RS485 is bidirectional does it mean it is fullduplex?

"Floyd L. Davidson" <floyd@barrow.com> wrote in message 
news:873brf5ovi.fld@barrow.com...
>>[1] I can provide a synopsis if needed.
>
> I don't expect that you could.  Not an accurate one.
>
> Note that the others in this discussion tend to keep the
> discussion technical, which you seem to try avoiding.
>
> I won't respond to any more of your articles if they do not
> contain technical discussion.

<plonk>

Steve
http://www.fivetrees.com

"Paul E. Bennett" <peb@amleth.demon.co.uk> wrote:
>Floyd L. Davidson wrote:
>> I pointed out why this idea that cable shields can *only* be
>> grounded at one end is wrong, explained why and under what
>> circumstances they might be engineered differently.  You two
>> claimed that was wrong.  I pointed out that every outside plant
>> telephone cable in the country does in fact have multiple points
>> where the shield is grounded.
>>
>> You've both been trying to deny that.  But it is a simple
>> *fact*.
>
>All we have seen is a lot of hand waving.

Why make claims like that.  It isn't true, and you *know* it.
I'm just astounded at how thick skulled some people can be.

In addition to detailed discussion of the theory, I've
previously posted references that make it clear enough that the
above is a fact.  Outside plant communications cables are
installed with a ground at every section splice.  At each end
where they enter a building they are grounded to the building
ground.

Here they are *again*:

  http://www.nrao.edu/evla/geninfo/memoseries/evlamemo41.pdf
  http://www.fac.ilstu.edu/Facilities_Services/Fac_Plan_Home/DesignGuidelines/16743%20-%20Telecom%20Risers,%20Closets%20and%20Entries.htm
  http://www.usda.gov/rus/telecom/publications/word_files/1751f810.doc
  http://www.tequipment.net/WilcomTeleTransT124.asp

Or, you can skip all of that again, and just respond to this
one quote (emphasis added):

     3.2.4 Grounding of Cable Shields/Support Strands
     (Messengers):  Normal construction practices and NESC
     provisions require that cable *shields*, *messengers*, and
     other *non-current carrying metallic hardware* be effectively
     grounded.  It is especially important to effectively ground
     cable *shields*, *messengers* and *non-current-carrying* metallic
     hardware at *dead* *ends* and other *junction* *points* for noise
     mitigation, personnel protection, and/or power contact
     protection.
         RUS Bulletin 1751F-815
         http://www.usda.gov/rus/telecom/publications/

It also says that cable shield should be bonded to messenger
cable or other support at least 4 times per mile if that can be
done without opening the sheathing otherwise (at points where
splices exist anyway), but for long runs with no such convenient
points the cable sheath should be broken every 1 mile to provide
such bonding.  The messenger cable or support, of course, is
required to be grounded.

If you are not aware of what RUS is, I would suggest reviewing
their homepage at http://www.usda.gov/rus, because they are a
major part of the interface between the Federal government and
the US telecommunication industry.

Rather than doing your own hand waving, you need to *read* the
cites that I provided.  There have been three people involved in
this discussion doing a lot of hand waving.  They either cite
*nothing*, or use examples that might best be described as
imaginative.  Sometimes those are indeed accurate, but none of
them are common enough that typical engineering requirements
would take them into account.

(I've seen locations where the outside air temperature can be
-40F, and the inside heat has been known to fail, therefore
*you* should only design equipment that can operate at -40C???
That is no different than claiming there are places where a
ground fault in the power system has melted a cable.  In one
case the problem was the *heater*, not the equipment that didn't
work at 40 below.  In the other case the problem is the power
system, not the RS-485 ground.)

>for your position. Steve and myself have stated that we seen fried cables
>because people disobeyed the rulings about screen connectivity.

A lot of people attribute any number of things to odd causes
that often have no significance.  The fact that you've seen
cables fried by ground faults should tell you something about
ground faults.

As I've demonstrated *very* well, communications cable is
grounded at regular intervals, including both ends.  I've
discussed the theory involved to show why that is done.

Yet you want to say it shouldn't be...

