Talk:RS-485/Archive 1

connectors used with EIA-485

Is there a standard pinout for putting EIA-485 on a RJ11 or RJ45 connector? (Does 10BASE-T specify this, or are the signals on a 10BASE-T cable *not* EIA-485 signals?)

The article currently says "The following table lists some typical RS-485 signal pin assignments" then literally lists the pinouts. That is good information, but is there a name for these pinouts? Is there a name for the "EIA-485 on a DE-9 connector" standard?

I would like to fill in the holes in this table:

(table moved to Talk:Physical layer).

The article says EIA-485 is just the signalling of bits, which means talking about the pinout of EIA-485, even typical pinouts of EIA-485, is nonsense. The author of the "connectors" section probably thought as I did until today that RS-485 referred to that RS-232-style character stream protocol. If not, I hope there is some other name for that protocol, and an article on that is where the connectors section and the pinouts belong.

Bryan Henderson 18:29, 13 September 2006 (UTC)

NMEA compatibility

The article currently states that NMEA signal naming is different from the RS483 standard (A- / B+). That is not correct. The NMEA standard says that " ... idle, marking, logical 1, OFF or stop bit states are defined by a negative voltage on line A with respect to line B" (in other words: A is the inverting signal); this is identical to EIA-485 as quoted above by Steve Fairhead. I also verified that NMEA V2.0, V3.0 and the corresponding IEC61162/1 are identical on this point. Heje/87.48.134.234 14:40, 17 October 2006 (UTC)

I have updated the article accordingly. I also added some details about the EIA standard naming (and removed the mentioning of 5V - that is a arbitrary voltage and not a general value; I'm using 3V drivers myself and they give me 0.4V and 2.6V when terminated). Heje/87.48.134.234 15:31, 17 October 2006 (UTC)

External links

I removed the external links to Modbus sites since those links belong on the Modbus page. I also removed some commercial links to companies providing EIA-485 products since it was just general advertising for them unlike the TI and Maxim IC links which go to specific information about EIA-485 and not their own products. Littleman TAMU 20:33, 5 February 2007 (UTC)

I replaced on of the modbus links you removed with a RS-485 specific page from the same site, as it provides useful practical information to implement RS485. Please remove it if you see no fit for this link. 88.204.248.235 18:18, 1 March 2007 (UTC)

That is a useful link and I've no problem with it. Thanks for adding it. The only reason I removed the others was that they were primarily about Modbus and only mentioned that Modbus could run on EIA-485.--Littleman_TAMU ^(talk) 00:58, 2 March 2007 (UTC)

Confusion

The tables "equating" RS 485 pins with RS 232 pins is completely confusing and meaningless - can you *wire* a working connector from this table? You can' tinterconnect 485 and 232 levels as this mish-mash seems to indicate. I'm pulling it out. --Wtshymanski (talk) 18:32, 30 December 2007 (UTC)

TIA or EIA?

Why is this standard referred to as EIA-485 when the document is called TIA-485-A? There doesn't seem to be any consistancy between any of the documents controlled by EIA/TIA, some seem to be common as prefixed by TIA and others as EIA. This one I thought was a bit more clear cut since the standard title doesn't have both like most do.206.169.227.226 (talk) 21:51, 5 February 2009 (UTC)

DMX over microphone cables

The "Applications" section appears to suggest that DMX-512 can be run over standard microphone cables. This is a very bad idea: the impedance of a standard microphone cable is way too low and degrades the DMX signal to the point where random errors become common. These are very hard to diagnose for anyone unfamiliar with the problem. It is possible to run microphone signals over 120 Ohm cable (Van Damme market an "AES/EBU Microphone cable" which is 120 Ohm and performs well in either role) but the converse is not to be recommended. I believe that the whole sentence should be replaced with a description of DMX which does not mention "standard microphone cables". I am happy to make the edit if there are no objections. —Preceding unsigned comment added by 80.176.190.231 (talk) 20:24, 25 August 2009 (UTC)

