On Thursday 14 May 2009, Gary P. Fiber wrote: >Gene Heskett wrote: >> On Thursday 14 May 2009, [email protected] wrote: >>> Hi >>> I want to ask about shielding of cable. >>> I am building bigger machine and for my new machine I need extend my >>> encoder cable. My encoder cable has DB 15 connector. >>> I do not want to cut cable and re-solder all 15 wires. I want to use >>> pre-made cable that already has male female connectors so I can use them >>> as an extension cord. >>> First, can I do that? >>> Are there any problems with using extension cable (10 ft) for encoder >>> signals? >> >> With adequate shielding I think you can. >> >>> Second question, that extension cable has aluminum wrapping around wires. >> >> ALU wires=Oops. mylar/alu foil=good but see next question. >> >>> That cable does not have shielding like woven wire mesh around aluminum >>> wrapping. >> >> Does it have a 'drain wire' that is about 22 or 24 gauge equ in stranded >> wire that can be brought out and tied (soldered) to the appropriate ground >> point? >> >> At frequencies we deal with, it can be as much as 60db quieter than an >> equivalent 98% braid covered cable. (That in math terms, is about 1 >> million times quieter, and is because the mylar foil is a 100+ % coverage, >> not the braids 98% at best, and is often 90% or less in cheap cable) This >> drain wire should be electrically bonded (soldered) to the DB15 shell on >> both ends, likewise your existing cable you want to re-use should also >> show continuity from end shell to end shell. I'd scrape it just to see the >> shielding before I'd use it. No shielding, nope.. >> >> Whether you tie that drain wire and the machine end of the DB15 shell to >> the machine depends on how the encoder is covered. If it is in a metallic >> housing that is not grounded to the machine, then connect the encoder >> housing to the DB15 shell at the machine end. This would be the ideal >> situation, but if the encoders housing is metallic, and grounded to the >> machine, you may have to opto isolate the encoder signals to get truly >> clean A-B-Z sigs. An oscilloscope will tell that tale. Or take more >> heroic measures to shield the spindle motor power & keep its noise under >> control if that is the primary noise src. >> >>> So, do those woven wire mesh around aluminum wrapping important? >> >> Yes, unless they too are alu (test by soldering, if it won't, its alu), >> which would be somewhat less useful than those belly appendages on a boar >> hog. You should find another brand of cable where that is tinned copper & >> solderable. The alu oxide, forms in .001 second or less when alu is >> exposed to the oxygen in the air, is a very very good insulator, and that >> is not what you want wrapped around that foil supplying its ground. >> >>> Will noise pass through single aluminum wrapping? >> >> With the 'drain wire' well grounded, the isolation will be somewhere in >> the range of 100-110 db. 120 with good terminations. Std braid only >> covered coax is in the range of 55-65 DB, and the FCC forced the cable >> companies to replace all that 20+ years ago because it leaked so bad. >> Those cable operators that drug their feet had a dead giveaway, visible to >> any savvy tech, the channels on their system that had local broadcasters >> on them were unusable because we also leaked _into_ their systems. The >> cable folks were hard to convince locally, particularly when they needed >> the channel we were on for the last 25 years, then discovered they could >> not get rid of the ghosts. We just as firmly replied that it was NOT our >> problem and that they should tighten up their systems to meet the FCC >> leakage specs, and it came home to roost about a year later when the FCC >> truck came to town and wrote them up a citation for every stop the truck >> made. Lets just say that Pandora's box was full. Some viewer called them >> I guess, it sure wasn't us. >> >> Yes, I'm a retired broadcast engineer with a 1st phone from 1962, and have >> been a C.E.T. since 1972. :) >> >>> For the spindle I am using 2.5 hp router. Not many electric power tools >>> around. >>> That spindle and 3 servomotors. >>> >>> Thanks >>> aram > >I am curious wouldn't you be concerned with creation of a ground loop >when grounding the "drain" wire at both ends? Or will the servo motors >ignore a ground loop? > This is why I said IF the encoder is isolated/ungrounded, which with the plastic molded housings favored today can be a possibility. If its grounded, the of course that would setup a ground loop and all bets are off. Then you have to treat it as a balanced differential pair at the receiver, running it through a comparator before handing it off to emc. A bit of hysteresis, a volt or so, also will be needed.
