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

Re: [802.3_NGAUTO] OEM Ethernet Cable fault detection and repair



Natalie/Eric/All-

We are on the cusp of an entire new set of functional requirements for automotive parts.
The issue that I worry about is not the factory or even the authorized dealer setting.
If we are to be realistic about world-wide maintenance requirements we have to take
a broader look at the problem.  Further, existing requirements for coax/high freq testing
are not really relevant as they are (AFAIK) only associated with entertainment systems.
The high frequency systems that have been presented to us as use cases will be used
in life critical systems.

We have to build systems that will be repairable in 3rd party mechanical and collision
repair shops around the world.  We can impose some test equipment requirements on
the post sale maintenance market but there are considerably lower limits on what the
market will tolerate.  Two examples from my experience:
  1. Tractor mfgrs tried to make it illegal in the US for owners to do essentially any engine diagnosis or repair other than filter and oil replacement.  There was a huge lobbying fight and the farmers won.  Tractor owners are now free of manufacturer restrictions when repairing (electronically controlled) systems on their own farm equipment.
  2. For many years (I don't know how many or if it still persists) automobiles manufactured for the Australian market did not have electronic ignition control.  This was the case long after the US and Europe had moved beyond traditional distributor and points.  This was done because of the primitive repair facilities and skills in the Outback.  I believe the same was true for fuel injection versus carburetors in Australian cars.

Natalie, I would appreciate a copy of your presentation that you mention.

Best regards,

Geoff Thompson

On Apr 10, 2018, at 2:09 PMPDT, NATALIE WIENCKOWSKI <nwienckowski@xxxxxxx> wrote:

All,

We do not use hand held testers to check cables built into the car.

We require the cables to be tested by the supplier at their facility before they are shipped to GM.

Once they are in the vehicle, we use the PHY/Switch TDR functions, if they exist, to determine where a cable fault may be.  We can also use information on which ECUs are participating in the communication to narrow down the issue.  

When it comes to repair, we will replace segments between connectors, in-line or end.  If we aren't sure which one it is, we will replace segments until the issue is fixed.  We will not try to repair a cable as it will likely not meet the requirements for the cable.  I gave a presentation on this topic for 100BASE-T1 at the 2016 Automotive Ethernet Congress.  If you are interested in this, please let me know and I can send you a copy.

Natalie Wienckowski

From: DiBiaso, Eric <eric.dibiaso@xxxxxx>
Sent: Tuesday, April 10, 2018 3:37 PM
To: STDS-802-3-NGAUTO@xxxxxxxxxxxxxxxxx
Subject: Re: [802.3_NGAUTO] Wednesday 4 April NGAUTO ad hoc cancelled
 
All,  

 

I do not have firsthand experience with field test equipment as Matthias, Geoff, and George have been describing, so I can’t really comment on their feasibility or relative cost as frequency increases.    However, my comment lies in the Automotive Industries need for such equipment.   

 

The reason I bring this up, is that cars built today have coaxial link segments requiring component validation and testing up to 6GHz, and to my knowledge I have not heard of such test equipment issues.   So that brings up the following questions that could best be answered by our OEM colleagues:    

 

  1. What is currently being done in the automotive industry today for testing these higher frequency links segments?   
  2. Do automotive repair shops around the world do RF testing on the above mentioned Co-axial link segments? 
  3. Is it an expectation for automotive repair shops to conduct full RF testing on NGAuto cables to diagnose a problem?   Or could the ECU’s be put into a diagnostic mode to test the links?     
  4. If a problem in the link segment is detected, is a repair of the cable expected?  Or would a completely new link segment be installed?

 

Best Regards,
Eric 

 

  

 

From: George Zimmerman [mailto:george@xxxxxxxxxxxxxxxxxxxx] 
Sent: Monday, April 9, 2018 12:55 PM
To: STDS-802-3-NGAUTO@xxxxxxxxxxxxxxxxx
Subject: Re: [802.3_NGAUTO] Wednesday 4 April NGAUTO ad hoc cancelled

 

The following message was sent from an external e-mail address. Be cautious when clicking on links, opening attachments or providing sensitive information.

