RJ45, RJ12, RJ14 cables, straight-through or crossover?

W

Winfield Hill

Guest
RJ45 network cables come in straight-through, and
in crossover versions. Does straight-through in a
patch cable mean that pin 1 in goes to pin 1 out?

What about everyday telephone cables? A flat cable,
with wires going straight into the plug on each end.
Physically, it looks like the very definition of
straight-through. But does pin 1 in go to pin 1 out?
NO!! Since the two plugs face in opposite directions,
pin 1 goes to pin 4, and etc. Pin 1 to pin 1, would
require a wire crossover or a 180-degree cable twist.

What about the straight-through RJ45 network cables?
Yes! These DO have pin 1 in going to pin 1 out. And
it's the crossover type that has pin 1 to pin 8, etc.

I thought about this at 3 AM, looked at a telephone
cable, and changed the PCB layout for my beehive's
five-sensor I2C jacks. But wait, are RJ11 RJ14 6P4C
phone and RJ12 6P5C data cables different? Checking,
Yes! An RJ11 voice cable can worked reversed, but
an RJ12 data straight-through is pin 1 to pin 1.
Some sellers of RJ11 RJ14 cables let you specify to
get straight pin 1 to pin 1, or "reversed".

Sheesh, changing my PCB jack wiring back again.


--
Thanks,
- Win
 
Winfield Hill <hill@rowland.harvard.edu> wrote in
news:qb9b4s01blb@drn.newsguy.com:


Sheesh, changing my PCB jack wiring back again.
Click to expand...

I found and (then) posted this:

<http://www.mediafire.com/file/pkcz36j3nd1ipi9/wiring-diagrams-lan-
cable-cat5e-ends-rj45-connector-within-lively-for-ethernet-
diagram.jpg/file>
 
On 12 May 2019 07:38:20 -0700, Winfield Hill
<hill@rowland.harvard.edu> wrote:

RJ45 network cables come in straight-through, and
in crossover versions. Does straight-through in a
patch cable mean that pin 1 in goes to pin 1 out?

What about everyday telephone cables? A flat cable,
with wires going straight into the plug on each end.
Physically, it looks like the very definition of
straight-through. But does pin 1 in go to pin 1 out?
NO!! Since the two plugs face in opposite directions,
pin 1 goes to pin 4, and etc. Pin 1 to pin 1, would
require a wire crossover or a 180-degree cable twist.

What about the straight-through RJ45 network cables?
Yes! These DO have pin 1 in going to pin 1 out. And
it's the crossover type that has pin 1 to pin 8, etc.

I thought about this at 3 AM, looked at a telephone
cable, and changed the PCB layout for my beehive's
five-sensor I2C jacks. But wait, are RJ11 RJ14 6P4C
phone and RJ12 6P5C data cables different? Checking,
Yes! An RJ11 voice cable can worked reversed, but
an RJ12 data straight-through is pin 1 to pin 1.
Some sellers of RJ11 RJ14 cables let you specify to
get straight pin 1 to pin 1, or "reversed".

Sheesh, changing my PCB jack wiring back again.
Click to expand...

Most ethernet/RJ45 gadgets nowadays are agnostic to the crossover
situation; they work either way.

In non-ethernet situations, like RS485 or such, I buy non-crossover
RJ45 CATx cables, pin1 to pin1, just like most cable assemblies. I
find that easier to understand on schematics.

I also find that I discover things at 3AM, or sometimes later in the
shower.


--

John Larkin Highland Technology, Inc

lunatic fringe electronics
 
Winfield Hill <hill@rowland.harvard.edu> wrote in
news:qb9b4s01blb@drn.newsguy.com:

RJ45 network cables come in straight-through, and
in crossover versions. Does straight-through in a
patch cable mean that pin 1 in goes to pin 1 out?
Click to expand...

Most ports have self negotiation Hdw at one end or the other or both
and they auto handshake.

But to answer you query:

Googled "Gb ethernet wiring" and the very first picture has all your
answers.

<https://www.google.com/search?q=Gb+ethernet+wiring>
 
DecadentLinuxUserNumeroUno@decadence.org wrote...
Winfield Hill wrote:

Sheesh, changing my PCB jack wiring back again.