>> And how that applies to RS-422/485 should be obvious to both of
>> you, but doesn't seem to be.  You've both been exposed to a
>> little bit of technology using twisted pair as a transmission
>> line, and have learned some rules of thumb, but not the basic
>> theory behind it.  If you would cease being so stubbornly hung
>> up on your simple rules of thumb, and learn something about it
>> on a broader scale, I'm sure that both of you would enjoy the
>> value you get from a better understanding.
>
>We are still waiting some serious attempt to convince us otherwise.

I can't get past people who tell me there are 7km loops on ships.
I probably can't convince you that there is no Santa Claus at the
North Pole either.

That doesn't mean that I haven't provided a very clear and very
logical proof of the above statement.

>> You continue to claim that what is done as a standard practice every
>> where can't be.  It still is, whether *you* can understand it or not.
>
>I never claimed it to be standard practice everywhere but certainly is in

*I* said it was standard practice.  *You* said it isn't.  I
showed where it is used, explained why it is used, and you say I
haven't convinced you.

Here's the water trough...  but *you* are the one who drinks or
not.

>situations that I have come across. I have, as I have stated, also seen the
>other side in the ship wiring side of things.

Yeah, sure.  7 km loops...

>> All the math in the world won't help if you don't understand how
>> to spot parameters that don't fit.  If you *start* with off the wall
>> numbers, you math is going to give you off the wall results.
>
>My numbers were not off the wall as I have the readings from my log book to
>back up those I used in the example. We still haven't seen you put numbers
>to your diagram so I wait.

Your 7 km loop is off the wall.

What do you mean I haven't put numbers to my diagram?  I posted
several diagrams, and have listed more numbers for a larger
variety of things than you seem to be able to comprehend, much
less remember.  I provided *detailed* job specifications describing
how it is done too!  Right down to the minimum size of the ground
wires.

You can't come up with anything that refutes what I've shown,
and the references I've cited...  why not just admit that it was
correct from the beginning.  Like I said, it really *is* useful
information to understand.

One of the most notable characteristics of this exchange is how
none of you want to discuss the details I provide.  I post
cites, nobody mentions them and you all drop whatever angle the
cites were posted to counter, and instead you try some other
equally inaccurate approach.

But the few cites that others have made *do* get a response from
me.  I show *in detail* why the cite does not support a point or
is not correct.  Nobody has answered that.  It, again, was
dropped immediately...  because it is *obviously* irrefutable.

So, tell me again who is doing the hand waving here?  And then
tell me about 7 km loops on a ship again too.  And all about
Santa while you are at it.

-- 
Floyd L. Davidson           <http://web.newsguy.com/floyd_davidson>
Ukpeagvik (Barrow, Alaska)                         floyd@barrow.com

Floyd L. Davidson wrote:
>
>Steve at fivetrees wrote:
>
>>Floyd L. Davidson wrote...
>>>
>>>  [many lines of imaginative fiction deleted]
>>>
>>>>While your claims about grounding the telephone cable at every 1-2 km
>>>>is believable, it does not justify your claim that RS-485 cables can
>>>>be directly grounded (without current limiting resistors) at both ends
>>>>in all cases.
>>>
>>> Ah, we are making progress.  You now admit that yes in fact that
>>> *is* standard engineering practice.
>>>
>>> When you get a better grip on it, you understanding of how to
>>> engineer RS-485 systems will improve too.
>>
>>Floyd, you're being unnecessarily insulting. It's a little unseemly.
>
>What is insulting about that?  A little bit of "I told you so,
>right at the start of this lengthy conversation" is not
>inaccurate at this point, nor inappropriate.
>
>What you have posted, right from the start, was *far* more
>insulting!

A gentle reminder: two wrongs don't make a right.  Our behavior 
should be controlled by our own values, not by those of the person
we are replying to.

You should also keep in mind that it is a common occurance to see
one person criticized for being insulting while the person who
insulted him did far worse.  This is caused by the reader concluding
that one person is a reasonable fellow who sometimes gets carried 
away and starts posting insults that are beneath him, and that the 
other fellow is a lost cause - a total loser who will never respond
to any plea to moderate his bad behavior.  

It is my considered opinion that you are one of those reasonable 
fellows I just described, and that phrases like "many lines of 
imaginative fiction" and so forth are inappropriate.  I think 
you are better than that.

