At 36 users the switch heat/BTU is still problematic. AE optics for 1Gbps SFP for 36 users is around $900 like was mentioning below. Switch is probably less than $1800 for a 48 port (I use Planet). That switch can do most of the heavy lifting unless you are doing fancy stuff at that location like MPLS. Maybe another $150 for 10Gbps SFP+ backhaul connection (the Planet has four SFP+ ports) That planet also has does really well on energy/Watt usage, but only has one power supply.
Battery backup of some sort that can handle the temperature changes. Then just direct splice or whatever the 36 to LC pigtails and plug in. Not much more than that. Again though, you’ve got to do a ton of venting for a fully loaded SFP switch, or have a small AC unit. I think the A/C units we use are about $1500. You just plug it in standard 120v and cut a square rectangle in the side of your enclosure to mount it and set the temperature. You’re well within the 15amp zone for all that. And if you want to switch it over to GPON you just pull out the AE stuff and put the GPON stuff in. You probably don’t need to worry about temp anymore, so sell/reuse the AC unit, and the switch. From: Af [mailto:[email protected]] On Behalf Of Josh Reynolds Sent: Thursday, March 30, 2017 4:46 PM To: [email protected] Subject: Re: [AFMUG] FTTH construction tiger team 36 users? Jesus. That would be hard to beat..Even though that puts you at $900 plus s/h just in optics, plus the switch cost. I could do some napkin math when I'm no longer at a stop light to figure out how many users per location the tipping point is for a low cost OLT vs something with 48 SFP ports. A quick guess says around 100. On Mar 30, 2017 3:19 PM, "Adam Moffett" <[email protected]<mailto:[email protected]>> wrote: I'm on a similar quest for low cost deployment methods. If you find your super contractor team, you can give me a quote. I was planning a switch cabinet with rackmount 2U enclosures with splice trays inside and a patch panel on the front.....now I'm looking at a plain Hoffman box, fanout the cables and terminate with SOC's so the cables go straight into a switch. Waaay fewer parts, waaay cheaper box. You can also put splice trays on the back plane and splice pigtails onto your fanouts rather than use SOC's. That's probably cheaper still, but then the box is bigger. With the SOC's I can keep the box down to 20" wide, and I can mount it right to the bottom of one of our poles. I could also (as someone said) put a splice enclosure above this box and splice a bundle of long pigtails onto the OSP cable. I suspect that would come out slightly more expensive. This box for 36 AE users will come in under $1800 and goes right on the pole. I've already got electric service where I'm putting it.....if I didn't, then I'd figure on another $1000-1500. I'm not ready for the PON rabbit hole yet, but I see a lot of room for scrimping there. The problem I see is to be efficient with the cost of the OLT you need to aggregate hundreds of customers to one spot. There won't be a cheap 36 user PON box....unless UBNT is really cheap. I don't know what there price is like. ------ Original Message ------ From: "Chuck McCown" <[email protected]<mailto:[email protected]>> To: [email protected]<mailto:[email protected]> Sent: 3/30/2017 1:50:18 PM Subject: [AFMUG] FTTH construction tiger team I am considering building a construction team that can drop into a small town or large subdivision, install fiber and move on. Something where I own the construction equipment and keep labor costs low. Want to identify the minimum equipment necessary and the methods of construction to provide the best value. HDD with mud truck Do we need a vacuum excavator? Mini excavator One of these drop plows. Whoever said their guy can do 5-8 homes per day.... yeah that one (to lazy to search who it was). What to use for hand holes? Plastic hand holes? What are the best values for splice cases? Perhaps try UBNT GPON. Can always throw it in the ditch if it does not work. So a best practices/FTTH in a box schedule of equipment and methods is needed. From that I will look at the ROI needed from the equipment as well as the labor costs to estimate the costs to do a subdivision. From that we will look at the ROI on a competitive ARPU to see if an area is worth doing. I keep getting asked to do this, so I guess I better do this.