I found and (then) posted this:

http://www.mediafire.com/file/pkcz36j3nd1ipi9/wiring-diagrams-lan-
cable-cat5e-ends-rj45-connector-within-lively-for-ethernet-
diagram.jpg/file
Click to expand...

Yes, thanks.

I shouldn't have ended my first sentence with a question mark.
Everyone knows about RJ45 pinouts, it's the 4 and 6-pin voice
cables and 6-pin data cables that are obscure. And I made
the mistake of actually looking at one for the information.


--
Thanks,
- Win
 
Winfield Hill <hill@rowland.harvard.edu> wrote:
DecadentLinuxUserNumeroUno@decadence.org wrote...

Winfield Hill wrote:

Sheesh, changing my PCB jack wiring back again.

I found and (then) posted this:

http://www.mediafire.com/file/pkcz36j3nd1ipi9/wiring-diagrams-lan-
cable-cat5e-ends-rj45-connector-within-lively-for-ethernet-
diagram.jpg/file

Yes, thanks.

I shouldn't have ended my first sentence with a question mark.
Everyone knows about RJ45 pinouts, it's the 4 and 6-pin voice
cables and 6-pin data cables that are obscure. And I made
the mistake of actually looking at one for the information.
Click to expand...

As you say, everyone knows about RJ45. Nonetheless allow me to note the
Cisco RJ45 RollOver cable's elegant eccentricity. [1]
FWIW, on premise, it's my practice to always crossover the Tip and
Ring in each pair of a 4-pin voice cable. Although my method may be
superfluous in this world of mostly agnostic data ports, it's one less
point to investigate after a failure.

Note.

1. http://pinouts.ru/NetworkCables/rj45_rollover_pinout.shtml

Thank you, 73,

--
Don Kuenz KB7RPU
There was a young lady named Bright Whose speed was far faster than light;
She set out one day In a relative way And returned on the previous night.
 
Ethernet cross-overs swapped pairs.

Voice cables by default swapped polarity within the pair.

And 8-conductor telephone cables were a crap-shoot as to which
way they were built, adding to the fun.

Telco cables cared about polarity so the (first generation) DTMF
pads would work. I wonder if it even matters if an Ethernet
cable has backwards polarity; not sure.


--
A host is a host from coast to coast.................wb8foz@nrk.com
& no one will talk to a host that's close..........................
Unless the host (that isn't close).........................pob 1433
is busy, hung or dead....................................20915-1433
 
On 12 May 2019 07:38:20 -0700, Winfield Hill
<hill@rowland.harvard.edu> wrote:

I thought about this at 3 AM, looked at a telephone
cable, and changed the PCB layout for my beehive's
five-sensor I2C jacks. But wait, are RJ11 RJ14 6P4C
phone and RJ12 6P5C data cables different? Checking,
Yes! An RJ11 voice cable can worked reversed, but
an RJ12 data straight-through is pin 1 to pin 1.
Some sellers of RJ11 RJ14 cables let you specify to
get straight pin 1 to pin 1, or "reversed".

Sheesh, changing my PCB jack wiring back again.
Click to expand...

"Standards are a good thing. Every company should have one."
I forgot who said that.

Before you invent a new wiring standard for I2C over RJ11 connectors,
perhaps you might want to search for prior art[1].

Finding the right connector for I2C
<https://hackaday.io/project/26823-lameboy-another-esp12-handheld/log/70606-finding-the-right-connector-for-i2c>

Is there any definitive I2C pin-out guidance out there? Not looking
for a "STANDARD"
<https://electronics.stackexchange.com/questions/47056/is-there-any-definitive-i2c-pin-out-guidance-out-there-not-looking-for-a-stand>

6P6C I2C Connector Standard
<https://www.iascaled.com/store/6P6C-I2C>

I2C Bus Connectors & Cables
<https://www.i2cchip.com/i2c_connector.html>

I2C - Community Recommended Pinout DRAFT 4th Feb 2015
<http://samepins.github.io/doc/i2c.html>
<http://samepins.github.io/doc/i2c%20&%20one-wire.html>

There are probably more such "standards".

Drivel: In the distant past, I was marginally involved in trying to
cram the 25 pin RS-232 wiring into an 8 pin RJ-45 connector while
having the pin numbering reversible between DTE and DCE. Not my idea
of fun. To make everyone happy, we had to use an (optional) 10 pin
RJ-50 connector.


