01-26-2019 08:41 AM - edited 01-26-2019 08:42 AM
Hello, I have 20+ Rocket Prism 5AC in the field, and most powered by ERPOE-5 routers or ES-8XP switches.
Most routers are only powering 1-2 Prism's and 1-2 PowerBeams.
Most Switches are 3-4 Prism's and 0-4 Power Beams.
Most are not temperature controlled and have very short cable runs to the equipment.
We have a small number of sites where cable runs to the Prism's are 160+ feet outdoors using ToughCable Carrier cabling, and when the temperature drops below -10F the radios start rebooting and/or having trouble communicating at full duplex/full data speeds. The fix we have found and has been proven to correct the issue has been to disable POE on the router/switch and install standalone power supply that originally came with the units. Of the 2 sites that had the problem, one was an ES-8XP heated to ~60F, and the other an ERPOE-5 heated to ~25F. The ES-8XP cable runs are 92M each to 3 Rocket Prism 5AC. The ERPOE-5 has a 54M run to a single Rocket Prism 5AC. All of the Prism 5AC were operating at higher EIRP channels. After replacing the power from the switch/router with the standalone we have had no reboots or dropped packets at 1gbps.
My question is why is this only happening when the radio and long runs cabling get to -10F or below, but the switch/routers do not and the cable runs are over 50M; It does not happen in any other scenario except where the cable runs exceed 54M. We have several 30M runs with the same equipment to both heated and unheated switches/routers that it does not happen to.
01-27-2019 07:00 AM
Frankly this doesn't make a lot of sense - not saying it's not what you are seeing, but the engineering doesn't track...
Cold should actually decrease the resistance of the cable slightly, which would in turn decrease any cable losses with longer runs. The switches themselves don't seem to be the issue since you have one staying at around room teperature but showing the same symptoms (plus all active devices generate their own heat so internally will always be above ambient internally). But if the only change to the system is using PoE injectors instead of the switch PoE, then that does indeed point to the switches somehow. And if the cable runs are just powering a single radio, the switch ports (with their 11 watt PoE limits at 24V) should be more than adequate...
We have a ton of radios in production, many hundreds of them, and all the APs (everything from RocketMs and NanioBridges to AC Rockets/Gen2s to LiteBeams and LTU radios) are powered from switches (only a few of the older Toughswitches/ES5XPs but lots of ES-8-150s, ERX/SFPs and ES10XPs) and we've never seen this kind of thing. Many are in unconditioned NEMA boxes outdoors, but we seldom reach temps below 0F these days, so not the same conditions you are seeing. I do know a lot of people who operate in way colder climates (like Canada) and I think we would have heard from them if this was a widespread issue...
How are you powering the switches? I wonder if they are not putting out (or being fed) the full 24.0VDC they need...
"Humans are allergic to change..They love to say, ‘We’ve always done it this way.’ I try to fight that. "Admiral Grace Hopper, USN, Computer Scientist
"It's not Rocket Science! - Oh wait, Actually it is... "NASA bumper sticker
"Just because you can do something doesn't mean you should."my mantra in the Programming classes I used to teach once upon a time...
01-27-2019 07:11 AM - edited 01-27-2019 07:22 AM
I'm confused by it as well. Both sites have battery backup/conditioning and temperature controls.
Site A (60F):
ES-8XP - AC Power
Config port -> AF24 Config Port
Ports 1-2 24v Powering PBE-5ac-Gen2
Port 3 24v Powering PBE-5ac-620
Ports 4-6 were 24v Powering Rocket Prism 5ac Gen2, now those are on oem psu
Port 7 was PBE-5ac-gen2 already on its own psu while troubleshooting a crashing issue
Port 8 unpowered to AF24 Data port
The only ones that rebooted were the Prism 5ac Gen 2. All dishes between 150-200ft. All Prism sectors 280ft. The only ones that rebooted were the sectors, but only when powered by the switch. This site has been setup like this for multiple years. Just newer feed horns with newer equipment.
Site B (25F):
ER-5POE - OEM AC-DC Adapter
Port 0 was 24v poe to Prism 5ac gen2, now is on its own psu
Port 1 24v to PrismStation 5ac
Port 2 no poe to additional switch
Port 3 24v poe to pbe-5ac-gen2
Port 4 no poe to customer equipment
Only radio that rebooted was the Prism 5ac, even though all radios are around 170ft. Again, been setup for years and this is the only time it has happend.
All that being said, this is the first time since the new AC gear went up on the towers that it has gotten that cold here. These are the only 2 instances of this happening out of 35+ sites so its by no means common, i would just like to know why its happening and if it can be corrected.
01-30-2019 04:53 PM
I am also having one of my RP 5AC Gen2 radio reboot. It was on an OEM power supply, one of the 24v 0.5 amp ones that came with the older rocket radios. There is about 200 feet of cable between power supply and radio.
After reading this thread, changed the power supply from another RP 5AC Gen2 that was not rebooting and came with the newer white power supplies, which I think is a 24v, 1amp. I have three radios on the tower, and only one of them is rebooting. It is currently -11F here and getting colder over night.
Any suggestions that don't involve climbing a tower in below zero temps to swap out the radio with another that may or may not reboot would be greatly apprecitated.
02-11-2019 09:44 AM
02-11-2019 09:58 AM
The temperatures haven't dropped like that again. After changing the radios with the 170+ feet cable runs to the OEM power supplies that came with the Prisms they held strong. I did have one reboot around -34, but only once.
02-12-2019 09:58 AM