Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Jason Lucas]

Didn't want to watch the MINUTE AND A HALF commercial to see the content, but yes, adding an Ethercon to a shielded cable with a metal RJ45 works just fine. I did it with a Monoprice CAT6, and sparks were irrelevant to it. I think Brian Wynn's initial run of videos showed that, too.

Interesting, no commercial when I watched it.

I remember Brian mentioning taking the ethercon off and getting drop outs, then putting it back on and drop outs stopping. So I hoped that simply adding them would work.

I think it's worth trying. A regular old 150' STP Cat5e cable is only like $45 and the ethercon connector is less than $5, so if it works we'll be saving a lot of money.

[Corey Scogin]

I think it's worth trying. A regular old 150' STP Cat5e cable is only like $45 and the ethercon connector is less than $5, so if it works we'll be saving a lot of money.

Be aware that two causes of dropouts have been shown (more may exist).  One is caused by electrostatic discharge and is fixed by shielded cable with Ethercons.  The 2nd is caused by movement of the cable or possibly movement of the pairs within the cable in relation to one another.  I have experienced dropouts while using STP with Ethercons and a $50 direct burial cable.  I think they were caused by someone stepping on the cable but I can't be sure.  I have since changed to ProPlex which is a heavy duty cable where the wire pairs are less likely to move around.  No problems to report yet but not enough shows to say one way or the other.

[Jason Lucas]

Be aware that two causes of dropouts have been shown (more may exist).  One is caused by electrostatic discharge and is fixed by shielded cable with Ethercons.  The 2nd is caused by movement of the cable or possibly movement of the pairs within the cable in relation to one another.  I have experienced dropouts while using STP with Ethercons and a $50 direct burial cable.  I think they were caused by someone stepping on the cable but I can't be sure.  I have since changed to ProPlex which is a heavy duty cable where the wire pairs are less likely to move around.  No problems to report yet but not enough shows to say one way or the other.

I would hope that since this cable would be installed under the slab and never touched again, that it wouldn't need to be all that durable.

[Brian Wynn]

I can confirm that adding a ethercon connector at both ends to a premade stp cat cable fixes the dropout issues. As far as the guys with dropouts because of cable movement I would suggest you buy a more robust cable.
You get what you pay for. Don't cheap out on the cable and you will be fine.

No problems to report since I changed over to proplex brand cat5.

[Chris Clark]

So what does it tell you that it's no help to add an additional ground wire between console and stage box? I thought that would indicate where the problem might be.

That's what I've been trying to figure out. Like I said without opening up the X32 to look under the hood (I won't be able to get at the one I have access to for another few weeks), particularly at the AES50 daughter card (I assume it's a DC, not built into the main board) I am only speculating that there's some level of disparity between the AES50 shielding and the chassis itself. When it was just the test of a ground wire between the two chassis (is that plural too???) My thought would be there was maybe a limiting resistance between the daughtercard and the chassis itself, and so the shocks could still plague the daughtercard. Now with the results from Dan, that connecting the ethercons together with a separate wire (that isn't the shield of an STP cable) that it is still affected, the only logical conclusion is there's a meaningful degree of full-blown shielding happening that can only be accomplished by the proper shielding around the cable for its entire length, not just as far as the ethercons. In pure electrical theory, since the interference isn't natively coming from around the cable itself, a jumper between the two ethercons should serve the same purpose as the shield of the STP - to eliminate the electrical potential difference between the daughtercard shielding and/or chassis. Given that we're talking extremely high frequencies here though, that starts to get outside of my days from electronics training 10+ years ago, so it is possibly something might still be induced into the UTP conductors when such a shock occurs, something that can't be prevented simply by tying them together, but by having the physical foil there.

At this point, however, despite my curiosity as well, I have to agree with JR (or, at least how I read his post) that this is getting overly complicated. The fact of the matter is STP w/ properly connected Ethercons seems to work >99% of the time, whereas any other method has a greatly increased potential for failure, and it seems YMMV with any of those other methods (as I said, we ran a whole summer on UTP without a dropout, other people can't go 5 minutes...)

[Tim McCulloch]

At this point, however, despite my curiosity as well, I have to agree with JR (or, at least how I read his post) that this is getting overly complicated. The fact of the matter is STP w/ properly connected Ethercons seems to work >99% of the time, whereas any other method has a greatly increased potential for failure, and it seems YMMV with any of those other methods (as I said, we ran a whole summer on UTP without a dropout, other people can't go 5 minutes...)

