So what happened? On 9/24/2014 1:45 AM, That One Guy via Af wrote:
nope, just swapped radios, the leads are handmade crimp on N connectors that are like 4-5 years old, the lightning welded our switch to out battery backup.I dont have a problem with ten minute shutdown, but it will end up being an hour or more. I have 477 to do tomorrow so it will be the boss going. I told him to take the power supply completely out of the box so the guy doesnt claim the power supply capacitors must still have power going to the radio I also told him to not let the guy powercycle the 650 unless our radio is powered on because it will probably come back up and perform, so if our radio is powered down, of course its our radio causing the problemsThe whole point of this thread was to say the interface on the 650 is really cool and to find out about ATPC on the 650, but when i got the email telling me it was relayed to the landlord that its a combination of our radio and local interference I got really pissed.Going out on a limb and saying maybe there is not directly a physical issue looking at the fluctuations on both sides output power (-15 to 21) and receive power (-47 to -78) with an ATPC threshold set to -35 (is this the default value?) The numbers make sense, output power is ranging 36db and receive powers are ranging 31db. EXCEPT that when i was on them the remote transmit was 21 and the local rx was -78, its not correlated to the range of numbers.so our radio was on 5755 i think at the time before i moved it, 10mhz. so for the sake of argument their 650 was also sitting on 5755 for whatever reason, and we will say it was recieving at the linkplanner target of -61 and had a -35 threshold on ATPC, if my ubnt had some sort of massive fart and hit the 650 antenna with more energy than -35, could the 650 assume that additional energy is coming from the remote and and issue an ATPC power down? would that account for all the tx power and rx power fluctuations? What I dont understand is since the peak rx was -47, why would the tx have even dropped if atpc was functioning especially don as low as a negative 15Are there any bugs with ATPC with the 650? I dont recall there being any real configurable atpc parameter in the 3/500 series.On Wed, Sep 24, 2014 at 1:18 AM, Josh Reynolds via Af <[email protected] <mailto:[email protected]>> wrote:I'm not talking about your issue, per say. Just commenting on the receiver front-end overload on rockets (and other UBNT AirMax radios). I'm sure this probably happens on MikroTik radios too. EPMP? Don't have any to test. Josh Reynolds, Chief Information Officer SPITwSPOTS, www.spitwspots.com <http://www.spitwspots.com> On 09/23/2014 09:52 PM, That One Guy via Af wrote:Im pretty sure there is no atheros chipset in a ptp650 to have this issue happen and since the rocket is a backhaul, if it were deaf im pretty sure it would be hard to manage, and the fsk customers beyond it would be calling in with concerns about the lack of internet. I highly doubt that a brand new 650 would go deaf the minute it is powered on, and had it gone deaf the minute it was powered on, I doubt the spectrum would show well defined hills and valleys so clearly you can tell the channel size of the interfering systems, it would more likely either be fairly flatline or constantly in flux On Wed, Sep 24, 2014 at 12:44 AM, Josh Reynolds via Af <[email protected] <mailto:[email protected]>> wrote: This is exactly what I am talking about. Josh Reynolds, Chief Information Officer SPITwSPOTS, www.spitwspots.com <http://www.spitwspots.com> On 09/23/2014 09:31 PM, George Skorup (Cyber Broadcasting) via Af wrote:Fundamentally, no. But what you can end up with is a receiver front-end overload. This happens far too often on Rocket radios. Isn't the 650 a whole-band radio, like 4.9-5.9? I hope it would have some spectacular filtering for the fify brazillion $ they want for it. I would shut your stuff down for 10 minutes and see what happens. On 9/24/2014 12:00 AM, That One Guy via Af wrote:INTERFERENCE DOES NOT ALTER RECEIVED POWER!!! On Tue, Sep 23, 2014 at 11:38 PM, Josh Reynolds via Af <[email protected] <mailto:[email protected]>> wrote: Just wait until you have people with AF5's in your neck of the woods. No overtly OOB emissions that I'm aware of, but it absolutely crushes anything on 5GHz in it's beamwidth and freq-use range. Atheros radios outside of the band also get overloaded and CCQ tanks. AF24 is amazing and firmware will only get better. AF5... kinda not a fan at this point. Same for just about every -AC radio from every manufacturer. Time will tell how Mimosa does though, I am mildly interested in those. Josh Reynolds, Chief Information Officer SPITwSPOTS, www.spitwspots.com <http://www.spitwspots.com> On 09/23/2014 08:29 PM, David Milholen via Af wrote:Since I run several of these in our networks as well as the new 650 units. Ubiquity has a bunch of OOBE even that low if the power requirements are not being met. I have had ubiquity on my tower colo'd with a ptp230 5.4 unit and I set the ubiquity in the 5.2range and it completely knocked off our ptp230 link. I had to turn the power way down below even min powerlevels before the 230 would come back up. If by turning your system down and levels do return to normal for them. Then I would take a closer look at your config on your AP to see if you can tweak it to meet standards and at the same time not mess with them. I tried running a ptp link colo'd on my tower using ubiquity and the Out of band noise was incredible. I had 50' sep and andrew dish with at least 120 deg out of center. The Ns5 was the one with 3' dish. Another thing to try is to get someone who make gutters and use sheet metal to make an extended shield placed between the ubiquity and the 600s On 9/23/2014 7:05 PM, That One Guy via Af wrote:but i do really