Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: [802.3_SPEP2P] EXTERNAL: Re: [802.3_SPEP2P] AW: AW: [802.3_SPEP2P] Number of link segment in-line connectors



Hi Peter,

 

I believe what Steffen says is that installation is reasonably managed for up to 100 m cables. A 1000 m run can be installed as 10 x 100 m. If we are not counting the connectors on the end as part of the link segment, then we really have 9 in-line connectors, not 10. Better to allow one extra I guess.

 

The longer the cable run, especially outdoors, the more likely you are to have surges coupled into the link segment by nearby lightning strikes. Common practice is to apply a Surge Protection Device (SPD) at bulkhead locations (either where a cable enters a cabinet or a building through a wall. A switch is likely to be in the cabinet or in the building. An end device is likely to be attached in the field at the other end of the link segment. The end device also requires surge protection, since a lightning surge is likely to be seen at both ends. The SPD may be bulky in comparison to a connector and would be an imposition to place in an end device. Nothing prevents the same end devices from operating indoors on short cables, so the SPD should be optional. There is generally some limit to the number of lightning strikes (nearby and not a direct hit) that a SPD can tolerate without replacement. So the SPD falls into a maintenance program for replacement. The surge itself may be picked up capacitively by the shield or from a ground voltage gradient into the shield termination. This has a secondary capacitive coupling from the shield as high voltage and current common mode transients into the communication pair. A separate and substantially rated SPD ends up having to attach to the communication pair to shunt the surge.

 

One method to accommodate a pair of SPDs, one on each end, having some effect on the communication pair, is to set some limit on the impairment to IL, RL, etc. A simple rule is to say that the SPD can have an equal effect to an in-line connector. Given this strategy, we ought to have allowed 11 in-line connectors for 1000 m. The last cable sections on each end might not be 100 m, but 90 m + SPD + 10 m for example.

 

Extrapolating to 500 m, and with this equivalence rule, and 100 m maximum section, we might want 4 + 2 = 6 in-line connectors. Or more generally:  L/100 + 1, where L is the maximum reach in meters, rounded up to a multiple of 100 m.

 

 

Regards,

 

David D. Brandt

Engineering Fellow

Rockwell Automation

1201 South Second Street

Milwaukee, WI 53204-2496

P: 1.414.382.4309

ddbrandt@xxxxxxxxxxxxxxxxxxxxxx

 

Expanding Human Possibility

  

 

 

From: Peter Jones (petejone) <00000b5d1d72f221-dmarc-request@xxxxxxxxxxxxxxxxx>
Sent: Thursday, July 1, 2021 12:49 PM
To: STDS-802-3-SPEP2P@xxxxxxxxxxxxxxxxx
Subject: EXTERNAL: Re: [802.3_SPEP2P] AW: AW: [802.3_SPEP2P] Number of link segment in-line connectors

 

[Use caution with links & attachments]

 

Steffen,

 

Thanks for the response.

 

I’m still wondering what the cabling structure looks like for the connectors within the cable run. I was thinking 10 connectors in 1000m as being the same as a connector every 100m. I thin you are saying that it’s more like a connector every 80-85m. What are the design rules behind this?

 

FYI – I appended the earlier part of the mail thread to this message.

 

Regards
Peter

_________________________________________________________

Peter Jones               Cisco Systems           

Distinguished Engineer    170 West Tasman Dr.

Enterprise Networks       San Jose, CA, 95134, USA.

Wrk: +1 408 525 6952      Mob: +1 408 315 8024

Email:                    petejone at cisco.com

Web:                      about.me/petergjones

Webex:                    cisco.webex.com/join/petejone

_________________________________________________________

 

From: Steffen Graber <steffen@xxxxxxxxxxxxx>
Sent: Wednesday, June 30, 2021 6:04 AM
To: STDS-802-3-SPEP2P@xxxxxxxxxxxxxxxxx
Subject: [802.3_SPEP2P] AW: AW: [802.3_SPEP2P] Number of link segment in-line connectors

 

Hi Peter,

 

when originally discussing the link segment for 10BASE-T1L at least I did not think about surge protection devices at this point in time. For Ethernet-APL this topic came up, when later writing the Ethernet-APL Port Profile Specification, but at that point the number of inline connectors was already fixed for 10BASE-T1L. For the 200 m Ethernet-APL spur, we were able to go up to 4 (2 + 2) inline connectors, as this was within the bounds of the 10BASE-T1L 1.0 Vpp operating mode link segment definition, but unfortunately not for the Ethernet-APL trunk.