We had a band carrying a pair of S16 and an X32.  They used DuraCat6e (UTP) with Ethercon.  I asked the BE if he'd had static-related drop outs.  Nope, he said.

Twice that night, he did.

[John Roberts {JR}]

Just to be clear JR is saying the problem is complicated. I can only speculate that subtle distortions to the very high frequency digital data signal cause the digital brains to lose their place when communicating so have to re-sync and start over.

Apparently problems can be caused by momentarily changing the geometry of the wire pairs inside the cable. Another source of problems is when a static charge transients hits one of the two chassis.

I can imagine multiple different strategies. A more robust communication strategy, involving some redundancy so erroneous packets could be discarded and replaced seamlessly. If the redundancy was added in parallel there would be added cost, if the redundancy was in the serial data stream this would reduce data throughput or bandwidth).  If the cable is sensitive to flexing, a more rigid cable.

I ASSume they are already using best practices for this digital communication interface. It mostly works well. Live sound reinforcement may be more stressful and less accommodating of short dropouts than other typical applications. 

JR

PS: For the poster planning to bury a cable in concrete.. perhaps bury a conduit so you could pull a different cable should that be indicated.

[Dan Mortensen]

...the only logical conclusion is there's a meaningful degree of full-blown shielding happening that can only be accomplished by the proper shielding around the cable for its entire length, not just as far as the ethercons.

In further thinking about it, my next test was going to be shielded for some distance beyond the Ethercon, with a chunk of UTP in the middle of the cable. Can a static charge cover the entire length of a cable, looking for places to get in and cause problems?

I'll have some time this afternoon to do this, and will edit this post with the result.

Edit: OK, now we are getting somewhere.

I used the Quad Proplex, with one end connecting Ethernet cable #1 to the console and #2 to the S16's (two connected with a 1' chunk of single Proplex) sitting a couple feet away.

The other ends of both #1 and #2 connected to separate Ethercon connecting barrels, with a 1' chunk of UTP in between.

With the console and S16's showing sync, zapping the S16 had no effect.

Yreka! So a continuous shield is NOT necessary.

Because I have limited STP on hand (a few 1' and one 100'), I connected a couple of the 1's together with barrels on the S16 end, and connected that via UTP to the console.

Zapping the S16 immediately caused loss of sync.

Maybe there's a reasonable length of STP that can be on each end of a UTP snake to keep the static from bleeding into the devices? That would let me use my expensive and flexible UTP.

Advice: if you don't already have expensive or difficult-to-replace UTP, just get the STP. But I am not in that position. Since I spend a lot of time on minutia anyway, when the various chunks of STP that are on their way to me arrive, I'll continue to pursue this. Or if someone has a couple of 35' chunks of STP and a couple of barrels and a snake-length chunk of UTP, you can try it first. A 35' max coil of STP on each end seems reasonable, IMHO, and means a 250' UTP could still be used. If it works.

[Jason Lucas]

Just had two services without drop outs, we'll see how the rest of the weekend goes.

edit:

Had 1 drop out during the whole weekend and it was before service started. So again, while they're not nearly as frequent as the first X32 we tried, they are still happening. Going to try a shielded cable soon and add ethercons.

[Dan Mortensen]

Edit: OK, now we are getting somewhere.

I used the Quad Proplex, with one end connecting Ethernet cable #1 to the console and #2 to the S16's (two connected with a 1' chunk of single Proplex) sitting a couple feet away.

The other ends of both #1 and #2 connected to separate Ethercon connecting barrels, with a 1' chunk of UTP in between.

With the console and S16's showing sync, zapping the S16 had no effect.

Yreka! So a continuous shield is NOT necessary.
SNIP
Maybe there's a reasonable length of STP that can be on each end of a UTP snake to keep the static from bleeding into the devices?

Today I tried a 100' STP on the S16 end, connected to a 100' UTP thence connected to the console.

Although it initially resists the sparks, enough sparking makes it lose sync, and any sparking soon after that results in immediate sync loss.

That makes me want to throw in the towel on this.

STP with Ethercons is the way to go.

If only Behringer had announced that at the beginning instead of six or nine months later....

 

[ProSoundWeb Community]