[1] If it does something disgusting when reversed, like destroy the
bus transceivers, you can blame whomever originally contrived the
wiring standard.

--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus
 
On Sun, 12 May 2019 13:53:50 -0700, Jeff Liebermann <jeffl@cruzio.com>
wrote:

(...)
Are you doing something like this?
<http://hivetool.net>
<http://wiki.hivetool.org/HiveTool.org>

--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

---
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus
 
Jeff Liebermann <jeffl@cruzio.com> wrote in
news:n92hdetbhvobn6rcvps0uuet7ad2cufak1@4ax.com:

On Sun, 12 May 2019 13:53:50 -0700, Jeff Liebermann
jeffl@cruzio.com> wrote:

(...)
Are you doing something like this?
http://hivetool.net
http://wiki.hivetool.org/HiveTool.org
Click to expand...

That is pretty cool.
 
Jeff Liebermann <jeffl@cruzio.com> wrote in
news:8vvgde5mfaet0rva8vjkacf4pcs8ueudma@4ax.com:

"Standards are a good thing. Every company should have one."
I forgot who said that.
Click to expand...

Probably Carnegie, Morgan, or Rockefeller... Naaah!

Sounds more like something Twain would have said.
 
On Sun, 12 May 2019 23:01:53 +0000 (UTC),
DecadentLinuxUserNumeroUno@decadence.org wrote:

Jeff Liebermann <jeffl@cruzio.com> wrote in
news:8vvgde5mfaet0rva8vjkacf4pcs8ueudma@4ax.com:

"Standards are a good thing. Every company should have one."
I forgot who said that.

Probably Carnegie, Morgan, or Rockefeller... Naaah!
Sounds more like something Twain would have said.
Click to expand...

It was much more recent. As I vaguely recall, it was said by a well
known computah industry luminary during some long forgotten patent
battle the 1990's. However, Google couldn't find it, so the quote
must not exist.

2nd best:
The good thing about standards is that there are so many
to choose from. (Andrew S. Tanenbaum)

I would guess some flavor of I2C wiring and color coding will
eventually become the de facto standard, where it doesn't need to be
the superior solution, it just needs to mostly work. Meanwhile, the
success of external I2C sensors and wiring will depend mostly on how
the technology is misused, abused, and priced.


--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558
 
Jeff Liebermann wrote...
>
See DropBox for my six-channel sensor board:
https://www.dropbox.com/sh/w1xbcwovma602he/AADzRhQgieifJ9z1FoSscDrHa?dl=0

RIS-788_BME280+Si1145+LED.JPG shows daughter
PCB with T + RH + mBar pressure, plus three
light channels, plus a bright LED heartbeat.
That was last year's I2C sensor board; this
year's new I2C board adds five ICs, with 12
channels, located in the middle of the hive,
see TH-stick_sensors-only_3D.PNG


--
Thanks,
- Win
 
Jeff Liebermann wrote...
(...)
Are you doing something like this?
http://hivetool.net
http://wiki.hivetool.org/HiveTool.org
Click to expand...

Yes, all that and much much more. 60 sensors so far,
and now adding 12 more, including gas sensors. E.g.,
we count every single bee trip in/out of hive, see up
to 170,000 trips (not under 10,000). Including the
back door. Two microphone sensor channels. Also
getting sugar-water level. LoRa real-time reporting.
As a scientific research tool, have learned a lot, I
hope, but not yet sure of value to experienced beehive
operators. Had 6 systems out last season, will deploy
15 more this season.


--
Thanks,
- Win
 
On Sun, 12 May 2019 23:04:20 +0000 (UTC),
DecadentLinuxUserNumeroUno@decadence.org wrote:

Jeff Liebermann <jeffl@cruzio.com> wrote in
news:n92hdetbhvobn6rcvps0uuet7ad2cufak1@4ax.com:

On Sun, 12 May 2019 13:53:50 -0700, Jeff Liebermann
jeffl@cruzio.com> wrote:

(...)
Are you doing something like this?
http://hivetool.net
http://wiki.hivetool.org/HiveTool.org

That is pretty cool.
Click to expand...

