The regulators in many radios chop the incoming POE current, run it through an inductor or transformer in a buck smps config and then filter the output. That chopping frequency along with the sharp edges might be radiating back down the cable. They do test for conducted EMI during the FCC approval process but it just has to be under a certain limit and that limit may still be strong enough to interfere with things.
From: Craig House Sent: Friday, March 1, 2019 7:52 AM To: AnimalFarm Microwave Users Group Subject: Re: [AFMUG] Interference with 911 repeaters POE when it’s powered affects the repeater equipment regardless of whether it’s from the manufacturer supplied power supply or a net tonics switch. Andean even one case it was from an old CMM micro. It would immediately cause noise as soon as the power lights on the radios would come on. What I did not try was seeing if there would be any noise while the cable had power but not plugged into the radio. I would tend to think if there is no power draw and no ethernet link there would not be interference. But now that I think about it most of those radios don’t establish an ethernet link that early in the boot up process. So I’m more inclined to believe it’s the power draw from the radio creating the interference but it seems to be present regardless of the POE device or the length of the cable Sent from my iPhone On Mar 1, 2019, at 08:39, Josh Luthman <[email protected]> wrote: POE when it's powered, but does that mean the POE on the ground is effecting gear on the top or do you have cat5 plugged in and it could be the DC in your cable to the radio? Josh Luthman Office: 937-552-2340 Direct: 937-552-2343 1100 Wayne St Suite 1337 Troy, OH 45373 On Thu, Feb 28, 2019 at 11:15 PM Craig House <[email protected]> wrote: I'm pretty sure it is the POE. The problem shows up on the spectrum analyzer as soon as the POE ports power on. Before the radio has even the remotest chance of fully booting up. I'm not sure however if that would be the case if the radio was not even on the other end of the cable that had POE power. If I recall we eliminated the noise when we removed the POE CAT5 from the radio even when it still had power. Possibly due to power draw on the cable I suppose. I think we probably should look at the double shielded cable I'll try that first. We already use metal tower boxes. Thanks again Daniel. You have always been a big help spanning the last 15 years. Craig ---------------------------------------------------------------------------- From: "Daniel White" <[email protected]> To: "AnimalFarm Microwave Users Group" <[email protected]> Sent: Thursday, February 28, 2019 10:06:57 PM Subject: Re: [AFMUG] Interference with 911 repeaters If these are all water tanks, you would probably be best served running Fiber + DC to a small box at the top and use double-shielded cable from the Netonix to the radios. I would use a metal enclosure so no noise from the switch should be an issue. Some FP-1204 from Shireen would do the trick for the cable run. This would be far easier and less costly then conduit, liquid tight, or running brand new cables to each radio. Of course those solutions only work if it is the cable, and not the radio itself causing the issue. Daniel White Co-Founder - Business Development & Operations direct: +1 (702) 470-2766 Craig House wrote on 2/28/19 20:44: It doesn't seem to be a particular radio. At one point we were convinced it was some old 900 FSK radios that we removed eventually. At one point we ran coax up to the omni and left the connectorized radio at the base of the tower which solved that problem at that time but it was the cable not the radio that was the problem. We later added other UBNT and Cambium radios to the tower and the problem came back. Added ferrite clamp beads and that reduced the problem but didn't totally eliminate it. It does seem that the problem is consistently with on particular guy that is installing the 911 equipment. No other towers that we have equipment on are as drastically affected. Maybe it is his equipment or method of installing but that doesn't solve our issue since the city and county are wanting us to fix it since we added the equipment that makes it not work. We are not using POE injectors in most cases. We used old CMM3 when we had the first problem but that has not been the case at the other sites. We have Netonix 24 and 8 port switches at the other sites. I am thinking about the liquidtite but that is going to be a pain to do on long runs. Grounding the shielded cable will be problem midspan because this cat5 is single shielded with the mylar inner jacket that is easy to cut and would probably not get a good ground. So replacing all my networks links BH cables in hopes that fixes it would be a lot of down time potentially for a "Hope it