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

Re: [STDS-802-3-400G] Electrical Ad-Hoc Straw Poll



I just did the comparison. A set of board transmission line parameters was created to emulate the a host with the proposed insertion loss curve.  This was scaled to remove the test fixture and emulate the 6.26 dB in the COM computation for cables compliance. The 10.2 dB host with the FLYOVER connector was 0.2 dB to 0.4 dB better than the original COM. I tested a worst case 26AWG 3m cable model from Megha Shanbhag  done for ‘.3by and 26 AWG cable model by Andrew Zambell done for ‘.3cd.

So the answer is it looks like it is compatible. I’ll present some more details next Monday.

… Rich

 

From: Ali Ghiasi [mailto:aghiasi@xxxxxxxxx]
Sent: Wednesday, April 26, 2017 5:17 PM
To: STDS-802-3-400G@xxxxxxxxxxxxxxxxx
Subject: Re: [STDS-802-3-400G] Electrical Ad-Hoc Straw Poll

 

Vinu

 

 

What I am suggesting is before any changes are made to 120e host requirements, lets verify compatibility with CL 136A operation with Cu cable to preserve universal host.

If there is no excess COM penalty for a link constructed from Cu cable from ball-to-ball then we have preserve universal host, but now we have the option of building line cards 

with Fly Over cable.

 

Thanks,
Ali Ghiasi
Ghiasi Quantum LLC

 

On Apr 26, 2017, at 12:48 PM, Vinu Arumugham <vinua@xxxxxxxxxx> wrote:

 

Agree with Ali on "universal host paradigm supporting optics and Cu cable."

Can we not have identical host requirements in 120E and 136A, both supporting twinax in host?

Thanks,

Vinu

 

On 04/26/2017 09:36 AM, Ali Ghiasi wrote:

Andre 

 

I vote for option 2.

 

Before agreeing to option 3, I would like to see potential COM penalty on the 50G-CR1/200G-CR4 due to strong SQRT(F) response from ball to ball!

If there is no COM penalty for 50G-CR1/200G-CR4 then I am open to option 4, as I don’t want to potentially break universal host paradigm supporting optics and Cu cable.

 

Thanks,
Ali Ghiasi
Ghiasi Quantum LLC

 

On Apr 26, 2017, at 1:28 AM, Andre Szczepanek <aszczepanek@xxxxxxxxx> wrote:

 

At today’s P802.3bs Electrical Ad-Hoc meeting I announced we would have a straw poll on options for the C2M Insertion Loss equation based on discussion following the presentation of mellitz_01b_042417_elect.

 

·       Option 1: Do nothing

·       Option 2:

       Change  “The supported insertion loss budget is characterized by Equation (120E–1) and illustrated in Figure 120E–4. “  to “The recommended insertion loss budget is characterized by Equation (120E–1) and illustrated in Figure 120E–4. “

       Add “recommended” to the title of Figure 120E-4.

·       Option 3: Change the insertion loss equation to the equation given in Slide 14 “Option B” of mellitz_01b_042417_elect.

·       Option 4: Implement the changes in both options 2 & 3 above.

 

Please send me an email specifying your preferred option (select only one).

-         Don’t get hung up on exact values. This is a straw poll, nothing is definitive – we just want to establish a direction for the group.

 

Regards

                 Andre Szczepanek (Electrical Ad Hoc chair)