like the interface on the 650 On Tue, Sep 23, 2014 at 7:04 PM, That One Guy <[email protected] <mailto:[email protected]>> wrote: This is really beginning to irritate me, Now the guy who replaced the gear is still blaming us for the problems here, I moved the ubnt gear clear down to like 5.1 or whatever the lowest channel is, the spectrum at this and the remote site are deplorable. The Signal/Noise ratio is moving around on the ptp650 and the Vector Errors are off the chart, but he still wants to blame our equipment. I can tell you it boils down to an improper system repair post disaster. I pulled screen shots, both before and after I moved our channels, showed them the issue with their own colocated radios, turned on assymetric channels, yes, they were running symmetric in a high noise environment, nothing could go wrong there, right? Now tomorrow, my boss is going there to unplug our radio, taking our customers down. Im betting some utter nonsense like capacitant power or our antenna shape ends up being to blame here. I know ubnt is shit and bleeds noise allover, this particular radio is a rocket m5 with the 30db dish and the shield kit. The link is 90 degrees off both of theirs (ours is west, they have one north and one south) I believe we have 30 foot vertical sep between it and their closest radio. I can see how a rocket would magically destroy the whole 5ghz spectrum and not have performance issues itself.I even cycled the UBNT radios to make sure that they actually did change channels. ATPC power ranging not matching current TX output and RX doesnt make any sense to me. Interference alone will not alter RX power unless its very very notable. And then to top it off its said it would be better to move completely off the band to 3ghz since it cant interfere. Yeah, great fucking idea, lets take the only semi clean spectrum left and burn it on a backhaul thats performing as it should because other people dont know how to troubleshoot their own damn gear. But the kicker to that would be "oh, you must still be interfering, that m365 is actually a 5ghz radio downconverted how bout this, climb the damn tower and fix the fuckup fucking meh On Tue, Sep 23, 2014 at 5:04 PM, That One Guy via Af <[email protected] <mailto:[email protected]>> wrote: Im not doing anything, this is a not my chair not my problem issue. This strike blew everything on the tower, if it was electronic, it cooked, the switch was sitting on back of the APC and welded to it even tripped the breaker Im just curious with these if theres any issue with the ATPC on these bas boys On Tue, Sep 23, 2014 at 4:42 PM, David via Af <[email protected] <mailto:[email protected]>> wrote: Inspect the cables or at lease switch one or both out at one end and see if a prevalent change is made. Could be a feed horn but unlikely I would shoot for pigtails first. On 09/23/2014 02:38 PM, That One Guy via Af wrote:I just got done troubleshooting a 650 link for our landlord we are coloed with on a couple towers. I had not looked at the ptp interface since the 500. This thing is freaking beautiful, and I never compliment anybody, especially on a web gui. Sooooo much information, so easy to find. one question though, They have atpc set to -35 on these, does that basically turn atpc off, or could it cause a problem? Im pretty sure they have a loose antenna or damaged feedhorn/patch cables (this was a lighnting replacement of a ptp500, reusing the cables/feedhorn) The system statistics showed a variation of received power ranging from -47 to -78 with a peak of -110 , -78ish being current. Transmit powers show a variation of -15dBm up to 21 dBm (I did not notice the negative value at first). This would account for the range of Received power except When the Status screenshots were taken, the transmit power on both units was at 21 dBm with a 77/78 receive power on each side. If the output power is accurate, the receive power on the remote end would be at the peak, not the mean.-- All parts should go together withoutforcing. You must remember that the parts you are reassembling were disassembled by you. Therefore, if you can't get them together again, there must be a reason. By all means, do not use a hammer. -- IBM maintenance manual, 1925-- All parts should go together without forcing.You must remember that the parts you are reassembling were disassembled by you. Therefore, if you can't get them together again, there must be a reason. By all means, do not use a hammer. -- IBM maintenance manual, 1925-- All parts should go together without forcing. Youmust remember that the parts you are reassembling were disassembled by you. Therefore, if you can't get them together again, there must be a reason. By all means, do not use a hammer. -- IBM maintenance manual, 1925-- All parts should go together without forcing. Youmust remember that the parts you are reassembling were disassembled by you. Therefore, if you can't get them together again, there must be a reason. By all means, do not use a hammer. -- IBM maintenance manual, 1925---- All parts should go together without forcing. You mustremember that the parts you are reassembling were disassembled by you. Therefore, if you can't get them together again, there must be a reason. By all means, do not use a hammer. -- IBM maintenance manual, 1925-- All parts should go together without forcing. You must rememberthat the parts you are reassembling were disassembled by you. Therefore, if you can't get them together again, there must be a reason. By all means, do not use a hammer. -- IBM maintenance manual, 1925--All parts should go together without forcing. You must remember that the parts you are reassembling were disassembled by you. Therefore, if you can't get them together again, there must be a reason. By all means, do not use a hammer. -- IBM maintenance manual, 1925
