[802.3_SPEP2P] AW: [EXTERNAL] [802.3_SPEP2P] [GT10MbSPE] Website update,

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

Dear George,

Hello together,

I like to share with our team some information and ideas regarding our SPE P-t-P enhancement project.

I start my business career in 2000. At this time the Ethernet technology jump over from the enterprise market into the industrial applications. It takes more than 10 years to change over the market from the 2-wire industrial BUS-System like (PROFINET, DeviceNet and others) to the 2-pair (/4-wire) Fast-Ethernet based Industrial Ethernet (IE) systems like Ethernet/IP, PROFINET, EtherCAT and others. Today Automation, Robotics, Machinery, and many other industrial applications use more Ethernet based networks compared to the still uses industrial bus systems. Some of this IE systems specify also Gigabit-Ethernet usage. But the majority of the devices still work with 100Mbs.

The feedback of a huge number of discussions with suppliers of industrial automation systems and industrial users in the last years, show up the following reasons why they still not change over to Gigabit:

For some applications 100Mbs are good enough
Installation of 4-pair Gigabit cabling is too complex and error-prone in industrial environments.
Special trained and skilled installers are needed.

Since SPE show up in the market, many end customers hope they can go back to a 2-wire system but now for Gigabit-Ethernet.

For the automotive industry we create at IEEE802.3 an optimal ecosystem for 15m link lengths. But this SPE protocols are not perfect for non automotive use cases (reach is with 15m to short, limited AUTO-NEG, long latency because of FEC).

For the industry including transportation and smart building applications, we start to build the SPE ecosystem already with the 10BASE-T1L protocol. Higher speeds are needed to complete the protocol ecosystem for more and more higher data rate applications.

Based on these facts I learned from the market and the presentations for use cases we see during the last moths this are my proposal for the

Managed Objects

Suggested Basic PHY Objectives

1. Preserve the IEEE 802.3/Ethernet frame format at the MAC client service interface.

2. Preserve minimum and maximum frame size of the current IEEE 802.3 standard.

3. Do not preclude meeting FCC and CISPR EMC requirements

4. Support for (optional) single-pair Auto-Negotiation

5. Do not preclude the ability to survive industrial fault conditions (e.g., shorts, overvoltage, EMC)

6. Do not preclude working within an Intrinsically Safe device and system as defined in IEC 60079

7. Be compatible with existing PoDL remote powering (IEEE802.3bu / ..cg)

Speed-Specific Objectives (100 Mb/s)

1. Support a speed of 100 Mb/s at the MAC/PLS service interface.

2. Support 100 Mb/s single-pair Ethernet operation in industrial environments (e.g., EMC, temperature).

3. Define performance characteristics of a link segment with a single balanced pair of conductors supporting up to 6 inline connectors for up to at least ≥ 400m reach, and a PHY supporting point-to-point full duplex operation over the link segment.

4. Maintain a bit error ratio (BER) at the MAC/PLS service interface of less than or equal to TBD (10-10)

5. Latency objective?

Speed-Specific Objectives (1000 Mb/s)

1. Support a speed of 1000 Mb/s at the MAC/PLS service interface.

2. Support 1000 Mb/s single-pair Ethernet operation in industrial environments (e.g., EMC, temperature).

3. Define performance characteristics of a link segment with a single balanced pair of conductors supporting up to 4 inline connectors for up to at least ≥ 100m reach, and a PHY supporting point-to-point full duplex operation over the link segment.

4. Maintain a bit error ratio (BER) at the MAC/PLS service interface of less than or equal to (10-10)

5. Latency objective? TBD

In this way we build up a perfect SPE ecosystem for all non-automotive Ethernet applications.

1000BASE-T1L is possible based on 5GBASE-T1 or 10GBASE-T we have already in the market. Here we use 4-pairs and we have more than 1Gbs per pair over ≥ 100m link length.

The market potential is valuable and if this standard exists in the market more application will show up.

It is very attractive for the market to move from 2-pair Fast-Ethernet to 1-pair Gigabit-Ethernet, compared to change only from 2-pair Fast-Ethernet to 1-pair Fast-Ethernet. Gigabit SPE offers better performance with less cost and easier to install cabling.

Thanks for your time to read this text and I hope you see the big benefit of Gigabit-SPE for 100m link segments.

Very nice greetings Matthias

Best regards / Mit freundlichen Grüßen

Matthias Fritsche
HARTING Electronics GmbH
- Senior Specialist & Global Product Manager Ethernet Connectivity -
Marienwerderstraße 3
32339 Espelkamp - Germany
E-mail: Matthias.Fritsche@xxxxxxxxxxx
http://www.HARTING.com
Phone: +49 5772 47-9104
Fax: +49 5772 47-909104
Mobile: +49 171 3389010

Von: George Zimmerman <george@xxxxxxxxxxxxxxxxxxxx>
Gesendet: Mittwoch, 27. Oktober 2021 21:04
An: STDS-802-3-SPEP2P@xxxxxxxxxxxxxxxxx
Betreff: [EXTERNAL] [802.3_SPEP2P] [GT10MbSPE] Website update, minutes, next meeting, actions requested

All –

Thank you for a productive meeting today. The minutes and meeting files are on the website.

Our next meeting is at the November Plenary (Nov 10), and at the closing 802.3 working group meeting we expect to be asking for rechartering of the study group.

[PLEASE BE ADVISED THAT AS PART OF THE PLENARY MEETING, REGISTRATION IS REQUIRED – see http://802world.org/plenary/ ]

I feel we made some good progress on cleaning up our draft CSD and potential objectives, at least for the 100 Mb/s speed. We still have work to do but the key points appear to be identified and ready for decision at the next meeting. If we get those, we are in a good position to request rechartering and proceed towards a completed project in a small number of cycles.

In order to be ready for that, we will need some consensus building, and some contributions (or email discussion on the reflector – always preferable if it is short) on a few points:

I expect straw polls, and possibly votes at the next meeting on at least the following things:

A straw poll to update potential participation and breadth of expertise in the project
Reach for 100 Mb/s: 400m or 500m.

There was significant discussion on this today, and the sense I got was that 400m was supported from technical feasibility, and in line with emerging directions in other standards. 500m may be desirable, (because it is longer and because it aligns with the 1Vpp mode of 10BASE-T1L), but could use more technical feasibility support, particularly in light of power and economic feasibility constraints. Contributions on technical and economic feasibility for both are encouraged.

Support and objectives for a 1000 Mb/s rate PHY

We are lacking contributions from PHY SMEs on a 1000 Mb/s PHY, and would need such contributions to support the rate. Additionally these should speak to reach, latency, constraints, and technical feasibility of such a PHY.

Participants are also encouraged to review the wording on the PAR, CSDs, and Objectives. Areas highlighted in yellow show those places remaining where it seemed the wording could use some review and possible rephrasing.

George Zimmerman, Ph.D.

President & Principal

CME Consulting, Inc.

Experts in Advanced PHYsical Communications

george@xxxxxxxxxxxxxxxxxxxx

310-920-3860

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

[802.3_SPEP2P] AW: [EXTERNAL] [802.3_SPEP2P] [GT10MbSPE] Website update, minutes, next meeting, actions requested