The DMX512-A article covers the wiring issues so I have deleted them here. -Tim D. (talk) 01:30, 22 October 2009 (UTC)

Lack of a basic summary

This article doesn't even explain what EIA-485 is! —Preceding unsigned comment added by 24.18.242.170 (talk) 04:14, 8 May 2009 (UTC)

Agreed - I visited to get an overview and a bit of history - I'm not clear how old the standard is or when it was invented, by whom, etc. - is it an 'old' outdated standard, or is it very much currently used? Something like the RS-232 page has would be good, although it doesn't have to be long - thanks --SteveBagpuss (talk) 08:33, 3 March 2011 (UTC)

Discusses the limitations of RS485

I'd like to see some comparison, like RS485 vs. RS232 vs. Ethernet, or something similar, so that it becomes apparent what RS485 does not implement. E.g. it does NOT implement flow control, collition avoidance, collision detection, or detection of corrupted data with checksums. I've spent the last few weeks working with people, that were obviously trying to use RS485 half-duplex as a full-duplex connection, caiming that the RS485 would wait until the wire is free when sending data. IMHO, it is worth mentioned that RS485 does not come with all fency mechanisms working in the background, that we're used to from using RS232, Ethernet, etc. 31.16.78.3 (talk) 14:17, 1 May 2012 (UTC)

Well, that's a bit like saying ASCII doesn't include spelling check. It's a signalling standard at the very lowest physical level of the OSI model - it realy only defines what "1" and "0" look like on the wire, everything else is for higher layers to resolve, outside the scope of the standard. --Wtshymanski (talk) 16:16, 1 May 2012 (UTC)

Buring of RS485

I am working on Canal ans Dam Automation Project (NSP)in Venukonda (A.P)India.I our project RS485 used in actuator card for gate opration i.e up,down and stop command from microcontrollar DASTU. but many time burn the 485 and communication is faulty. nothing any command to gate operation. I am trying all possibilitys.but still same promble is comming.

So please give me some solution to working the system satisfactory. —Preceding unsigned comment added by 59.99.17.20 (talk) 07:43, 10 October 2009 (UTC)

Sorry you're having difficulties, but this isn't the place for that kind of help. This talk page is for discussion relating to improvement of the article. Try http://answer.yahoo.com or http://tech.groups.yahoo.com/group/industrial_networking/ or one of the electronics newsgroups. —EncMstr (talk) 15:55, 10 October 2009 (UTC)

Use a real BUSz u n00b. RS-485 iz old and busted. 68.144.191.97 (talk) 00:33, 18 July 2012 (UTC)

Confusion in A/B labelling

I have a problem with this article: The Waveform Example gets the polarity naming convention exactly wrong, and helps confuse people. Here is a good treatment of the confusing polarity issue: http://www.bb-europe.com/tech_articles/polarities_for_differential_pair_signals.asp -dgi, 12 June 2006

Can anyone provide a direct citation to the standard? I think the articles on the net like this one are just 50:50 stating that or the other. Petr Matas 23:53, 7 September 2006 (UTC)

EIA-485 spec, section 3.2:

"a) The A terminal of the generator shall be negative with respect to the B terminal for a binary 1 (MARK or OFF) state.

b) The A terminal of the generator shall be positive with respect to the B terminal for a binary 0 (SPACE or ON) state."

It's important to note that the MARK state is the rest state, i.e. the input to a transceiver from an inactive UART will be a logic high, i.e. a 1. I concur completely that the transceivers from the 75176A onwards are incorrectly marked. I am unable to find anything within the spec that supports the contention that the "A line can be designated by a '-' and the B by a '+'". (Steve Fairhead, sfdesign.co.uk) 195.112.48.103 18:31, 9 September 2006 (UTC)

The standard TIA-485-A (Reaffirmed March 28, 2003) does not include the terms MARK, SPACE, Inverting, Non-Inverting. It does include + and - but only in reference to voltage levels, not in reference to the A and B lines.