Just my opinion; take it for what it is worth.

Floyd L. Davidson wrote:

> Here they are *again*:
> 
>   http://www.nrao.edu/evla/geninfo/memoseries/evlamemo41.pdf

This first one seems to be about Fibre Optic Cable. This is not the topic 
of discussion. The thread is about RS485 data links and whether or not to 
connect the shield at end of the cable run to frame earth connections, and 
the precautions that may be needed if both ends really had to be connected.

 http://www.fac.ilstu.edu/Facilities_Services/Fac_Plan_Home/DesignGuidelines
/16743%20-%20Telecom%20Risers,%20Closets%20and%20Entries.htm

This one seems to be about telephone system risers in buildings and the 
practices that apply to that. Why should that be relevent to discussion 
about RS485 networking?, especially in industrial settings.

>   http://www.usda.gov/rus/telecom/publications/word_files/1751f810.doc

I am not able to read that on this system (I'll have to wait until I am 
using a different one). I only have the segment you provided below for now.

>   http://www.tequipment.net/WilcomTeleTransT124.asp

Why should I be interested in that continuity tester?

[%X]

A lot of people attribute any number of things to odd causes
> that often have no significance.  The fact that you've seen
> cables fried by ground faults should tell you something about
> ground faults.

Except that I am fully aware of the leaps in the difference potential 
between pieces of equipment that are not attributable to ground loops but 
are a fact of their normal and expected operation.

> As I've demonstrated *very* well, communications cable is
> grounded at regular intervals, including both ends.  I've
> discussed the theory involved to show why that is done.

Except that you keep going back to telephony/telegraph systems and straying 
away from this topic which is RS485 networks.

> So, tell me again who is doing the hand waving here?  And then
> tell me about 7 km loops on a ship again too.  And all about
> Santa while you are at it.

You still seem to be. Put your diagram up again and go through putting 
numbers up on that and show what happens to the energy in the system when 
the earth potentials differ. Until you do that I am leaving it lie there.

-- 
********************************************************************
Paul E. Bennett ....................<email://peb@amleth.demon.co.uk>
Forth based HIDECS Consultancy .....<http://www.amleth.demon.co.uk/>
Mob: +44 (0)7811-639972
Tel: +44 (0)1235-811095
Going Forth Safely ....EBA. http://www.electric-boat-association.org.uk/
********************************************************************

Guy Macon <_see.web.page_@_www.guymacon.com_> wrote:
>Floyd L. Davidson wrote:
>>Steve at fivetrees wrote:
>>>Floyd L. Davidson wrote...
>>>>
>>>>  [many lines of imaginative fiction deleted]
>>>>
>>>>>While your claims about grounding the telephone cable at every 1-2 km
>>>>>is believable, it does not justify your claim that RS-485 cables can
>>>>>be directly grounded (without current limiting resistors) at both ends
>>>>>in all cases.
>>>>
>>>> Ah, we are making progress.  You now admit that yes in fact that
>>>> *is* standard engineering practice.
>>>>
>>>> When you get a better grip on it, you understanding of how to
>>>> engineer RS-485 systems will improve too.
>>>
>>>Floyd, you're being unnecessarily insulting. It's a little unseemly.
>>
>>What is insulting about that?  A little bit of "I told you so,
>>right at the start of this lengthy conversation" is not
>>inaccurate at this point, nor inappropriate.
>>
>>What you have posted, right from the start, was *far* more
>>insulting!
>
>A gentle reminder: two wrongs don't make a right.  Our behavior

I don't plan to start posting either fiction or articles
filled with gratuitous insults.

If posting *facts* insults him, that isn't *my* doing.

...

>It is my considered opinion that you are one of those reasonable
>fellows I just described, and that phrases like "many lines of
>imaginative fiction" and so forth are inappropriate.  I think
>you are better than that.

You want me call it what it is?  I was being *diplomatic*!

-- 
Floyd L. Davidson           <http://web.newsguy.com/floyd_davidson>
Ukpeagvik (Barrow, Alaska)                         floyd@barrow.com

Floyd L. Davidson wrote:

>If posting *facts* insults him, that isn't *my* doing.