In no event, is the servo power common, or the shielding to the servo inputs to share the grounding of the encoder. If you have a ground on any shield which is disconnected at the center of the 'star', then that has to be found and isolated. That star point must be the only common. >During my railroad days working on Defective Equipment Detectors in the >1980's, most of my shielded cable was grounded to the rack but not at >the Bolometer end. The circuits for the infra red sensors were real high >impedance so any noise at all would translate to false train stops at >the recorder. And by the time the operating engineer has walked back to check the hot boix it thought triggered the stop, he finds it maybe 10F over ambient, converting him into an unhappy camper at best. >Here's something you'll appreciate, Once in a trip into Illinois we had >to move a detector readout location . I think we ran 16 and 18 KHz line >overlay carrier and shot that 105 miles down # 6 solid copper, 600 >balanced dispatcher line. #6? that is more then just power for a sensor. >In that 105 Miles we lost 10db checked on a HP >3551 transmission test set and a Harmon Selective voltmeter.. At one >point we got so much AC off the lines due to overhead power running >parallel we had to install 60 Hz drains on the open wire line. Which probably should have been cross-over swapped as it passed a given pole. But the wire stringers probably wouldn't have grokked the reason for that all that well. :( Another time I was in on checking how well one could ship video over common 16 gauge Packard Automotive wire, so we bought a 1000 gallon stock tank, tossed in a couple hundred pound bags of salt and filled it up with a puddle gulper sucking in the dregs of Mission Bay (San Deigo) which was in our back yard. Then we unwound and tossed 100' of this wire into it and did a video sweep. Darned thing was plumb usable as long as our gear was grounded to the tank. Based on that, we went ahead and installed the two cameras the Navy had bought from us onto the Trieste. A couple of months later those two cameras went down to the bottom of the mohole, 37,000 feet deep in the pacific, and at an outside pressure of about 18,000 psi, they worked well. No such thing as a VCR in Feb of 61 when they did that, so Jacques C. (Yup, THAT Jacques C.) and a little bitty navy guy (see wikipedia->Trieste) used several cameras to shoot pix off the 8" monitors, and its those pix that were used in the next several editions of the encyclopedia's. The only pictures ever from that depth as they didn't trust the Trieste to ever do it again. Having my fingerprints on the innards of those two cameras didn't hit me as an ace in my hat band for several more years though. And I'm feeling better about it all the time cuz it appears they will not visit that place ever again. They have now had from Feb 1961 to May 2009, 2 years short of 50 years to do it again. >Oh ya I am just getting my little PC Router set up using EMC2 and a >Shuttle computer. The mechanics for the router / engraver was made by >Paul Jones. Have his PCB spindle too. I am using Probotix motors and >drivers. Huge table 5.5 inches by 5.5 inches :) 3 inch Z axis travel. That is just a wee bit smaller than my lashup. I have about a foot of X, 5.5" of Y, and the whole micromill post for Z, 15" maybe. Overkill, but it gets the bit out of the back of my hand when I'm working on the table. Z drive is a fresh design, and can put 150+ pounds on a drill bit. <http://gene.homelinux.net:85/gene/emc> >I have to measure the X and Y axis travel yet to set up the stepper conf >but it does move and will simulate the snowflake example in EMC2 until >it hits the ends of the travel. The router should be just right for me >for now since I don't make to large of circuit boards. I want to use >gEDA and PCB for the circuit design but may use Eagle and PCB-GCcode. I have contemplated some pcb work, but how does one go about keeping it flat on the table? I've not seen any material that will stay flat as it warps upwards as soon as the copper is penetrated. I don't have a vacuum table obviously. :( >Gary Fiber K8IZ > >2nd Phone 1979, FCC converted to a GROL in 1984 with Ship Radar >Endorsement since 1987. > > > > > >---------------------------------------------------------------------------- >-- Crystal Reports - New Free Runtime and 30 Day Trial >Check out the new simplified licensing option that enables >unlimited royalty-free distribution of the report engine >for externally facing server and web deployment. >http://p.sf.net/sfu/businessobjects >_______________________________________________ >Emc-users mailing list >[email protected] >https://lists.sourceforge.net/lists/listinfo/emc-users -- Cheers, Gene "There are four boxes to be used in defense of liberty: soap, ballot, jury, and ammo. Please use in that order." -Ed Howdershelt (Author) Speak softly and own a big, mean Doberman. -- Dave Millman ------------------------------------------------------------------------------ Crystal Reports - New Free Runtime and 30 Day Trial Check out the new simplified licensing option that enables unlimited royalty-free distribution of the report engine for externally facing server and web deployment. http://p.sf.net/sfu/businessobjects _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