The Actual wording of section 4.0 (English version)

a) The A terminal of the generator shall be negative with respect to the B terminal for a binary 1 (OFF) state.

b) The A terminal of the generator shall be positive with respect to the B terminal for a binary 0 (ON) state.

The logic function of the generator and the receiver is beyond the scope of this Standard, and therefore is not defined.

Any reference to these terms in the article should clearly state that they come from documents other than the TIA-485-A standard.

The above statement (Section 4.0) is very clear that the specification does not include the "logic function". It only specifies the physical layer.

The scope (Section 1) is also clear

This Standard does not specify other characteristics, such as signal quality, timing, protocol, pin assignments, power supply voltage, operating temperature range, etc., that are essential for proper operation of interconnected equipment. Any devices complying with this Standard shall do so within the ranges of those factors appropriate for the device operation, such as power supply voltages, and ambient temperature. It is intended that this Standard be referenced by other standards and specifications that specify the additional characteristics necessary to assure satisfactory interoperation of equipment.

Since most '485 implementations drive the '485 tranceiver with an EIA-232 tranceiver or from '232 USART logic, too many people assume that the MARK/SPACE function of '232 apply to the '485 spec.

If "A" is more positive than "B" (within the voltage limits and termination impedance stated in the spec) then the signal is a binary 0. That's all the spec says.

EE JRW (talk) 16:50, 4 March 2013 (UTC)

The generator symbols in the standard use a diff amp triangle with the A terminal straight and the B terminal with the inverting bubble, which seems to go against the inverting/non-inverting convention stated in the article. However, A<B=1 would make the A terminal inverting in my understanding. I, too, didn't find reference to the words "inverting" or "non-inverting" in the standard. My copy of the spec (ANSI/TIA/EIA-485-A-1998, affirmed 3/3/98, reaffirmed 3/28/03) has Fairhead's quotation in section 4, and it goes on to say that the logic function is not defined.

As to whether some company's pin definition is correct or not, if A<B=1(OFF) and B<A=0(ON) then it's correct, regardless of which pin you call inverting. —Preceding unsigned comment added by 149.32.192.33 (talk) 22:30, 20 July 2009 (UTC)

What is in the standard vs. application information

The standard has a very limited scope. Much of what is in the wiki article is from applications information/guides or just practical information. I have no problem with including practical information in the article, but I do have a big problem with implying that this information is in the standard.

My original edit added a section with information as to what was actually in the standard. I intended to use the new section as a springboard to correct other information in the article. Unfortunately the new section was immediately edited so severely that it no longer has information about what is in the standard.

When a quote from the standard of:

"The logic function of the generator and the receiver is beyond the scope of this Standard, and therefore is not defined."

is changed to:

The standard does not define any logic function to the two states.

It simply confuses the reader as to what is in the standard.

I would be happy to work with an experienced Wikipedian to bring this article up to Wikipedia standards, but when an exact quote from the standard is reworded to say something that the standard does not say, ... well there is no point in my wasting time trying correct this article when is is going to be promptly incorrected.

As stated in the the standard "This Standard specifies the electrical characteristics of generators and receivers ..." and "This Standard does not specify other characteristics, such as signal quality, timing, protocol, pin assignments, power supply voltage, operating temperature range, etc., that are essential for proper operation of interconnected equipment. "

I am open on how to present that the standard only has the following:

Section 1 - SCOPE

"This Standard does not specify other characteristics, such as signal quality, timing, protocol, pin assignments, power supply voltage, operating temperature range, etc., that are essential for proper operation of interconnected equipment."

Section 2 - DEFINITIONS, SYMBOLS AND ABBREVIATIONS

Section 3 - APPLICABILITY

Section 4 - ELECTRICAL CHARACTERISTICS

This section and sub-sections define the electrical characteristics of the generator (transmitter or driver), receiver, transceiver, and system. These characteristics include: definition of a unit load, voltage ranges, open circuit voltages, thresholds, and transient tolerance.

