Building Automation…
The existing standard for building automation systems is 1200 m reach. I am personally persuaded that changing from multidrop to point-to-point will make 300
m reach feasible. To the extent that I err, more than 300 m means less risk. I am wary of subdividing the market, so I would support 1000 m reach as agreed on by the industrial people.
For BER, building automation networks are more tolerant than automotive or industrial networks. Most devices can handle a one-minute communication outage. (My
thermostat at home runs off 2 AA batteries. It works happily for three years. In the week before I put in new batteries, it electrical performance is probably very poor, but the system runs.) The argument for BER = 10^-9 is that it allows me to use the same
communication inside a piece of equipment (such as a chiller) as I use to network that piece of equipment.
David
From: Peter Jones (petejone) [mailto:petejone@xxxxxxxxx]
Sent: Saturday, October 29, 2016 2:49 PM
To: STDS-802-3-10SPE@xxxxxxxxxxxxxxxxx
Subject: Re: [802.3_10SPE] Objective - BER
Folks,
It seems to me that we have at least 3 clear use cases:
1.
Automotive Automation up to at least 15 m
2.
Industrial Automation up to at least 1000 m
3.
Building Automation up to at least N m where N is bigger than 15, maybe 300-500?
I think the Automotive Automation folks have been clear that they need 10^-10 for the 15 m reach. The other two cases are happy with 10^-9.
George proposed
Maintain a bit error ratio (BER) at the MAC/PLS service interface of less than or equal to:
10^-10 on link segments up to at least 15m, and
10^-9 on all link segments up to at least 1km
I read this as requiring 10^-10 on all link segments 15 m or less, and 10^-9 above this. To me it does not say anything about the number of PHYs the TF will define.
I’d prefer to decide how many PHYs we need when we have the right technical data in the Task Force.
Regards
Peter
_______________________________________________
Peter Jones Cisco Systems
Principal Engineer 560 McCarthy Blvd.
Campus Switching S/W Milpitas, CA, 95035 USA
Wrk: +1 408 525 6952 Mob: +1 408 315 8024
Email: petejone at cisco.com
Twitter: @petergjones
LinkedIn: /in/petergjones
_______________________________________________
From: Pat Thaler [mailto:000006d722d423ba-dmarc-request@xxxxxxxx]
Sent: Friday, October 28, 2016 2:55 PM
To: STDS-802-3-10SPE@xxxxxxxxxxxxxxxxx
Subject: Re: [802.3_10SPE] Objective - BER
I don't see a difference in the implications of #1 and #2. In either case, the first part reads as requiring 10^-10 when the link segment is 15 m or less. #1 doesn't mention anything about which PHY so it says that even a PHY that can drive
1 km should 10^-10 or better BER when running on a 15 m link segment (and perhaps that would be true).
If then intent is that PHYs under proposed objective 9 support 10^-10 BER and those under proposed objective 10 support 10^-9 BER regardless of where in their distance range they are used, then that would need a different wording.
An editorial nit: we put a space between the number and the unit because IEEE style requires it.
On Fri, Oct 28, 2016 at 1:57 PM, Geoff Thompson <thompson@xxxxxxxx> wrote:
George-
I don't know. I see two interpretations.
Maintain a bit error ratio (BER) at the MAC/PLS service interface of less than or equal to 10^-10 on link segments up to at least 15m.
Maintain a bit error ratio (BER) at the MAC/PLS service interface of less than or equal to 10^-9 on all link segments up to at least 1km.
(which to my eyes would imply but not require two PHYs)
Maintain a bit error ratio (BER) at the MAC/PLS service interface of less than or equal to:
10^-10 on link segments up to at least 15m, and
10^-9 on all link segments greater than 15m up to at least 1km
(which to my eyes would imply a single PHY)
...but would be clearer wording.
We have had some round-abouts on how to write the BER objective. Ludwig had proposed a nice split, where the short-reach/automotive PHYs had a 10^-10 BER objective and the
longer reach ones had a 10^-9 BER objective. With that in mind, I propose the following objective for discussion:
Maintain a bit error ratio (BER) at the MAC/PLS service interface of less than or equal to:
10^-10 on link segments up to at least 15m, and
10^-9 on all link segments up to at least 1km
Experts in Advanced PHYsical Communications