On Saturday 17 May 2014 05:06:08 EBo did opine And Gene did reply: > On May 16 2014 11:26 PM, Marius Liebenberg wrote: > > On 2014-05-17 01:43, EBo wrote: > >> On May 16 2014 11:14 AM, Gene Heskett wrote: > >>> On Friday 16 May 2014 12:15:06 Marius Liebenberg did opine > >>> > >>> And Gene did reply: > >>>> Look under the High Speed Mode - up to 3.4Mb/sec > >>>> > >>>> > >>>> http://www.esacademy.com/en/library/technical-articles-and-documen > >>>> ts/mi scellaneous/i2c-bus.html > >>> > >>> Thank you Marius. But nowhere in that link is any mention of using > >>> differential pair cabling, its all single ended all the way down. > >>> There > >>> may be chips that can convert it to cat5 differential pair's and > >>> back. I > >>> haven't researched it. Needing to tristate them when not busy > >>> complicates > >>> it. > >> > >> Isn't it possible to just tie the twisted pair to a single pin (for > >> noise cancellation)? I worked with something in the past that I > >> remember as having that arrangement. Then you would not have to > >> have > >> logic for each line, but each pin. I can see the benefit of adding > >> isolation logic between the pins and cable thought... > >> > >> EBo -- > > > > Not worth all the effort. I2c is not really meant for long distance > > communications but rather a protocol that is applied on the same > > circuit > > board or close proximity boards. It is an high speed inter device > > protocol that useses very little hardware resources (pins). My goal > > is > > to provide an expansion bus that can hook up several functions that > > are > > commonly desired by machine builders making use of just the parport. > > Things like AtoD for plasma, IO expander, pwm to analog. Mostly non > > mission critical stuff. > > Fair enough on the port buffer, but tieing two wires together to reduce > EMF induced current seems trivial -- although I have no idea how far > that would add to the line caring capacity. > > EBo --
And would have even in a twisted pair quad cable such as cat-5, a minimal effect. That stuff is made for differential signalling, and there is nothing differential about i2c, any version. I don't use it for anything but ethernet, no shielding at all. The shielding on a "star-quad" cable would reduce the working range because of the capacitance, only the shielding of such a cable could be measured as resulting in better noise immunity, but only if its grounded at one end only. Both ends grounded could generate ground loop currents, generally not a good idea. I try to attain a star topology ground system all the way to the frames of the machines that BOB is controlling. That means one bolt point where all the grounds come together. With modern IEC line cords on everything, that often can't be done because both the computer and the motor psu/BOB box will have a separate connection back to the static ground in the power strip. However, I treat the ground in that box as isolated, so while the box is connected to the 3rd pin of the power cord, the actual ground the whole system uses is made thru the DB25 connectors ground. The only place I use a star-quad cable This is not 2 twisted pairs, but 4 wires layed up in a twist, inside a 100 db rated shielding) is in stepper motor wiring, and I wire that cable up to use the opposite corners of the cable layup for each winding of the motor. With the steppers alive, I can't find more than 1 or 2 millivolts of the chopper noise on the outside of the cable insulation with my oscilloscope. I've looked at the Belden version of that cable but its slightly stiff, its plastic insulation is fairly hard compared to Clark's version. Susan Clark (Clark wire & cable in Windy City) imports a far eastern made version of that cable in sizes of 20 gauge and up, 22 being heavy enough for NEMA 23 stepper motor currents. I use the 22 gauge version for my motor wiring because its very supple, it just lays there like cooked pasta. Intended for microphone cable, I've put at least a mile of it into the system at WDTV, where it lowered the noise floor to let the air conditioning in the studio be the noise floor, 75 or 80 db down. Out in the production van, we have had the usual Sony lapel mic out on the end of 300 feet of it, with people walking on it, without problems. And even in the greasy/oily environment of my stuff, it seems to be destined to work forever as some of it is over a decade old now. 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) Genes Web page <http://geneslinuxbox.net:6309/gene> US V Castleman, SCOTUS, Mar 2014 is grounds for Impeaching SCOTUS ------------------------------------------------------------------------------ "Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE Instantly run your Selenium tests across 300+ browser/OS combos. Get unparalleled scalability from the best Selenium testing platform available Simple to use. Nothing to install. Get started now for free." http://p.sf.net/sfu/SauceLabs _______________________________________________ Emc-developers mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-developers