It also defines three generator interface points (signal lines); "A", "B" and "C". The data is transmitted on "A" and "B". "C" is a ground reference.

"The signaling sense of the voltages appearing across the interconnecting cable are defined as follows and is shown in figure 2" (figure 2 of the specification is not shown in this article)

"The A terminal of the generator shall be negative with respect to the B terminal for a binary 1 (OFF) state."

"The A terminal of the generator shall be positive with respect to the B terminal for a binary 0 (ON) state."

"The logic function of the generator and the receiver is beyond the scope of this Standard, and therefore is not defined."

Annex A - (informative)

I am also open on how to present that the following is typical application information, and is not part of the standard:

Termination resistor values and biasing

Master/Slave

Full Duplex (Actually this section is confused and is really a part of EIA-422)

Everything in the Applications section

All of the pin labeling (see also "Confusion in A/B labelling" in this talk)

The waveform example (includes protocol which is required for communications to function but is not part of the standard)

EE JRW (talk) 14:09, 8 March 2013 (UTC)

RS vs. EIA vs. TIA

RS is an obsolete reference and is no longer used by EIA / TIA http://eetimes.com/electronics-news/4163492/Trim-the-fat-off-RS-485-designs

The TIA-485-A standard (Reaffirmed: March 28, 2003) cover page has the following:

[ANSI logo]

ANSI/TIA/EIA-485-A-1998

Approved: March 3, 1998

Reaffirmed: March 28, 2003

TIA STANDARD

Electrical Characteristics of Generators and Receivers for Use in Balanced Digital Multipoint Systems

TIA-485-A

(Revision of EIA-485)

MARCH 1998

RS does not appear anywhere on the cover page. RS-485 appears once in the forward of the TIA-485 standard. In the forward it states:

No technical changes have been incorporated into TIA/EIA-485-A which will create compatibility problems with equipment conforming to previous versions of this Standard (EIA RS-485).

The title of the wiki article is RS-485 which starts the confusion. Further use of RS in the article causes more confusion.

A note (and citation) has been added to the article that RS is obsolete. Using RS-485 in the article simply confuses the issue further. All of the Wiki articles that are titled RS-XXX are incorrectly titled.

I will not argue with the use of TIA or EIA or TIA/EIA, but I do have a problem with the use of RS.

Radiojon seems to have tried to get the title of this article corrected in 2005 with no success. I am hoping that at least we can keep the obsolete "RS" out of the article itself.

Well, we can bring the article up to the bleeding edge of whatever the last whim of TIA was to call it, but there's no way to update the (hundreds of thousands? millions?) of references out there to the previous titles. Google Books tells me it can find "RS 485" 26,000 times, "EIA 485" 1430 times, "TIA 485" 125 times and "EIA/TIA 485" a lonely 65 times, beat out by "EIA RS 485" which shows up 678 times. The article title should probably remain at "RS 485" with redirects from all the various other permutations and combinations. It's usual practice in a Wikipedia article (as much as we have "usual" practices) to pick one name for the subject of the article and use it throughout. --Wtshymanski (talk) 15:52, 8 March 2013 (UTC)

I'm not talking about correcting references, just the article. I think that it would be best to use the reference that is on the standard, but an engineer is going to know what RS-485 means. The EE Times article is "Trim the fat off RS-485 designs" and uses RS-485 throughout, but then has the information that RS is no longer correct. The Wikipedia article uses RS-485, EIA-485 and TIA-485. I think it would be better to use just one. The question is which one? You said Wikipedia articles aren't supposed to talk about themselves. So if the official reference is not used, then how do you tell the reader that the official standard is TIA-485-A, but this article is going to call it RS-485 or EIA-485?

I know that not everyone has a subscription to the latest edition of the standards. I don't either, but I do happen to have the latest version of this standard. That's why I am quoting from the standard so that people know what is in it and what is not. --EE JRW (talk) 20:03, 8 March 2013 (UTC)

In my opinion, let's call it "RS 485" throughout, which is the article title, and in the lead or early on mention that, like so many other standards, the document has been renamed a few times and the current edition is called "TIA 485" in the current title and designation. Standards wonks already know that the prefix changes, and normal people should find it less confusing with only one term used throughout.

