309/gene> > > > > > > > > > _______________________________________________ > > > Emc-users mailing list > > > Emc-users@lists.sourceforge.net > > > https://lists.sourceforge.net/lists/listinfo/emc-users > > > > _______________________________________________ > > Emc-users mailing list > > Emc-users@lists.sourceforge.net > > https://lists.sourceforge.net/lists/listinfo/emc-users > > _______________________________________________ > Emc-users mailing list > Emc-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/emc-users Cheers, Gene Heskett -- "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) If we desire respect for the law, we must first make the law respectable. - Louis D. Brandeis Genes Web page <http://geneslinuxbox.net:6309/gene> _______________________________________________ Emc-users mailing list Emc-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/emc-users