Yep. Some of the details on how bee hive monitor is used are
interesting. I didn't know that the life of a bee hive is so
complicated:
Hive Management
<http://wiki.hivetool.org/Hive_Management>

Instead of one connector for each PCB designers idea for creative I2C
wiring and pinout, he simply crimps the cable to match his wiring
standard. See photos of PCB at:
<http://wiki.hivetool.org/Hardware:_Hive_Interface_Board_1.0>
and wiring at:
<http://wiki.hivetool.org/Hardware:_Sensor_Wiring>
I've been through the DHT22 humidity sensor color coding mess with my
various weather stations.

Methinks he made a mistake with the TSL2591 Lux Sensor wiring, which
shows that the cable is wired for "standard" telco crossover wiring,
which might a problem since the 3.3v and ground are on opposite ends
of the RJ14 connector and reversing them might become a smoke test. My
rule of thumb is that if it's used for telco, then a "rollover" or
mirror image connector wiring should be used. For everything else, I
wire it pin 1 to pin 1.

Oops. I don't think the author thought this out very well:
<http://wiki.hivetool.org/Hardware:_Sensor_Wiring#TSL2591_Lux_Sensor>
Consider cutting a 10 or 12' cord in half for two
sensors, crimping will be unnecessary.
If he uses "standard" telco flat cable, cutting it in half will result
in two different (mirrored) RJ11 wiring configurations for each half.
I suspect he wrote that before actually trying it.

--
Jeff Liebermann jeffl@cruzio.com
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558
 
Jeff Liebermann <jeffl@cruzio.com> wrote in
news:rgfhdeler6jljhqmijs5qdnj3mnhhlncom@4ax.com:

On Sun, 12 May 2019 23:04:20 +0000 (UTC),
DecadentLinuxUserNumeroUno@decadence.org wrote:

Jeff Liebermann <jeffl@cruzio.com> wrote in
news:n92hdetbhvobn6rcvps0uuet7ad2cufak1@4ax.com:

On Sun, 12 May 2019 13:53:50 -0700, Jeff Liebermann
jeffl@cruzio.com> wrote:

(...)
Are you doing something like this?
http://hivetool.net
http://wiki.hivetool.org/HiveTool.org

That is pretty cool.

Yep. Some of the details on how bee hive monitor is used are
interesting. I didn't know that the life of a bee hive is so
complicated:
Hive Management
http://wiki.hivetool.org/Hive_Management

Instead of one connector for each PCB designers idea for creative
I2C wiring and pinout, he simply crimps the cable to match his
wiring standard. See photos of PCB at:
http://wiki.hivetool.org/Hardware:_Hive_Interface_Board_1.0
and wiring at:
http://wiki.hivetool.org/Hardware:_Sensor_Wiring
I've been through the DHT22 humidity sensor color coding mess with
my various weather stations.

Methinks he made a mistake with the TSL2591 Lux Sensor wiring,
which shows that the cable is wired for "standard" telco crossover
wiring, which might a problem since the 3.3v and ground are on
opposite ends of the RJ14 connector and reversing them might
become a smoke test. My rule of thumb is that if it's used for
telco, then a "rollover" or mirror image connector wiring should
be used. For everything else, I wire it pin 1 to pin 1.

Oops. I don't think the author thought this out very well:
http://wiki.hivetool.org/Hardware:_Sensor_Wiring#TSL2591
_Lux_Senso
r
Consider cutting a 10 or 12' cord in half for two
sensors, crimping will be unnecessary.
If he uses "standard" telco flat cable, cutting it in half will
result in two different (mirrored) RJ11 wiring configurations for
each half. I suspect he wrote that before actually trying it.
Click to expand...

With so many units involved, buying a set of 500+ connectors and
500+ sockets for PCB mounting and using a modern, micro-sized form
factor would seem to be far better. Much higher integrity
connections and one can perform any pair twisting or coax run or any
other schema for the interlinks. Even a set of sata pairs (plug and
socket) would seem better as they can be stacked on a PCB layout one
right next to the next one. Ten sata connector sockets would fit on
an interface pcb in a couple square inches, whereas those old phone
interlink jacks and plugs are simply far to big for such a huge
sensor unit count.

Very similar to the old audio interconnect cables on the first CD
players for PC installation. They are like 3 or 4 wire inline
miniature connector heads. They were white and they had god
retention in their deisgn.

