Hi Jonathan

There are a number of possible causes for a problem like this. Here are a few ideas:

Try looking in the Touch Screen event logs. Before a disconnection, are there ACE QoS warnings?

How long and what type of cable are you using for your ACE link? A cheap or faulty strobe light could emit a lot of rogue radio energy, which with unshielded or cheap cables, could cause electrical interference on the Ethernet lines, which in turn could cause a disconnection. We have a list of recommended cables that we have tested for use with ACE here https://www.allen-heath.com/ilive/system/cables. These have been tested at 120 meters with no errors reported even with 2 kV mains line transients being injected into the rack or surface PSU. Even with good cable, a damaged or poorly fitted RJ45 connector on the cable can have a significant impact on the noise susceptibility of the ACE link.

Proximity of power cables to a data cable can also be important. If you have a noisy, high current power line running parallel to an Ethernet cable, this will increase the chance of noise on the power line affecting the Ethernet link. In this case it does not matter whether the power line is on a separate circuit.

In general, the connection should be fine even with quite a few errors; perhaps even 10s of errors per second could be OK, but you might notice a meter stutter, or sluggish response from the control surface. Did you notice this kind of symptom at the show?

If you are in this sort of situation again, you could try having a look in the Mixer Preferences page of the Mix Rack Setup screen. Select ‘Port A/B Setup’, then press the ‘Advanced’ button at the bottom of the Port A ACE area. You will then see various counters that are updated every second. There is one counter (Error Count 1, Error Count 2 is only relevant if you are using an ACE Option Card) showing you the number of errors per second for each of surface and rack (scroll down to see the second one). If these counts are high, this means that there is some problem with the ACE link that could cause a disconnection.

Hope this helps.

Cheers

– Jeff, A&H

Hey Jeff, thanks for all that background info, it makes understanding the system easier in terms of sorting a cat5 cable issue.
For the ‘badly fitted connector’ scenario, I have a cable tester that just checks continuity across the 4 pairs. Is this adequate to find a fault or are we needing something a bit more involved to check a cable….. Say something that measures the impedance across the cable and fitted connectors?
My system currently works fine in this respect but I’m always aware that it’s possible to be on the limit without knowing, and then an extra long cable or other variant pushes your luck too far. And then it’s Game over!
Jim

R72 & IDR32

Hi Jim

A simple continuity tester will not identify subtle faults in the cable like a badly fitted connector. For this you need a more sophisticated cable analyser, but these are expensive. You can get some reassurance that you are not on the edge of the connection failing from the ACE QoS indicators. On the system info window on the touch screen (the page with the clock on it), you should see traffic lights associated with the ACE link. If these ever change to orange or red, this means you have experienced a few errors on the ACE link. These should also be reported in the touch screen event log. There isn’t that big a line between ‘working perfectly’ and ‘starting to go wrong’ with Ethernet, but if you never see any errors on the ACE link, you can be reasonably confident that you have a solid setup. If the lights are going orange or red, you can look at the exact error rate (see my earlier post). If the reading is rarely zero, this would be a fairly strong indicator that there is an issue with your cable.

Jonathan; Moving the strobe light to be powered from a different room does not necessarily rule it out as the cause of the problem, as it could still be emitting a lot of radio energy. If you don’t get the kind of sluggishness and dropouts at other venues, then this of course points to some issue with equipment at this particular venue. It could also be a combination of two problems that by themselves do not cause an issue, for example a slightly substandard cable, along with a noisy piece of lighting equipment.
This information doesn’t help you much, but my main point is that although I would always recommend using a good quality cable, I can’t predict whether or not a better cable will solve this particular issue that you are having.

Cheers

– Jeff, A&H

Hi Jonathan

That tester looks like it just checks for continuity, but does so between incorrect wire combinations to identify a cable that has been wired up the wrong way round. I was thinking of something more like this https://www.flukenetworks.com/datacom-cabling/copper-testing/dtx-cableanalyzer-series, but somehow I don’t think the average user is going to want to fork out for that! This is part of the reason we have an approved cable list, as it is almost impossible for the end user to predict how well a particular cable will work. If you want to try out some other cable, the cheap solution is to soak test it, to see if any errors are reported by the software when your system is running in your target venue. If you can do this with high power and/or noisy equipment like strobe lights running, this sort of test should give you a good level of confidence that the system will work well in the real show.

In general, you should not encounter any problems due to electrical interference or noise if you are using a good cable. The fact that you had some problems in this particular venue indicates to me that there was something wrong, but as before, I can’t predict whether a better cable will solve this particular problem.

Hope this helps

– Jeff, A&H

Hi Jonathan

By ‘soak test’, I just mean running the system for a period of time, to reassure yourself that it is working properly. If you can have a system running for a couple of hours during a period that a problem would not matter, and experience no problems, this will give you more confidence that you will not get problems during the show, when it does matter. Soak testing is a good way to discover problems that do not have an easily measurable or predictable cause.

The Belden 1305A cable does have good performance, but our system does not pass EMI regulations using unshielded cables, so we cannot recommend it.

Cheers

– Jeff, A&H