"get a better grip" and "many lines of imaginative fiction deleted"
are not facts.  They are mild insults.  

>I was being *diplomatic*!

All I can do is to express my considered opinion that using 
phrases such as "many lines of imaginative fiction deleted" 
is not an example of being diplomatic.

Floyd L. Davidson wrote:

>Or, you can skip all of that again, and just respond to this
>one quote (emphasis added):
>
>     3.2.4 Grounding of Cable Shields/Support Strands
>     (Messengers):  Normal construction practices and NESC
>     provisions require that cable *shields*, *messengers*, and
>     other *non-current carrying metallic hardware* be effectively
>     grounded.  It is especially important to effectively ground
>     cable *shields*, *messengers* and *non-current-carrying* metallic
>     hardware at *dead* *ends* and other *junction* *points* for noise
>     mitigation, personnel protection, and/or power contact
>     protection.
>         RUS Bulletin 1751F-815
>         http://www.usda.gov/rus/telecom/publications/

RUS Bulletin 1751F-815 concerns "aerial and buried cable."  
While I agree that this applies to outside plant telephone cable, 
I question whether it is appropriate for RS-485 digital cables,
which have a 40 foot distance limit at 10Mbps and a 4,000 foot 
distance limit at 100Kbps.  That's quite short for outside plant 
use.

RUS Bulletin 1751F-810 concerns "digital and lightwave 
telecommunications systems", which  would seem to be more
appropriate for RS-485 applications.  It says (emphasis added):

1.3  The *Single Point Grounding System*
described in this bulletin meets the protection requirements of most
central office and other electro-optic equipment manufacturers.  The
described methods should be followed unless there are compelling
reasons for change

2.13  Isolated Ground Zone (IGZ)
A dedicated area within an office building where all equipment is
*electrically insulated from all external grounds* except through *a
single ground connection* between the GWB and the MGB.  The isolated
area should preferably extend a minimum of six feet (two meters) on
all sides from the equipment frames and framework and where practical
be separated from other equipment by permanent walls.  The IGZ will
normally house sensitive electronic components (Isolated Area).

7.2  Transmission Equipment Termination and Protection
Digital carrier equipment and sensitive electronic pair gain systems
should normally be located inside the IGZ.

7.2.1.1  Shields of intraoffice cable connecting the MDF to carrier
equipment bays *should be open at the MDF end and grounded at one 
point* to the MGB or GWB.  This grounding arrangement provides 
electrostatic shielding and maintains GWB integrity.

2.9  Ground Loop
Ground loops exist when there is more than one electrical path to a
ground connection.  Such parallel paths to ground are normally not a
problem if associated with nonsensitive circuitry located outside the
Isolated Ground Zone (IGZ.)  *Ground loops are undesirable for
equipment located in the IGZ.*

2.21  Single Point Grounding
A grounding system using a *single point*, usually the MGB, for a zero
reference potential to ground for an entire system.  While the voltage
at this connection point may rise above zero volts-to-earth-ground
under fault conditions, the entire system will also rise at the same
rate to the same voltage.  This helps minimize any circulating
currents between components from lightning or power surges.

-- 
Guy Macon <http://www.guymacon.com/>

"Paul E. Bennett" <peb@amleth.demon.co.uk> wrote:
>Floyd L. Davidson wrote:
>
>> Here they are *again*:
>>
>>   http://www.nrao.edu/evla/geninfo/memoseries/evlamemo41.pdf
>
>This first one seems to be about Fibre Optic Cable. This is not the topic

*READ IT!*

The primary topic might well be Fiber, but it is all compared to
the copper cable they've had in place since the 1970's.  It
discusses a number of specifics, such as ground resistance, that
have been mentioned in this thread, including the fact that NEC
code *requires* the grounding you say shouldn't be done.

[drivel deleted]

> http://www.fac.ilstu.edu/Facilities_Services/Fac_Plan_Home/DesignGuidelines
>/16743%20-%20Telecom%20Risers,%20Closets%20and%20Entries.htm
>
>This one seems to be about telephone system risers in buildings and the
>practices that apply to that. Why should that be relevent to discussion
>about RS485 networking?, especially in industrial settings.