Google images or just google

jst sh jumper

<https://www.google.com/search?q=jst+sh+jumper>
 
DecadentLinuxUserNumeroUno@decadence.org wrote...
Winfield Hill wrote:

See DropBox for my six-channel sensor board:
https://www.dropbox.com/sh/w1xbcwovma602he/AADzRhQgieifJ9z1FoSscDrHa?dl=0

RIS-788_BME280+Si1145+LED.JPG shows daughter
PCB with T + RH + mBar pressure, plus three
light channels, plus a bright LED heartbeat.
That was last year's I2C sensor board; this
year's new I2C board adds five ICs, with 12
channels, located in the middle of the hive,
see TH-stick_sensors-only_3D.PNG

Outdoor?
Click to expand...

Last year's I2C board is inside the monitor box,
exposed to the outside. This year's additional
I2C sensor board is in the middle of the hive.

TE connectivity.
"SLIMSEAL"
https://www.te.com/usa-en/plp/slimseal-miniature-connectors/Xpmyza.html
Click to expand...

Very interesting, I hadn't found that one. But,
what's with only up to 3 positions available?


--
Thanks,
- Win
 
Winfield Hill <hill@rowland.harvard.edu> wrote in
news:qbahcj0chh@drn.newsguy.com:

Jeff Liebermann wrote...

See DropBox for my six-channel sensor board:
https://www.dropbox.com/sh/w1xbcwovma602he/AADzRhQgieifJ9z1FoSscDr
H
a?dl=0

RIS-788_BME280+Si1145+LED.JPG shows daughter
PCB with T + RH + mBar pressure, plus three
light channels, plus a bright LED heartbeat.
That was last year's I2C sensor board; this
year's new I2C board adds five ICs, with 12
channels, located in the middle of the hive,
see TH-stick_sensors-only_3D.PNG
Click to expand...

Outdoor?

Te connectivity.

"SLIMSEAL"

https://www.te.com/usa-en/plp/slimseal-miniature-
connectors/Xpmyza.html
 
Winfield Hill <hill@rowland.harvard.edu> wrote in
news:qbakbv0gl0@drn.newsguy.com:

DecadentLinuxUserNumeroUno@decadence.org wrote...

Winfield Hill wrote:

See DropBox for my six-channel sensor board:

https://www.dropbox.com/sh/w1xbcwovma602he/AADzRhQgieifJ9z1FoSscD
rHa?dl=0

RIS-788_BME280+Si1145+LED.JPG shows daughter
PCB with T + RH + mBar pressure, plus three
light channels, plus a bright LED heartbeat.
That was last year's I2C sensor board; this
year's new I2C board adds five ICs, with 12
channels, located in the middle of the hive,
see TH-stick_sensors-only_3D.PNG

Outdoor?

Last year's I2C board is inside the monitor box,
exposed to the outside. This year's additional
I2C sensor board is in the middle of the hive.

TE connectivity.
"SLIMSEAL"
https://www.te.com/usa-en/plp/slimseal-miniature-
connectors/Xpmyza.
html

Very interesting, I hadn't found that one. But,
what's with only up to 3 positions available?
Click to expand...

automotive target 'demographic'. I didn't notice.

OK so somebody needs to tell them we need more positions.

I looked. Their ip67 unit with 8 positions is a CAT7A unit for
long haul and is a couple hundred each. Wow.
 
On Sun, 12 May 2019 18:34:38 -0700, Jeff Liebermann <jeffl@cruzio.com>
wrote:

I would guess some flavor of I2C wiring and color coding will
eventually become the de facto standard, where it doesn't need to be
the superior solution, it just needs to mostly work. Meanwhile, the
success of external I2C sensors and wiring will depend mostly on how
the technology is misused, abused, and priced.
Click to expand...

There are several issues why the unbalanced I2C is not a good slut ion
for large scale sensor deployment:

* The demand for longer distances. Increases the EMC problems.
* The demand for higher speeds. Impedance levels are not well defined,
reflections.
* The demand for lower supply voltages (and possibly large currents).
Ground potential issues will reduce the noise immunity.

Any differential system will address these issues even when trying to
interface with processors and sensors with only I2C interfaces. Look
for dual RS-485, dual CAN transceivers or dual LVDS multidrop
transceivers.
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top