EmbeddedRelated.com
Forums
Imagine Conference

RS485 is bidirectional does it mean it is fullduplex?

Started by Swizi June 15, 2005
On Fri, 17 Jun 2005 00:22:38 +0100, "Steve at fivetrees"
<steve@NOSPAMTAfivetrees.com> wrote:


>Absolutely. Connecting to local ground at both ends of the run is a recipe >for HUGE ground currents. I've seen (very) melted ground wires. It is indeed >a safety hazard.
Yes, this is a severe problem when the TN-C wiring system (L1, L2, L3 and N) in which the neutral is connected directly to each equipment frame. In this case quite a few Amperes can flow in the cable shield. However, with TN-S wiring (L1, L2, L3, N and PE) with the PE connected to each frame, grounding the data cable at both end is not so bad, since large AC currents will flow in the cable shield only during ground faults (Lx to PE) until the fuse is blown. Unfortunately pure TN-S systems are hard to find, since often the N, PE, the grounding electrode and all metal piping is connected to the same main grounding bar (equipotential bonding) close to the mains cable entry. From the main grounding bar to the distribution transformer, the system is just TN-C. Thus, most wiring systems claiming to be TN-S are actually TN-C-S, thus grounding the data cable shield at both ends would not be that bad _within_ the building, but connecting a data cable shield between two buildings, will suffer from the same huge currents as a TN-C system would. Paul
On Thu, 16 Jun 2005 18:49:55 -0800, floyd@barrow.com (Floyd L.
Davidson) wrote:

> >Here's a ground loop: > > Signal Source > > o > | > Rload > | > +---------> connection =======//======= <------+ > | to cable | > Rwire shield | > | | > | | > ----- Earth Ground ----- > --- --- > - -
In the TN-C electric distribution system, the "Signal source" is the mains phase voltage (100-240 V depending on country), Rload perhaps 10-1000 ohms, Rwire perhaps 0.2 ohm. An AC current up to 25 A could flow though Rload and through Rwire. The voltage drop across Rwire would be up to 5 V. Before connecting the data cable, there would be a 5 V ground potential difference between the left and right side equipment. When the cable shields are connected, a part of the up to 25 A flowing through Rload will be diverted through the data cable and into the right hand equipment ground. The magnitude of this "ground loop" current depends on the ratio of the Rwire compared to the combined resistance of the cable shield and the right hand side equipment grounding resistance. Especially, if the right hand side equipment is close to the main distribution panel, the total path resistance is dictated by the shield resistance. If the alternate path total resistance is as low as Rwire, more than 10 A could flow in the cable shield, which could even cause a fire hazard. Paul
Paul Keinanen wrote:
> floyd@barrow.com (Floyd L. Davidson) wrote: > >> Here's a ground loop: >> >> Signal Source >> >> o >> | >> Rload >> | >> +---------> connection =======//======= <------+ >> | to cable | >> Rwire shield | >> | | >> | | >> ----- Earth Ground ----- >> --- --- >> - - > > In the TN-C electric distribution system, the "Signal source" is > the mains phase voltage (100-240 V depending on country), Rload > perhaps 10-1000 ohms, Rwire perhaps 0.2 ohm. An AC current up to > 25 A could flow though Rload and through Rwire. The voltage drop > across Rwire would be up to 5 V.
You should also realize that 'signal source' above includes a myriad of possibilities, including lightning, pools of acid forming batteries, whatever. We can never really get at a true ground except through some sort of impedance. Luckily most (but not all) of our controllable signal sources are imposed between the marked signal point and the junction of Rwire and Rload. Also the effects of Rwire, and the IR drop across it, can be present at either the near or far end, or both. Those R things are not necessarily resistive, nor passive. -- Chuck F (cbfalconer@yahoo.com) (cbfalconer@worldnet.att.net) Available for consulting/temporary embedded and systems. <http://cbfalconer.home.att.net> USE worldnet address!
On 2005-06-17, Floyd L. Davidson <floyd@barrow.com> wrote:

>>The cable shield was grounded. What wasn't grounded was the >>RS-4xx driver/transceivers. >> >>----+ +------+ +------------/ /-- Shield >> | | Opto | +-----+ \ >> uP|-Tx--| Iso |--Tx--| 485 |--|---A-----/ /-- >> |-Rx--| |--Rx--| xcvr| | >> |-dir-| |-dir--| |--|---B-----/ /-- Mirror >>-++-+ | | +-++--+ | Image >> |`--Pwr--| Pwr |-IsoPwr-'| | >> `---Gnd--| Iso |-RScom---+-----|---com---/ /-- >> | +------+ / >> /// +--+--------/ /-- >> | >> /// > > > Okay. That's just fine. > > Hmmmm... It just occurred to me that if this goes into > customer premise locations, the benefit is obvious. That is > *much* less complex than telling someone they have to install > an appropriate ground system!
I think that was the general idea. The people who installed this stuff were much more used to installing 4-20mA current loop instruments, which were always isolated from ground, and the 4-20mA current loop could be grounded at any point in the loop (but hopefully only at one point).
>> That's sure not the way I remember it in the process control >> systems I worked with (it's been 6-7 years). The RS-485 bus >> was pretty much always floating w/ respect to ground, with A, >> B and a floating common wire between the two RS485 >> transceivers. There may have been people that grounded the >> RS-485 common node at some point, but it was expected to work >> if it was floating. > > I suspect that was engineered around less than the best customer > premises. They can't control the customer's environment, so > designing it to avoid that altogether is a smart thing to do.
RS-485 was pretty darned high-tech at the time. We were using it to replace a weird proprietary busd that used 48V differential signalling (with a floating common) that ran at two different baud rates. The "high speed" version was 250 baud, IIRC. A big advantage of RS-485: it doesn't hurt nearly as much as 250Hz at 48V. The 48V was current-limited but it still stung. -- Grant Edwards grante Yow! Okay... I'm going at home to write the "I HATE visi.com RUBIK's CUBE HANDBOOK FOR DEAD CAT LOVERS"...
"Floyd L. Davidson" <floyd@barrow.com> wrote in message 
news:873brhefvn.fld@barrow.com...
> "Steve at fivetrees" <steve@NOSPAMTAfivetrees.com> wrote: >>"Floyd L. Davidson" <floyd@barrow.com> wrote: >>> >>>>As I've just said elsewhere, this is simply nonsense. >>> >>> And you are out of your field... >> >>I've pointed out several factual errors with your posts. I've demonstrated > > You have yet to point out a single factual error. Just *you* > saying it is, without supporting your comments with discussion > that explains a bit of it, is indicative.
<snip>
> But you are going to have to support it with > more than pontification...
I just looked up the dictionary definition of "pontification". It has two meanings, but I'm ignoring one, as I have no immediate plans to become the next pope. The other involves opinions. The only opinion I've expressed in this thread is that you're misguided. Everything I've said about RS-422/485 is factual, and you are very welcome to check these facts against the published specs. (I have them here, and am very familiar with them. You don't appear to have read them.)(I can't point you at them on the web; the specs are not in the public domain.) Equally, everything I've said about double-grounding is also factual, and demonstrable. I'm not aware of any "specs" as such on this issue, but it is hard to argue with a melted signal cable. Finally, look around. Do you see anyone agreeing with your somewhat strange point of view? Does this not tell you something? Whatever it is you're talking about, it's certainly not RS-422/485. Steve http://www.fivetrees.com
Paul Keinanen <keinanen@sci.fi> wrote:
>On Thu, 16 Jun 2005 18:49:55 -0800, floyd@barrow.com (Floyd L. >Davidson) wrote: > >> >>Here's a ground loop: >> >> Signal Source >> >> o >> | >> Rload >> | >> +---------> connection =======//======= <------+ >> | to cable | >> Rwire shield | >> | | >> | | >> ----- Earth Ground ----- >> --- --- >> - - > >In the TN-C electric distribution system, the "Signal source" is the >mains phase voltage (100-240 V depending on country), Rload perhaps >10-1000 ohms, Rwire perhaps 0.2 ohm. An AC current up to 25 A could >flow though Rload and through Rwire. The voltage drop across Rwire >would be up to 5 V.
No. "Signal Source" means whatever your *desired* signal is. Not the AC power distribution... simply because how much that is perterbed is of no significance. Across Rload... is your Hifi Amp, your telephone set, the PA system, computer data, gas tank level sensor output, whatever... However, we could add that and many other sources of voltages. It makes the diagram and the discussion more complex, but the one particular example you chose is an excellent one because it is commonly seen and often has significant effect.
>Before connecting the data cable, there would be a 5 V ground >potential difference between the left and right side equipment. When >the cable shields are connected, a part of the up to 25 A flowing >through Rload will be diverted through the data cable and into the >right hand equipment ground.
But lets not suggest that it would be a significant part of that 25 A. We are *not* describing ground *faults*, but ground *loops*! But, in a typical arrangement, it could certainly have enough effect to cause one heck of a 60 Hz hum in the "signal" as described above.
>The magnitude of this "ground loop" current depends on the ratio of >the Rwire compared to the combined resistance of the cable shield and >the right hand side equipment grounding resistance. Especially, if the >right hand side equipment is close to the main distribution panel, the >total path resistance is dictated by the shield resistance. If the >alternate path total resistance is as low as Rwire, more than 10 A >could flow in the cable shield, which could even cause a fire hazard.
We need to make the distinction that this is a ground system fault, not an inherent characteristic of the indicated connections. And it can happen whether the arrangement is properly designed or not. -- Floyd L. Davidson <http://web.newsguy.com/floyd_davidson> Ukpeagvik (Barrow, Alaska) floyd@barrow.com
Paul Keinanen <keinanen@sci.fi> wrote:
>On Fri, 17 Jun 2005 00:22:38 +0100, "Steve at fivetrees" ><steve@NOSPAMTAfivetrees.com> wrote: > >>Absolutely. Connecting to local ground at both ends of the run is a recipe >>for HUGE ground currents. I've seen (very) melted ground wires. It is indeed >>a safety hazard.
I hate to mention this, but there is a difference between comm cable and power cable. The Neutral in three phase power systems is *not* the equivalent of the shield on a comm cable.
>Yes, this is a severe problem when the TN-C wiring system (L1, L2, L3 >and N) in which the neutral is connected directly to each equipment >frame. In this case quite a few Amperes can flow in the cable shield. > >However, with TN-S wiring (L1, L2, L3, N and PE) with the PE connected >to each frame, grounding the data cable at both end is not so bad, >since large AC currents will flow in the cable shield only during >ground faults (Lx to PE) until the fuse is blown. > >Unfortunately pure TN-S systems are hard to find, since often the N, >PE, the grounding electrode and all metal piping is connected to the >same main grounding bar (equipotential bonding) close to the mains >cable entry. From the main grounding bar to the distribution >transformer, the system is just TN-C. > >Thus, most wiring systems claiming to be TN-S are actually TN-C-S, >thus grounding the data cable shield at both ends would not be that >bad _within_ the building, but connecting a data cable shield between >two buildings, will suffer from the same huge currents as a TN-C >system would. > >Paul
-- Floyd L. Davidson <http://web.newsguy.com/floyd_davidson> Ukpeagvik (Barrow, Alaska) floyd@barrow.com
"Steve at fivetrees" <steve@NOSPAMTAfivetrees.com> wrote:
>Equally, everything I've said about double-grounding is also factual, and >demonstrable. I'm not aware of any "specs" as such on this issue, but it is >hard to argue with a melted signal cable.
So you declare yourself correct, eh? No discussion, no facts, nothing but pontification. If it is demonstrable, why don't you! -- Floyd L. Davidson <http://web.newsguy.com/floyd_davidson> Ukpeagvik (Barrow, Alaska) floyd@barrow.com
"Floyd L. Davidson" <floyd@barrow.com> wrote in message 
news:87d5qkdfw4.fld@barrow.com...
> "Steve at fivetrees" <steve@NOSPAMTAfivetrees.com> wrote: >>Equally, everything I've said about double-grounding is also factual, and >>demonstrable. I'm not aware of any "specs" as such on this issue, but it >>is >>hard to argue with a melted signal cable. > > So you declare yourself correct, eh? No discussion, no facts, nothing > but pontification.
I repeat: please see the RS-422/485 specs.
> If it is demonstrable, why don't you!
I, and others, have been trying. I'm about ready to give up. Steve http://www.fivetrees.com

