Telecom Netonix DC powered switch - looking for input

[Coco-Bill]

Sorry I missed this forum before!

We have this problem all the time, as we generally build our sites with plus-ground, or co-locate on larger sites who always are "plus-ground".
The reason for this is that we use industrial standard 48V-rectifiers/ battery systems (supplied by the Norwegian Eltek.com), and the fact that we use a lot of licensed radio links like NEC, SIAE, NERA and so on.
All these radio links are built to telecom industry standard, that is "plus-ground" DC systems.
Until now we had to use a dc-dc converter feeding each switch to get galvanic insulation.
Our latest project had 25 sites, each with two WS-12-DC and each of them with a converter.
Not ideal, to say at least!

We love the new 26-IDC, it's just what the doctor ordered.
And we really would love to see a WS-12-IDC.

And could you please have a second thought about those DC terminals?
They are just fine for lab and development work, but it screams MECANO to everyone looking at our racks, and I get a bit embarrassed when trade colleagues comments those terminals

Keep up the good work, and give us more IDC switches.
Why not make all DC switches IDC? Just a thought.

[sirhc]

We love the new 26-IDC, it's just what the doctor ordered.

And could you please have a second thought about those DC terminals?

They are just fine for lab and development work, but it screams MECANO to everyone looking at our racks, and I get a bit embarrassed when trade colleagues comments those terminals

Have you looked into using Banana Connectors which plug into our DC terminals?

[Coco-Bill]

No, that will not work well for us.
It would be way to "flimsy" and insecure, I think.
And what if someone happens to rip one of those out of the terminal? That is an open live 48V plug on the loose.
We do now use this "fork" kind of connectors, it works ok but its not as professional as I would like it to be.
And I have to get around to make me a tool of some kind to be able to torque those terminals better than by hand (reminder to myself )

[sirhc]

No, that will not work well for us.
It would be way to "flimsy" and insecure, I think.
And what if someone happens to rip one of those out of the terminal? That is an open live 48V plug on the loose.
We do now use this "fork" kind of connectors, it works ok but its not as professional as I would like it to be.
And I have to get around to make me a tool of some kind to be able to torque those terminals better than by hand (reminder to myself )

Be careful about over torquing those terminals as they are in a plastic housing.

I do not know, the banana connectors will only come out if someone pulls them out.

If someone want to be a jerk and pull your DC banna connector out what prevents them from simply unplugging Ethernet or FIber connectors?

They also make a 90 degree banana connector.

[Coco-Bill]

I get you.
But what about a 12-port IDC?
We build a lot of small POPs where the 26P is really an overkill, and use of 2 ws-12-idc would be perfect (we use two for redundancy).

[sirhc]

I get you.
But what about a 12-port IDC?
We build a lot of small POPs where the 26P is really an overkill, and use of 2 ws-12-idc would be perfect (we use two for redundancy).

Keep in mind that every model requires tens of thousands of dollars to bring to market even if based off an existing model due to testing and certifications from regulatory agencies.

The WS-26-400-IDC is NOT a large seller as not many WISPs install in Telco NOCs with negative 48V DC.

So we only wanted to make 1 model knowing the low volume so we chose the WS-26 over the WS-12.

We sold LESS than 1,000 WS-26-400-IDC in 2017.

Most WISPs run normal positive DC so they want the SMART DC.

What we might do in the future is improve the SMART DC power supplies to be ISOLATED with reverse polarity protection so people do no blow them up.

2018 is the FIRST year Netonix is operating in the black, up to this point it was in the RED operatining off money loaned to Netonix from the founders which as of December 2017 has been repaid finally. Also keep in mind that Dave and I worked the first 2+ years without pay which as of December has also been paid back.

As long as people continue to support us we will grow and become cash positive and thus be able to do more things.

Once again all it takes to do anything is MONEY, LOTS OF MONEY.

Ubiquiti became flush with cash on selling a simple add in card people used with MT boards.

Robert / UBNT was already worth tens of millions of dollars from building and selling a handful of PCMCIA add in cards like the SR-71 that did not even need firmware (just a plain add in card) before they ventured in their first 802.11b airMAX line.

We had to start off building 10 models that were complete systems with chassis, boards, certifications, and firmware.

[troy]

Moving forward, if there are hardware revisions, I hope you will give serious consideration to ensuring that all your DC products have a floating ground so they can be used anywhere without issue.

In our case, as a telco for 50+ years, we like -48vdc. We use it everywhere. It allows us to use the same power supplies and battery plants between DSL, FTTH, and Fixed Wireless sites, which helps keep costs down. In the past, we've used small inverters to power up AC switches when/where necessary, but since Netonix, we've successfully deployed several WS-8-150-DC switches (which appear to have a floating ground) and now the WS-26-400-IDC switch at a brand new site that's 100% fixed wireless, but runs on -48vdc for consistency.

Our next project is a 199' tower loaded with 4 UBNT and 4 Baicells, with a mobile operator colocating with us, and a new hub for our FTTH services. It will, of course, be 100% -48vdc. Unfortunately, the WS-8-150-DC is too small and the WS-26-400-IDC is too big. A 12 port with a floating ground would be perfect. 12 ports would cover >90% of all our sites, but without isolated grounds, we can't use them.

Either way though, great products! We've been loving them much more than the ToughSwitches (er... EdgeSwitch XPs).


Thanks!

[sirhc]

BE VERY CAREFUL the WS-8-150-DC is NOT isolated, your just lucky using radios that do not have DC negative bonded to the rail/tower or your tower grounds are not bonded to the hut rails or your not bonding the DC positive 0V to reference to the rail / earth ground?

The WS-8-150-DC power plant is no different than the WS-12-250-DC power plant other than the WS-8-150-DC has polarity correction on the input lugs via a bridge rectifier.

[troy]

The difference we found, is that there is no path from the lugs to ground on the WS-8-150-DC. I wasn't happy with it, but it's above my pay grade :)

In the meantime, I've seen mention of DC-DC power supplies to convert -48vdc to +48vdc, but I've not had any luck finding them. Any pointers?

[Coco-Bill]

The difference we found, is that there is no path from the lugs to ground on the WS-8-150-DC. I wasn't happy with it, but it's above my pay grade :)

In the meantime, I've seen mention of DC-DC power supplies to convert -48vdc to +48vdc, but I've not had any luck finding them. Any pointers?

We do have the same challange as you, with a lot of our radio sites with +ground.
What we had to, was to use an insulated dc-dc converter as a galvanic insulator.
We use one from MeanWell, type SD-200C-48 (32-76V in, 48V out, max 4.4A out).

It works ok, but with 2 Netonix on each site we also have two MeanWell, one SD-200C-48
for each Netonix. And each SD-200C-48 is fed from a separate fuse.
We pay about $75 for theese.