You've elsewhere pointed out a more serious problem: factoring out what the standard actually *says*, with the application folklore and legends that have grown up around it. If the bias network is not in the literal text of the TIA standard, it should be moved down to "applications", or something like that.

It would be nice to get some history - even issue dates of the standard revisions would be some help. --Wtshymanski (talk) 22:31, 8 March 2013 (UTC)

All right, you have convinced me. Even though it is technically "TIA-485", the more common usage is "RS-485" (the dash is silent). I'll make a first attempt at updating the article to reflect why RS-485 is being used.

I also would like to get some history into the article. Some of the comments below also ask for it. I have done a lot of googleing and have finally found some information. I'm not guaranteeing the accuracy, but it's probably pretty close: http://www.softelectro.ru/rs485_en.html

I also ran across what I consider to be one of the better articles about RS-485 (and yes its title has RS not EIA or TIA) Its a little old so it does not mention the reaffirmation in 2003. But the standard did not change in 2003, it was just reaffirmed. Strongly recommended reading for anyone who is editing this Wikipedia article. One note about the Circuit Cellar article, it discusses line length. Line length is not part of the 485 standard but it is definitely part of the application of 485. http://www.rcsystemsco.com/Download/rs485.pdf
EE JRW (talk) 00:09, 10 March 2013 (UTC)

Dead reference

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

---

The link to "Application Note 847 FAILSAFE Biasing of Differential Buses" (www.national.com/an/AN/AN-847.pdf) is dead. That link should go to the TI website (www.ti.com/lit/an/snla031/snla031.pdf) F01x2 (talk) 19:57, 30 May 2014 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Star unavoidable with PTZ cameras?

The article claims, "Star and ring topologies are not recommended because of signal reflections or excessively low or high termination impedance. If a star configuration is unavoidable, such as when controlling multiple pan tilt zoom cameras from a central video surveillance hub..."

This makes no sense. There is nothing inherently unavoidable about using a star topology for controlling PTZ cameras. What is wrong with having a daisy-chained bus from camera-to-camera and to the hub? There may certainly be convenience or wiring issues, but I don't see any way they can be "unavoidable." I am dubious that there is any way this can be correct; so marked. jhawkinson (talk) 13:09, 8 September 2014 (UTC)

RS-485 does not talk about topology. The closest it comes to topology is figure 1 which shows a transmission line with a transmitter, receiver, and transceiver connected to the transmission line via stubs. It does state: "Maximum data signaling rate is typically limited by ..., maximum allowable stub length,...", but it does not define the maximum allowable stub length. There is nothing about PTZ cameras, hubs, signal reflections or excessively low or high termination impedance.

RS-485 Annex A states: The multipoint system should be configured in the form of a daisy chain. Star, tree, or branch configurations are generally not recommended."

The forward in RS-485 states: "Annex A of this Standard is informative and provides a minimum set of guidelines for application, and is not considered part of this Standard." So technically the "not recommended" is not part of the standard. Unfortunately adding a bunch of weasel words saying that "some pages at the end of RS-485, but are not part of RS-485 say that star and ring topologies are not recommended" is rather awkward. The forward also states "The Telecommunications Systems Bulletin TSB-89 contains application guidelines on TIA/EIA-485-A. Topics covered include: data signaling rate vs. cable length, stub length, configurations, and other important considerations."

I personally have no problem with including application information in this article, but I would like to see it clearly stated that the information is application based and not from RS-485. That would help to reduce the folklore surrounding RS-485. There is already too much application information (not to mention folklore like the full-duplex section) in this article which is implied as being part of RS-485 but is not anywhere in the standard.

There may not be anything inherently unavoidable about using a star topology, but there are certainly circumstances where daisy-chaining is not practical, and converting to a daisy-chained network could potentially break a functioning network. PicList Thread: "Reflections in RS485 PIC Network?" 1998.

