GLD-80, AR2412, and AB168 QOS error

GLD GLD Troubleshooting
1

The rig is a GLD-80 connected to an AR2412 with the AB168 connected to the expand port on the 2412.

My issue is I am intermittently losing audio from the AB168 with QOS errors matching the time stamps of the dropouts…

The cable going to the AR2412 is a CAT5 and passed on the cable tester. The cable expanding out to the AB168 is a combination of CAT6 and CAT5 hitting a patchbay and all of those cables and patchbay ports and stage pockets involved all passed on the cable tester

The console was on a previous version of firmware and this was happening (can’t recall which version) so I tried updating the firmware and this didn’t change anything. it is very intermittent…

While I was working I plugged an iPod into the AB168 so i could hear if there were any dropouts… i worked on my computer for 4 hours or so with no issues or dropouts… few minutes into band practice and i lose all the drums (the AB168 is our drum snake box in this setup) and get a QOS error.

Now, here’s some more info. The expand port doesn’t run straight into the AB168… it goes into a CAT 6 patch bay and where ever the kids want to setup the drum kit that day is where it gets patched to. It’s not an ethernet switch it is actually a patch bay where each port is hard wired to a specific spot on stage. I tested each port with a cable tester and they all check out fine. Everything i’ve tested including the jumper cables and looms all test out just fine and dandy but for some reason I’m getting these errors and dropouts.

All my tests were done with all of the amplifiers and keyboards on stage plugged in and powered on to re-create a somewhat voltage accurate scenario for testing… The only difference i can think of is vibrations from the drum kit during playing but i can’t see that causing too much of an issue. For what it’s worth the AB168 is on its side and when I ran a home run from the AB168 to the expand port of the AR2412 (as a last minute “OH SHIT we have to hurry up” fix) it solved the problem. I just don’t understand how stuff is testing fine on the cable tester - working all day during testing/troubleshooting, then screwing up as soon as practice starts… and then magically getting solved by a new cable… The cable tester is not broken either… it’s helped me identify a few bad cat 6 cables during my week at the church.

Does anyone have any clue???

I was hired in to help this church out with a bunch of issues and ive been doing just fine getting rid of lots of bugs but this one has me scratching my head!

Is it possible that going from cat 5 to cat 6 is screwing with the setup???

ie/further calrification: out of the AB168 is a 5 foot shielded/grounded cat6e into the stage pocket. The backside of the stage pocket (under the stage) is a 50’ shielded/grounded cat6e which connects to a specific port on the patchbay… then on the front of the patchbay i jump into the expand port of the AR2412 with a 3’ Cat 5 cable… could it be that going from CAT 6e to CAT5 is fucking with it???

2

In theory you should be ok. That said I have had situations where a cable tested good then caused problems. For no apparent reason. You have a lot of connection points. Like i started with, in theory it should work. Not sure if the addition of all the points is equivalent to a long cable. I have ran 6 shielded over 350’ from QU series to AB168x2 with no problems. But only one connection point in between.

3

I think it is a combination of cable quality, connection points and cable quality.

Every patch/connection gives a small amount of loss.

Cable tester is only that there is connection, not the quality of connection.

Richard

4

I believe you need a minimum of a cat 5E.

5

What kind of cable tester do you have? Just a cheap continuity tester or some expensive one showing you quality of cable?
What kind of connectors does the patchbay and ports on stage have? Normal RJ45 or Ethercons?
Can you perform a “speed test” on your connection from FoH to stage with no devices attached?

I am using Neutrik Ethercon Cat6a for cables, patchbays and so on plus mostly Cat7 cables and had no troubles so far with the GLD.
But I run into troubles when using normal RJ45-connectors for use with our HDMI extenders (which have no Ethercon) - they clearly have a mechanical problem which leads to dropouts when moving but sometimes also without movement. So it seems the bandwith/quality is just a little bit to low when not plugged in correctly. A cheap continuity tester can’t find that problem.

Maybe dSnake protocol uses less bandwith when using only two channels for your iPod but increases when having multiple microphones. And then the quality of the connection is just to low.

6

An ethercon is just a regular RJ45 inside a metal shell but does offer a little more solid connection. I will say I’ve had problems with some RJ45s. I just finished a video install where we had all kinds of problems getting good connections. I finally realized I had a bag of bad connectors. They were premium but bad. I ditched them and tried some others I had and they worked fine.

I also did a job few years ago where we ran a 300’ foot CAT5 for an internet connection for streaming. We had to extend it about 3/4 of the way with a terminal block. My tester showed the whole path fine but we had consistent problems. The church paid an IT company to test it with a fancy unit. It showed a severe quality loss in bandwidth through that block. Eventually the church installed a fiber optic line.

7

A graphical TDR can show you the quality of your interconnects, shorts, breaks, and any other changes in impedance along the cable. There’s a small learning curve to using one, though :slight_smile:

DC