works" attempt. I dont think we have our lines anywhere near the feed lines on the water tower at one site. Putting them all in Liquidtite is doable but on one water tower that means about 20 runs of it and I'm afraid of how messy that will end up looking vs just the 20 cat5 runs. How do the coils of cable help? you cant put them in the liquid tite if they are coiled. Seems to be counter to the idea of containing the RF but I'm not an RF engineer. Thanks guys -------------------------------------------------------------------------- From: "Daniel White" mailto:[email protected] To: "AnimalFarm Microwave Users Group" mailto:[email protected] Sent: Thursday, February 28, 2019 9:19:23 PM Subject: Re: [AFMUG] Interference with 911 repeaters Craig, While I've not experienced this on a network I was responsible for, I have helped a few times while at various manufacturers troubleshoot issues like this. Some tips in addition to what Chuck has written: - First make sure to localize which radio it is. Some radios are particularly nasty in these cases. Go to the site and turn off everything, then turn on each radio individually. You may find it is a common radio at every site, and replacing those radios with another brand/type might be easier. - Bad switching power supplies can sometimes be the culprit. An easy change is to replace the power supply/PoE injector.. ideally with a different model (instead of the vendor supplied one try maybe a McCown Tech injector). This can be the cause of the radio or cable affecting the repeater. - If the interference is adjacent or a harmonic notch filters or band filters on the 911 repeater may be an easy solution - Double shield cable works well if the noise is from the cable and not the radio. Shireen DC-2022 or Superior Essex BBDGE would be best bets... and these are also easy to ground with GKCAT5 from SitePro (or something equivalent): https://www.sitepro1.com/store/cart.php?m=product_list&c=14. For best effect, use every 50ft of cable run and wrap around the outside double shield. While these issues suck to troubleshoot... narrowing it down to the radio causing issues is key. Good luck Craig (hope your doing well BTW!) Daniel White Co-Founder - Business Development & Operations direct: +1 (702) 470-2766 [email protected] wrote on 2/28/19 15:20: I would first attempt to discover if it is the CAT5 or the radio itself. Ask the 911 guys if they have a spectrum analyzer they can lend you to puzzle this out. Shielded CAT5 helps. Putting in coils at the ends helps. Ferrite chokes help. Some run it in liquidtight or full metal conduit. You can put fiber media converters on the ends of a CAT5 and perhaps fix it. But every single ethernet device, every switch, hub, router, radio does radiate noise. Just because they passed FCC certification tests does not mean they are noise free. The 911 repeater probably has a high gain omni and probably a super low noise pre amplifier too. So it will naturally be more sensitive to noise than a normal 2 way repeater. Isolating it to a particular radio or cable is a good place to start. -----Original Message----- From: Craig House Sent: Thursday, February 28, 2019 1:45 PM To: [email protected] Subject: [AFMUG] Interference with 911 repeaters I think I’ve posted this question on here before but I want to try again. We have multiple water towers where we have various types of equipment a mix of Mimosa cambium and ubiquity products. It seems that on many of these water towers our cat five power over ethernet creates RF interference for the repeater. In one instance we were able to install magnets that clamped around the cable which did help enough they stopped complaining. In other cases we’ve attempted to move our equipment from the top of the water tower down to the catwalk to create some distance and installed the ferrite beads with little success at all. I’m tired of playing with ideas that might or might not work. Can someone give me advice that will solve this problem once and for all. I understand that grounding the CAT5 might help I’ve also been told that putting our wires inside of conduit might help or shortening the wires might help. I don’t want anymore might help I would like someone who has actually done this to be able to give me some advice Sent from my iPhone -- AF mailing list [email protected] http://af.afmug.com/mailman/listinfo/af_af.afmug.com -- AF mailing list [email protected] http://af.afmug.com/mailman/listinfo/af_af.afmug.com -- AF mailing list [email protected] http://af.afmug.com/mailman/listinfo/af_af.afmug.com -------------------------------------------------------------------------------- -- AF mailing list [email protected] http://af.afmug.com/mailman/listinfo/af_af.afmug.com
-- AF mailing list [email protected] http://af.afmug.com/mailman/listinfo/af_af.afmug.com