My suggestion would be to simply state "The network should be configured in the form of a daisy chain. Star and ring topologies are not recommended. The forward in RS-485 recommends TSB-89 for application guidelines."

If you heart is set on mentioning hubs, you could add a section discussing specialized equipment that can be used on a RS-485 network, but is not part of the standard. EE JRW (talk) 00:59, 13 September 2014 (UTC)

First off, this is mostly moot because of this edit pair led off by User:Wtshymanski.

I think you might misunderstand my point, which was the article suggested star topologies were impractical for PTZ cameras, and that doesn't make any sense to me. I think it's appropriate for the article to explain star topologies are discouraged for RS-485, and that it's reasonable to give examples where star topologies might be more convenient. But it should not stay that star topologies are "unavoidable" for some application unless they truly are. And that's not true for PTZ cameras.

It sounds like you should cite the Annex next to the claim if you like. Since you have the reference that should be easy for you. jhawkinson (talk) 02:00, 13 September 2014 (UTC)

I did see the edits you refer to, but there was nothing on this talk page so I decided to add some discussion about what is actually in the standard.

You're correct. I did not understand that your only concern was the article suggested star topologies were unavoidable for PTZ cameras. I think you also misunderstood my point. RS-485 does not discuss topologies, (the annex does, discussed above). Nor does it include star/hub repeaters (or bit rate limits or line length limits, etc.) Yet the article's overview includes all of this information as if it were part of the standard. My thought is that the RS-485 article should be about the RS-485 standard. Application information could be added in a section(s) where it was made clear that the information in them is not part of the standard.

I started working on this article over a year ago with the intention of making it about the standard and removing folklore and false information, but gave up after less than a week. If you are interested in why, just read this talk page. My last edit to the article added historical information. It was promptly reverted because the format wasn't acceptable ("should be in prose, not bullet points"). I decided at that point that it was more productive to sit outside and watch the grass grow than to try fixing this article. So I am not planning to add anything to it.EE JRW (talk) 22:06, 14 September 2014 (UTC)

For what it's worth (not much), I would disagree somewhat strongly. I do not think the article should be about the standard -- I think it should be about the use and applicability of the standard. So, to me, any accurate information that is pertinent (hopefully well-sourced) is welcome. If the article were just scoped to the standard, rather than its utility, well, we might as well just tell readers to read the standard itself (except its not free). But please feel free to disagree, I don't mean to force my opinion on you. Thanks for the explanation on your editing motivations. jhawkinson (talk) 22:26, 14 September 2014 (UTC)

A+ or B+?

This is messy. As far as I know the standard says that A is - and B is + but it is paywalled and presumably little read so I cannot verify this. All the chip manufacturers say A+ B- which is probably wrong according to the standard but given that the standard is obscured and that all the chip manufacturers say this then which is right? What defines a standard? The obscured official documentation or what 99% of datasheets say? IMHO I would be tempted to say that the standard has escaped to public domain and is now A+ B- but I admit that this is way beyond the remit of Wikipedia to define. Mtpaley (talk) 21:24, 29 June 2015 (UTC)

This isn't messy, it's misunderstood. This issue has been discussed elsewhere on this page see #Confusion in A/B labelling. In addition there is a pretty good writeup of this issue on Wikibooks (IMBO) [That Pesky] Polarity:RS-485 Technical Manual.

In the interest of putting this to bed, let me try one more time to state what RS-485 says and does not say about this.

RS-485 does have "+" and "-" in the standard, but these characters are only used to specify positive and negative voltage. They are not used to define the polarity of the A and B pins of the driver. The terms "inverting" and "non-inverting" are not in RS-485. The standard states; "The logic function of the generator and the receiver is beyond the scope of this Standard, and therefore is not defined."

203.206.162.148 made edits changing the statement from "This is in conflict with the A/B naming..." to "This is in accordance with the A/B naming..." and added reference 6 "Polarity conventions" (PDF). to a TI application note. The reference is good. Neither the original statement nor the new one are correct.