 

Therefore I would recommend taking these two additional auxiliary devices/connectors into account, as otherwise we could use up pretty fast the maximum possible connector amount (having 8 remaining for 1000 m is less critical than just having 3 remaining for 500 m).

 

From my knowledge, the market is requiring surge protection for many reasons, but for me personally it is hard to tell, how many applications will use surge protection or not (this typically depends on many factors, like if it’s a chemical plant within a building or “open frame” without surrounding building, what environmental conditions and what legal aspects and safety measures have to be taken into account and so on).

 

So hopefully we also could get some more information until the next meeting.

 

Regards,

 

Steffen

 

----------------------

 

From: Peter Jones (petejone) 00000b5d1d72f221-dmarc-request@xxxxxxxxxxxxxxxxx
Sent: Tuesday, June 29, 2021 2:09 PM
To: STDS-802-3-SPEP2P@xxxxxxxxxxxxxxxxx
Subject: Re: [802.3_SPEP2P] Number of link segment in-line connectors

 

Steffen,

 

Thanks for this. a couple of things spring to mind:

  1. With existing 10BASE-T1L being 10 connectors over a 1000m, I assume the model is the connectors/surge protectors at each end, and 8 connectors in the middle with 100m nominal distances between them.
  2. If this is the case, then it seems that we are OK with 10 “connectors” (really connecting devices) in 10BASE-T1L and so 5 should be find for a 500m 100BASE-T1L line.
  3. On this general topic, it seems like maybe we should get some more info on “deployment considerations”. I’m wondering how common surge protectors and similar devices will be in deployments. Is it the case that these are “normally” deployed? Does this factor into the link segment in any other ways?

 

Regards
Peter

_________________________________________________________

Peter Jones               Cisco Systems           

Distinguished Engineer    170 West Tasman Dr.

Enterprise Networks       San Jose, CA, 95134, USA.

Wrk: +1 408 525 6952      Mob: +1 408 315 8024

Email:                    petejone at cisco.com

Web:                      about.me/petergjones

Webex:                    cisco.webex.com/join/petejone

_________________________________________________________

 

From: Steffen Graber <steffen@xxxxxxxxxxxxx>
Sent: Tuesday, June 29, 2021 12:55 AM
To: STDS-802-3-SPEP2P@xxxxxxxxxxxxxxxxx
Subject: [802.3_SPEP2P] Number of link segment in-line connectors

 

Dear all,

 

during our last meeting we discussed in the SPE enhancements study group the question about how many in-line connectors will be needed for a maximum length link segment.

 

While I agree that one in-line connector per 100 m link segment length is a good choice, I recommend to increase the overall number of allowed in-line connectors per link segment by an additional two connectors, allowing to substitute these two in-line connectors by Ethernet-APL auxiliary devices. In Ethernet-APL up to two auxiliary devices (typically surge protectors) can be added, while reducing the number of allowed in-line connectors by one per auxiliary device.

 

This is very similar to adding a PoE midspan PSE, which can be installed instead of one of the connection pairs in the floor distributor, as described in clause 33.4.9. Allowing for two additional in-line connectors in this context would also allow to use (potentially future defined) PoDL midspan devices being able to “upgrade” unpowered SPE switches by adding an external power coupling/decoupling circuit.

 

Just allowing e.g. 5 in-line connectors for a 500 m link segment, this would result in just having 3 remaining in-line connectors in applications, which use surge protection or future midspan PoDL devices, which would be 1 connector per 167 m only and thus require pretty long cable runs between two in-line connectors. Going for up to 7 in-line connectors (5 + 2) for a 500 m link segment would bring us back to the 1 in-line connector per 100 m range.

 

Best regards,

 

Steffen Graber

 



To unsubscribe from the STDS-802-3-SPEP2P list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-3-SPEP2P&A=1


To unsubscribe from the STDS-802-3-SPEP2P list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-3-SPEP2P&A=1


To unsubscribe from the STDS-802-3-SPEP2P list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-3-SPEP2P&A=1