It discusses cable entrance specifications, and specifically how
the cable is ground at *each end* to the building ground.  The
grounding that you say doesn't need to be done, that they
specify according the NEC requirements.

>>   http://www.usda.gov/rus/telecom/publications/word_files/1751f810.doc
>
>I am not able to read that on this system (I'll have to wait until I am

Go to the web page I gave you previously

 http://www.usda.gov/rus/telecom/publications/

and download the pdf version then.  Why do you need this much
hand holding?

>using a different one). I only have the segment you provided below for now.

That is of course from an additional document, that you made no
attempt to check at all.  And we notice you aren't interested in
discussing the quoted text either.  Can't blame you for that
head in the sand posture...

>>   http://www.tequipment.net/WilcomTeleTransT124.asp
>
>Why should I be interested in that continuity tester?

Because if those splice boxes weren't grounded just like I said
they were, nobody would be interested in testing how well
grounded they are.  But, unlike yourself, others *are*
interested in good engineering practice.

>A lot of people attribute any number of things to odd causes
>> that often have no significance.  The fact that you've seen
>> cables fried by ground faults should tell you something about
>> ground faults.
>
>Except that I am fully aware of the leaps in the difference potential
>between pieces of equipment that are not attributable to ground loops but
>are a fact of their normal and expected operation.

You probably should *very* carefully read some of the other RUS
documents then.

>> As I've demonstrated *very* well, communications cable is
>> grounded at regular intervals, including both ends.  I've
>> discussed the theory involved to show why that is done.
>
>Except that you keep going back to telephony/telegraph systems and straying
>away from this topic which is RS485 networks.

I keep explaining comm cable engineering to you.  Why not try to
learn something that is useful instead of putting your head in
the sand with trivial comments like "Oh, my it's about fiber."
Or "Oh my, it's a continuity tester." etc. etc.

>> So, tell me again who is doing the hand waving here?  And then
>> tell me about 7 km loops on a ship again too.  And all about
>> Santa while you are at it.
>
>You still seem to be. Put your diagram up again and go through putting
>numbers up on that and show what happens to the energy in the system when
>the earth potentials differ. Until you do that I am leaving it lie there.

I've gone through everything necessary to understand the diagrams.
Use Google and *read* it.  If you won't actually read the material
how do you expect to understand it?

Here's the water, whether you drink or not is up to you...

-- 
Floyd L. Davidson           <http://web.newsguy.com/floyd_davidson>
Ukpeagvik (Barrow, Alaska)                         floyd@barrow.com

Guy Macon <_see.web.page_@_www.guymacon.com_> wrote:
>Floyd L. Davidson wrote:
>
>>If posting *facts* insults him, that isn't *my* doing.
>
>"get a better grip" and "many lines of imaginative fiction deleted"
>are not facts.  They are mild insults.

Guy, you've been on Usenet almost long enough to claim gray
whisker, eh?  Or, maybe not...  you probably don't realize that
before you were here, nobody cussed at each other on Usenet!
That certainly hasn't been the case for a lot more than any 10
years now.  Regardless, there is a *huge* difference between
gratuitous insults, and saying something pertinent to the posted
material that just also happens to be less than complimentary.

You initially admitted that abuse *is not* something I initiate.
It is something I've tolerated.

If you actually do have a problem with the words exchanged, why
weren't you chiming in with comments about 100 articles back?


  [excess drivel deleted]

...

-- 
Floyd L. Davidson           <http://web.newsguy.com/floyd_davidson>
Ukpeagvik (Barrow, Alaska)                         floyd@barrow.com

Guy Macon <http://www.guymacon.com/> wrote:
>Floyd L. Davidson wrote:
>
>>Or, you can skip all of that again, and just respond to this
>>one quote (emphasis added):
>>
>>     3.2.4 Grounding of Cable Shields/Support Strands
>>     (Messengers):  Normal construction practices and NESC
>>     provisions require that cable *shields*, *messengers*, and
>>     other *non-current carrying metallic hardware* be effectively
>>     grounded.  It is especially important to effectively ground
>>     cable *shields*, *messengers* and *non-current-carrying* metallic
>>     hardware at *dead* *ends* and other *junction* *points* for noise
>>     mitigation, personnel protection, and/or power contact
>>     protection.
>>         RUS Bulletin 1751F-815
>>         http://www.usda.gov/rus/telecom/publications/
>
>RUS Bulletin 1751F-815 concerns "aerial and buried cable."
>While I agree that this applies to outside plant telephone cable,
>I question whether it is appropriate for RS-485 digital cables,
>which have a 40 foot distance limit at 10Mbps and a 4,000 foot
>distance limit at 100Kbps.  That's quite short for outside plant
>use.

