Long run power injection

The B24 looks very flexible with regard to powering, and the fiber data port has no serious distance limit. But I have run into a place where the cable run seems to be about 500 feet long, too long for standard PoE. Since it’s just power, and the data can be on fiber, would a PoE++ 802.11bt 4-pair ~50-volt injector work? Or does someone recommend a different device to do the injection up top of or inside a pole? Thanks.

I have never gone that long, technically you could do the math and calculate your losses to make sure you don’t inject more power then the B24 can handle, but my preference would be to have a PoE injector closer to the radio. I would double check the spec sheet to make sure that the B24 can take 4 pin power… I think it can, but I am not running mine off of it.

You might look at an option called a mid-span injector. The intention of this is, it is connected to the primary POE source, and it "enhances? (not sure that’s the best description) the power to make it work for longer runs.

I think the caveat here is the 802.11bt needs. I’ve not seen anything specifically listed for that power level. Not that it doesn’t exist, just that I’ve not seen one personally.

You might also take a look at TYCON as they have quite a few options in broad terms for unique situations like yours.

1 Like

The problem with mid-span is that there probably isn’t access – somehow the conduit from the building to the pole (a light tower at a stadium) is going to be longer than originally anticipated. So the mid-span point would have to be the tower top, with parallel lines pulled in the conduit. From the raw numbers, though, the voltage drop from the Mimosa 56-volt supply over Cat6 802.3at should leave enough to run a B24. I suggested that they try this on the ground (spool of cable) before going into the field. Using 4 pairs would be nicer (the B24 seems to support that) but 2 pairs seems to meet the requirement.

Alas I do have to concur on the installations likely situation… Where’s Nicolas Tesla when you need his wireless power!

We’re always thinking outside the box.
The suggestion of testing on the ground is awesome, and simple.

Light poles tend, not always, to have power at them… I would consider powering the radio off the power feed.

Or, you could just be brave and run 120 AC along with the fiber then do you Power insertion at the light pole.

TBH, both are not great ideas, but they at least don’t make you try to keep a 500 ft Ethernet connection live…

When I have really long runs, I run either of a 2 conductor 12Ga or 10Ga outdoor lighting cable (Lowes carries this) up the tower to a Mikrotik PowerBoxPro. I cut the power plug off the Mikrotik POE and solder it to the power cable. The PBP can run on either 24V or 48V(56). Then wire nut the bottom end of the cable to whatever POE you are using, wire nut it because the power to the PBP is polarized meaning you have to have the + go to the + and the - to the -. Once you have it right for sure (by logging into the PBP) solder it to your POE leads. I also use the power cable Ubiquiti sells and it has an extra layer of UV and water shield and the conductors are color coded. These power cables eliminate a lot of the voltage drop you will experience with Ethernet cable. With the PBP you can see what the voltage is at the top of the tower or the end of the run without climbing up there with a meter. I have also used variable power supplies instead of POEs so that I can fine tune voltage as desired.

Then plug you fiber either directly into the B24 with a short ethernet jumper for power or plug it into the PBP and do a very short ethernet jumper over to the B24 for power and data. I like having a smart box up top and will hook up the radio and PBP depending upon my requirements.

The PBP has 5 ethernet ports 4 of which support POE out, if you put a big enough power supply on it you can power several more devices. You can only do 24V or 48V but not both at the same time. The PBP is about a hundred bucks, the cable more.

1 Like

Have you tried any of the newer Ethernet cables that are coming out and allow you to exceed the 100m limit? I think, but don’t know for sure, that they’re using a larger wire gauge to accomplish this, but it should help with PoE also. I haven’t tried it, so I can’t comment beyond what I’ve heard so far.

Actually @Wayne2 has a decent idea. Cat7 has a significantly thicker gauge of wire and the tighter twist should help with noise cancellation.

I was thinking about it and something like the GPEN 21 from Mikrotik would be another option. Run power like @Dave1 mentioned and fiber then power the GPEN 21 through it’s barrel jack and use the PoE output on the GPEN 21 to power the B24… Should work…

Thanks, all, these are interesting ideas. I do notice that Cat7 is 23ga and Cat6 can be 23 or 24ga, though when I look at the specs the DC resistance is usually about the same on either size, or not different enough to matter.

There is, of course, power on the pole – at least when the lights are on – but it’s not available. Stadium lights aren’t like street lights, they’re a huge array of bright who-knows-what – do they make big LED arrays now? Probably – it’s a brand new stadium. The power lines might be very noisy, though, so the signal will be best on fiber.

It’s not obvious to me how the GPEN will work with a B24 but I recommended testing on the ground first, and if that fails something like that might be necessary. A PBP puts another router in the circuit, which complicates matters. I’m advising a client (a broadcaster) who is in turn depending on the WISP who serves the stadium to be cooperative. So nothing too fancy.

I wonder if the 300m limit on Ethernet is electrical or some kind of timing, because a Cat 6 or 7 wire might work at 100 Mbps if it’s just electrical issues. Nice for backup.

Here’s an example of the cable I was mentioning:

Thanks! I wasn’t aware of that model cable, with 22 ga conductors. I see that it comes in an OSP STP form. I will recommend that. I don’t see it from the usual distribution sources but Paige Electric will probably tell us where to get it.

The video from Lawrence points out that the Cat6 spec officially allows conductors between 22 and 26 ga, but 23 and 24 are most common, and this may be the only 22 ga one out there.

This is a basic issue with DC power. Carriers do this all the time. No, you don’t need to run AC or some funky solution. You just need 2 wires up the tower for DC power. You SHOULD certainly use proper surge protection and grounding.
The problem is that the POEs that ship with Mimosa can barely handle the current required. And at about 300ft, you are running into limits of the 802.3af/at protocol. (And minor voltage drop.)

Source this power supply TYCON POWER P/N TP-DCDC-4848G-HP 50W.
Then source the PN “C000000L184A” It is a DC to RJ-45 power adaptor to convert from 2 wire to the radio POE port.

wavonline.com can get these parts.

I’ll be wrapping up at Shentel on 12/31/21. So if you need a microwave engineer, drop me a line.

Charles Crizer | Senior Network Engineer II
Wireless Engineering - Microwave Engineering & Design
Office 540-984-5286 | Mobile 202-210-6346
Work: Charles.Crizer@emp.shentel.com
Personal: ccrizer@hotmail.com

PO BOX 449, 500 Shentel Way, Edinburg, Virgina, 22824
FCC Commercial Radio Operator’s License w/ RADAR (GROL Radar): PG00032312
Amateur Radio: KF4MNE
Microsoft Certified Professional MCP, MTA, MOS
U.S. Navy – Missile Radar (FC)

Timezone = Eastern, (EST UTC-5, EDT UTC-4)