Open the TI reference document and hold on to your seat while I make one more attempt to explain this.

The TI document includes a chunk of RS-485 in the blue box. This chunk (figure 2) is word for word and diagram and symbol exactly what is in RS-485 with one mistake. The second statement "b)" should say "for a binary 0 (ON) state" (just like it does in the text above the excerpt from RS-485, and in the excerpts' signaling diagram). You do not have to pay to get RS-485, you've already got all that RS-485 says about it. It does not have + or -. It does not have the terms inverting or non-inverting. It shows the a driver symbol with an inverting and non-inverting output, but does not include the input to the driver. Then it states when A is negative with respect to B, a binary 1 is on the bus. Note that the standard states "The logic function of the generator and the receiver is beyond the scope of this Standard, and therefore is not defined." This means that if your driver gets it's signal from a UART, the signal may or may not be inverted before it is output on the RS-485 bus. Also note that if the driver symbol is driven with a 1, you would expect that A would be positive with respect to B for the bus state, exactly opposite of what the standard shows in the signaling diagram. Finally TI defines their convention of "a logic HIGH on the driver input “D” will create an “ON” state, also called a 0". This means that (for TI) a logic 1 into their driver will result in a RS-485 bus state of binary 0. TI is defining the "logic function" of their driver and receiver. But, it does not matter whether they define the logic function as inverting or non-inverting, both ways meet the RS-485 standard.

I originally found this RS-485 Wikipedia page when I was desperately trying to figure out if my circuit had the A and B connections swapped. The circuit had a UART pin from a microcontroller connected to the RS-485 driver, and I connected pin A of the driver to the "A" wire of the RS-485 bus. This worked fine until I switched to a different "RS-232 to RS-485" device. I ended up calling Linear Technology (who's 485 driver I was using) and asking for help. I was told that I needed to speak to their "RS-485 guru" and he would call me back. When he called, I started by telling him which IC I was using and pointed out the RS-485 figure 2 and said something like ".. your IC will output a high on pin A, but RS-485 shows ..." and he cut me off and said; "Why do you thing that pin A of our IC has anything to do with the connection A of RS-485." After I picked my jaw off of my desk and reinserted it in my mouth, I babbled incoherent words for a minute while he continued on trying to tell me what my brain could not accept. RS-485 does not define the logic function of the driver or receiver. It took me re-reading the standard over-and-over (and an embarrassing amount of time) to actually get this through my head. I hope this discussion helps others to understand in a shorter time.

It would be nice to see the article updated with a synopsis of this discussion, but as I stated previously, I am not planning to add anything to it.

EE JRW (talk) 01:22, 1 July 2015 (UTC)

Edit:Try to fix bad linkEE JRW (talk) 01:30, 1 July 2015 (UTC)

What does it LOOK like?

Some people just want to know the connector geometry; (which of all these ports is the RS-485?) It's great to have all the technical stuff, but remember to show a picture of the physical port. 173.164.209.28 (talk) 18:26, 4 April 2019 (UTC)

There is no defined connector in the standard, so it could look like anything from screw terminals to a mini-DIN plug to ...whatever? --Wtshymanski (talk) 18:51, 4 April 2019 (UTC)

Release date

The article suggests RS485 was first released in 1998, however Modbus uses RS485 and was released in 1979 so there must have been an earlier revision of the standard. I've seen references to the early 1980s and some to the 1960s (when RS232 was published) so I'm having trouble finding out when the first version of RS485 was actually made available. Does anyone know when the standard was first released? -- Malvineous (talk) 03:06, 15 September 2019 (UTC)

I think it's at least a little older than that. I have a book called "Telebyte Technology Data Commnciations Standards Library" which contains reprints of several EIA and ITU standards. The title page for EIA standard RS 485 is dated April 1983, and the text within doesn't indicate any prior issues. --Wtshymanski (talk) 23:31, 15 September 2019 (UTC)