Remember, only a virgin can tame a unicorn. https://en.wikipedia.org/wiki/Unicorn#/media/File:DomenichinounicornPalFarnese.jpg
From: Mike Hammett Sent: Sunday, October 25, 2015 4:28 AM To: [email protected] Subject: Re: [AFMUG] B11 copper v fiber I want a unicorn in every box I order. ----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com Midwest Internet Exchange http://www.midwest-ix.com -------------------------------------------------------------------------------- From: "Jaime Fink" <[email protected]> To: [email protected] Sent: Saturday, October 24, 2015 10:11:47 PM Subject: Re: [AFMUG] B11 copper v fiber Matt - that's what we're working on. 802.3at works different than most are aware since all 8 wires are used for 1000BASE-T, so we had to add circuitry to make powering over fewer pairs work additionally, and we need a clean way that maintains the sealing of the enclosure. So it's not off the shelf, and I'd have even more people yelling at me if we didn't have PoE that aren't fiber up the tower users yet. Hang tight, we'll have something for y'all that takes direct DC into the PoE port and maintain the environmental sealing. Wish it were as easy as shoving raw fiber and power cables magically into a unicorn connector ;) Jaime On Oct 24, 2015, at 12:47 PM, Mathew Howard <[email protected]> wrote: Is there any reason you couldn't just feed DC directly into the ethernet port? I'm thinking just wire the appropriate pairs in a cat5 cable directly to DC and skip using a power injector. On Sat, Oct 24, 2015 at 2:14 PM, Jaime Fink <[email protected]> wrote: We isolate Ethernet from power internally so even if the ESD protection we put on Ethernet for any reason failed to protect the Ethernet PHY you should be in good shape if you're using fiber. Should continue to take power. Nonetheless we're looking at some clean solutions to adapt direct DC easily in to avoid full PoE solutions at the top of the tower. To Stefans point any converter would work too. On Oct 24, 2015, at 10:17 AM, Jeremy <[email protected]> wrote: I can see your point, but it seems that you are missing part of the idea of why we run direct DC and fiber, to avoid ESD issues frying the sensitive Ethernet components. We already have direct -48vdc and fiber sitting in our tower box, ready for future expansion. You simply needed one additional small connector and you could have made everyone happy. There are already a number of manufacturers out there who not only support POE, but also have direct DC connectors as well. AFAIK, The connector part is less than a buck. Consider this my request for a future hardware revision that adds the small direct DC connector as an option. On Sat, Oct 24, 2015 at 10:34 AM, Jaime Fink <[email protected]> wrote: Yes it is powered via 802.3at PoE. You do not need Ethernet PHY operational though once you've configured the SFP, it just accepts power through the circuit at that point. Unfortunately not everyone has gone to fiber yet so PoE is still needed for some transitionally. Jaime On Oct 24, 2015, at 9:19 AM, Jeremy <[email protected]> wrote: I do not see the direct DC connector on any of the marketing photos. Do these still need to be powered via POE? On Wed, Oct 21, 2015 at 4:40 PM, Jaime Fink <[email protected]> wrote: Thanks Mike. Yes the aggregates Mike quoted are roughly correct as well, but…it’s a bit different when you’re using our Auto-TDMA mode. Assuming you’re not fully loaded, users running speedtests get results at a Gigabit in up and down direction and around 1 ms RT latency, so it feels higher speed aggregate since we’re adapting to actual usage on the fly. In other words we fill up the underutilized directional demand with the direction that’s in demand (usually downstream of course). That’s opposed to locking down the MAC/PHY layer at 50/50 like FDD does, or 75/25 TDMA modes, etc. Jaime From: Af <[email protected]> on behalf of Mike Hammett <[email protected]> Reply-To: "[email protected]" <[email protected]> Date: Wednesday, October 21, 2015 at 2:33 PM To: "[email protected]" <[email protected]> Subject: Re: [AFMUG] B11 copper v fiber Yes. 1200 - 1500 aggregate. ----- Mike Hammett Intelligent Computing Solutions http://www.ics-il.com -------------------------------------------------------------------- From: "TJ Trout" <[email protected]> To: [email protected] Sent: Wednesday, October 21, 2015 2:00:16 PM Subject: Re: [AFMUG] B11 copper v fiber Jamie; Is the b11 based on 802.11 silicone? Is the b11 capable of 750fd or 1500fd? Thank you On Wed, Oct 21, 2015 at 11:29 AM, Rory Conaway <[email protected]> wrote: And Mike quits sending you nasty emails because you think copper was sent to us by an evil entity from another dimension. Rory From: Af [mailto:[email protected]] On Behalf Of Adam Moffett Sent: Wednesday, October 21, 2015 11:27 AM To: [email protected] Subject: Re: [AFMUG] B11 copper v fiber Two reasons: 1) Peace of mind regarding future EMI problems 2) It's pretty sexy these days to tell people you have fiber to the antenna, On 10/21/2015 1:35 PM, Scott Vander Dussen wrote: Mimosa recommends these SFP modules: AFBR-5710APZï¿1Ž2 1.25 GBd MMF Transceiver for Gigabit Ethernet, SFP, Bail de-latch, Ext Temp (-40 to 85C) AFCT-5715ALZ 1.25 GBd SMF Transceiver with DMI for Gigabit Ethernet, SFP, Std de-latch, Ext Temp (-40 to 85C) FTLF8519P3BTL Fiber Optic Transmitters, Receivers, Transceivers GigE 1x/2x FC, 2.129 Gb/s trnscvr, 550m FTLF1318P3BTL Fiber Optic Transmitters, Receivers, Transceivers 1310nmFP GigE 1x FC 1.25Gb/s trnscvr10km ï¿1Ž2 Assuming cable length and EMI are not issues, is there any benefit to using fiber over copper? ï¿1Ž2 `S
