[802.3_GT50GBIDI] APD receiver sensitivities

[Frank J Effenberger <frank.effenberger@xxxxxxxxxxxxx>]

 

All,

I wanted to prime the discussion on the BR20 and BR40 budgets. In my view, where we stand at this point is:

1. The BR40 discussion group (for lack of a better name) is favoring -12.8 dBm as the APD sensitivity. This is based on their opinion on what is the most feasible APD sensitivity including the diplexer losses.
2. There are a couple of commenters who would prefer -13.8 dBm as the sensitivity. The reasons are that an unamplified EAM transmitter has a hard time reaching the output level required if we use -12.8 dBm.  Also, -13.8 dBm matches the 100G MSA sensitivity (albeit that is a two fiber system).

Of interest, I would like to highlight the values from G.9806 Amd. 3. The sensitivities for the S_lower class (0 to 10 dB link loss) is -12.8 dBm, and for the B_lower class (10 to 20 dB) is -13.5 dBm. There is an obvious near coincidence between the two views above and the sensitivities in G.9806.

 

Therefore, I would suggest that we consider the compromise where the BR20 budget is based on a sensitivity of -12.8 dBm, and the BR40 budget is based on a sensitivity of -13.5 dB. This would put the BR40 optics within striking range (2 dB) of the B_lower class, and the industry could likely benefit from commonality of specifications there. Meanwhile, BR20 would become nearly identical to the S_lower class, and gain economies of scale also. 

I’d like to discuss this in our upcoming meeting. I will be preparing a contribution restating this argument, and if you would like to have inputs to this contribution, I’d be happy to entertain them. I think we are getting to the point where we’d like to nail down these budgets into a complete plan, so we can progress our draft.

Sincerely,

Frank E>

 

 

---

To unsubscribe from the STDS-802-3-GT50GBIDI list, click the following link: https://listserv.ieee.org/cgi-bin/wa?SUBED1=STDS-802-3-GT50GBIDI&A=1