The point was, originally, that the flat statement that cables
should have the shield ground *only* at one end, is not correct.

When that was challenged, I pointed out that comm cables are *all*
grounded at 3000 or 6000 foot intervals.  The response was that it
is not true.

Obviously, it *is* true.

From that one can conclude that clearly the flat statement was
in fact incorrect, just as I originally stated.  The *reasons*
that it is not correct *do* apply to RS-485, and certainly to
anything that approaches 4000 feet (some claimed to be going
far beyond that too).  Specifically, it applies because in many
of the applications being discussed the *very* *same* reasons
to ground comm cables at multiple points exist for these short
RS-485 loops.

It's good engineering.

>RUS Bulletin 1751F-810 concerns "digital and lightwave
>telecommunications systems", which  would seem to be more
>appropriate for RS-485 applications.  It says (emphasis added):
>
>1.3  The *Single Point Grounding System*
>described in this bulletin meets the protection requirements of most
>central office and other electro-optic equipment manufacturers.  The
>described methods should be followed unless there are compelling
>reasons for change

I have discussed that in *detail*, in several articles.

>2.13  Isolated Ground Zone (IGZ)
>A dedicated area within an office building where all equipment is
>*electrically insulated from all external grounds* except through *a
>single ground connection* between the GWB and the MGB.  The isolated

And the specifics of that single ground connection, are what
I discussed in detail.

>area should preferably extend a minimum of six feet (two meters) on
>all sides from the equipment frames and framework and where practical
>be separated from other equipment by permanent walls.  The IGZ will
>normally house sensitive electronic components (Isolated Area).
>
>7.2  Transmission Equipment Termination and Protection
>Digital carrier equipment and sensitive electronic pair gain systems
>should normally be located inside the IGZ.
>
>7.2.1.1  Shields of intraoffice cable connecting the MDF to carrier
>equipment bays *should be open at the MDF end and grounded at one
>point* to the MGB or GWB.  This grounding arrangement provides
>electrostatic shielding and maintains GWB integrity.

I discussed *that* in detail.  You do understand what
"intraoffice cable" is, right?  An how that fits what I said
about the *single* ground connection to a building ground?

>2.9  Ground Loop
>Ground loops exist when there is more than one electrical path to a
>ground connection.  Such parallel paths to ground are normally not a
>problem if associated with nonsensitive circuitry located outside the
>Isolated Ground Zone (IGZ.)  *Ground loops are undesirable for
>equipment located in the IGZ.*

So just how many of these 4000 foot, or longer, RS-485 runs are
totally contained within the IGZ?  For that matter, how many of
the runs that have been described in this thread have been in an
IGZ, regardless of the length?

>2.21  Single Point Grounding
>A grounding system using a *single point*, usually the MGB, for a zero
>reference potential to ground for an entire system.  While the voltage
>at this connection point may rise above zero volts-to-earth-ground
>under fault conditions, the entire system will also rise at the same
>rate to the same voltage.  This helps minimize any circulating
>currents between components from lightning or power surges.

You can't cherry pick things that do not apply, without
understanding the terms and the limitations being discussed.

As I've described...  each building should have a single ground
point, and each floor should also have a single ground point.
Generally each *row of equipment* has a single ground, though
sometimes individual racks are isolated.

Within a single row of equipment, or within a single rack, that
has a single ground connection, you have an IGZ.  There should
not be two grounds on each end of a shielded cable.

I'll leave the rest as an exercise for the reader...

-- 
Floyd L. Davidson           <http://web.newsguy.com/floyd_davidson>
Ukpeagvik (Barrow, Alaska)                         floyd@barrow.com