Steve at fivetrees wrote:
> "Steve at fivetrees" <steve@NOSPAMTAfivetrees.com> wrote in message > news:p5ydnfE1eqqnoS_fRVnyjw@pipex.net... > > "Floyd L. Davidson" <floyd@barrow.com> wrote in message > > news:87zmtpelyk.fld@barrow.com... > >>>You've mentioned telephone system cabling: I'm no expert, but I had > >>>thought > >> > >> If you have no expertise in this topic, please *do not* pontificate. > > > > You'll forgive me for laughing out loud here. > > Figured I should clarify that. > > Floyd, I'm really trying hard (now) to follow you. I've taken every diagram > you've drawn, viewed it with a fixed-width font, and tried hard to > understand your point. (Perhaps I'm misunderstanding something.) I've gone > back and re-read all your posts. I'm utterly convinced you're sincere, > you're passionate, and you know your stuff. I'm equally convinced that said > stuff does not include RS-422/485, or medium-haul datacomms in general. I > really believe you're missing the point (common-mode). > > I'm also enjoying this thread. Keep 'em coming ;).
IMHO, This is related to the difference between mechanical and electronic ground. In the systems we made years ago, the electronic ground was isolated from the mechanical ground. A jumper was provided that enabled the customer to tie both grounds if he wished. He was responsible for the earthing of the system. You will find perhaps the following useful: (http://www.spheresystems.com.au/RS485.html) Ground and Earth Connections The grounding and earthing connections in RS485 provide two separate functions. The second is related to safety and the first to establish a reference voltage Safety The RS485 *cable screen* must be bonded to the protective earth system of a building at *one point* only. The cable screen must be electrically continuous throughout the entire cable run Voltage Reference The screen of the RS485 cable establishes a ground reference voltage for the RS485 signal conductors. For this reason the cable shield must be connected to the *ground reference* for *each node* on the network. It is not acceptable practice to tie the node ground reference to the building protective earth as this will introduce electrical noise into the system and may lead to equipment damage in the event of electrical fault currents

Imagine Conference