Geoff, and all – I would add that my (independent) experience with field testers has been similar to what Matthias reports. I can confirm field testers in the 2-3 GHz band, but beyond that is substantially more difficult.  Many of us remember the work done in conjunction with Cat 8 cabling to get field testing to 2 GHz.  It is not trivial, but, at this point multiple vendors have gotten to that point. While 3 GHz is not a huge extension, extending the frequency of practical testers for field and assembly environments by another octave or more is a nontrivial task.  Those involved in the cabling industry know that specifications for high performance cabling are often limited by the capabilities of test.   And that is not to mention the extra difficulties brought in by the harsher automotive environment.
-george
George Zimmerman, Ph.D.
President & Principal
CME Consulting, Inc.
Experts in Advanced PHYsical Communications
310-920-3860

 

 

 

From: Geoff Thompson [mailto:thompson@xxxxxxxx] 
Sent: Wednesday, April 04, 2018 11:16 AM
To: STDS-802-3-NGAUTO@xxxxxxxxxxxxxxxxx
Subject: Re: [802.3_NGAUTO] Wednesday 4 April NGAUTO ad hoc cancelled

 

Colleagues-

 

On Apr 4, 2018, at 6:04 AMPDT, Fritsche, Matthias <Matthias.Fritsche@xxxxxxxxxxx> wrote:

 

Thanks George for the information.

 

Hello to all,

 

I review the last days for another project the market for field testers. The best handheld cable testers I found in the market can measure up to 2GHz or 2,5GHz bandwidth. Some of the HF experts in our lab and the market pointed out only below 3 to 3,5GHz bandwidth such handheld tester are possible and economical feasible. For higher frequencies the high-end lab equipment (Vector Network Analyzer) is needed. Following the discussion about the bandwidth for Multi-Gig Ethernet we should have this point in mind. All cable harnessing’s must be tested from the harnessing production before the wire harness installed into the car. And the price for the test equipment and the testing cost are also important for the car industry. Also for maintenance in the field test equipment is needed. From this point of view my understanding is we should use a bandwidth below 3 to 3,5GHz.

 

Matthias point is well made.
Please remember that harnesses will have to be tested in automotive body shops and repair shops around the world
including lesser developed countries.  The automotive market is a world-wide market.
Maintainability of high speed networks will be a major factor and challenge to their adoption.

 

Geoff Thompson
GraCaSI S.A.
 

 

So far my comment on this point. Any comment and other arguments are welcome.

 

Very nice greetings from Germany
Matthias

 

Best regards / Mit freundlichen Grüßen

Matthias Fritsche
HARTING Electronics GmbH
Dept.: Device Connectivity
- Product Manager -
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

 

 

 


HARTING Electronics GmbH, Marienwerderstraße 3, D-32339 Espelkamp; Registergericht: Amtsgericht Bad Oeynhausen; Register-Nr.: HRB 8808; Geschäftsführer: Dipl.-Kfm. Edgar-Peter Düning, Dipl.-Wirtsch.-Ing. Ralf Martin Klein, Dipl.-Ing. Uwe Gräff, Dipl.-Ing. (FH), Dipl.-Wirtsch.-Ing. (FH) Andreas Conrad

From: George Zimmerman <george@xxxxxxxxxxxxxxxxxxxx> 
Sent: Wednesday, April 04, 2018 2:05 AM
To: STDS-802-3-NGAUTO@xxxxxxxxxxxxxxxxx
Subject: [802.3_NGAUTO] Wednesday 4 April NGAUTO ad hoc cancelled

 

Due to lack of contributions, the 4 April ad hoc is cancelled. Our next ad hoc will be on 18 April. Please let me know if you wish to make a contribution.

George A. Zimmerman, Ph.D.
CME Consulting, Inc.
Experts in PHYsical Layer Communications
310-920-3860

 


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

 

 


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

 


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

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

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

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


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