ProSoundWeb Community
Sound Reinforcement - Forums for Live Sound Professionals - Your Displayed Name Must Be Your Real Full Name To Post In The Live Sound Forums => LAB: The Classic Live Audio Board => Topic started by: Jason Branden on May 15, 2018, 12:11:04 PM
-
**EDIT--> Solved
Hey Folks,
First time poster here. Sorry in advance if somebody has touched on this topic before but I haven't been able to find a post that covers the exact issues I've run into and am looking to see if anyone has heard or otherwise run into this bizarre issue.
The issue:
I ran a stage at the Hyatt Hotel a number of months ago where I had ZERO dropouts throughout bands, guest speakers, people moving around on and over the cables, etc. All day, No issues. Later that evening, there were four "DJ's" lined up. First 80's style set was fine. Second set were EDM style music events. This is where I ran into massive issues. Every time there were (persistent) low end frequencies running though the PA I would lose AES50 sync and get dropouts. sometimes one right after the other The only thing I could do is turn down the volume and/or cut the frequency at 40hz (and even then I would sometimes get a dropout at LEAST every 15mins)
Next day, everything was fine. -Until the DJ sets again. Same issue. Wash, rinse, and repeat. lots of troubleshooting, no resolve.
Troubleshooting steps taken
* measured power at amp rack to check for voltage dip. None. Rock Solid.
* measured (mains) power at supply (150A 3 phase 208) -all phases during event. -Perfect
* measured power going directly TO the DL32 -rock solid
* checked ground bonding between everything in the rack. -completely fine.
* checked the integrity of the shielded cat5 line (every line and the shield for continuity) -even replaced with a spare. -all was fine.
* moved FOH power to and from house power (checked voltage AND ground differential between the two supplies) -All looked fine.
* monitored voltage differential between DL32 AND M32 during event -Nothing stood out.
* strapped extra grounding between amp racks and FOH for grin giggles. -no difference.
* pulled DL32 out of amp rack in am attempt to limit electromagnetic interference -no change
* moved from AES50 "A" to AES "B" on both the snake and console. -No change
* put the DL32 on its own power supply outside of the rack mounted PDU in the amp rack. (right to mains power) -No change.
(loss was observed by readout on the console going "red" then "green" during these outages so for sure was a AES50 dropout)
I'd be at a complete loss except I've had some time to reflect on this a little bit and have a few thoughts as to what might have been happening, but still don't feel warm and fuzzy with my assumptions.
Before I get into that, I'm curious as to what everyone is doing with their DL32's/DL16's. Are you guys putting them in the back of your amp racks or otherwise near amplifiers and speakers? I can say that I've run the X32 board and S16 in this same same rig without issue whatsoever, same event previous year, same music, etc. That said, I upgraded this rig to a Midas M32 along with its DL32 counterpart last year and that's when the issues started. (same configuration, new board and snake) As you can imagine, I am a little dismayed by my upgrade.
relevant equipment;
FOH:
Midas M32
AMP RACK 1:
MIDAS DL32
Driverack Venue 360 (running in analog NOT AES input mode)
Driverack 260 (hot spare)
Powersoft Digium 5000 amp (driving four labsubs)
Six PAS TOC RSLA's being driven by 2000 watt"ish" (peak) amps.
three phase power distro rack (3P 208v 30A {90A total power @120v})
AMP RACK 2:
<same as above except no dl32 or driverack>
I can only come to the conclusion here that the DL32 does not have as good of shielding as the S16 snake does. The fact that it was near four labsubs, a Digium powersoft amp, along with other magnetic inducing equipment is the reason why I was losing sync during the EDM events. Its when the system is working its hardest. The amps and driving tons of power, subs are producing massive amounts of magnetic output, TONS of SPL. Etc, etc. All of this sustained due to the type of music it is. Just... Really low, really long low end sound. Its the only thing I've got at the moment. And still, why this all worked with the s16 and x32 is confusing. (as mentioned above, I did at one point take the snake out of the rack but had the same issues)
All of that said, I am planing on moving the DL32 to the rack that houses all of the wireless mic receivers and an in-ear system stuff. This rack is usually further from the business end of the PA power system(s) and closer to the stage anyway, which is good. I am also purchasing a pair of KLARK TEKNIK DN9620 which will support fiber optic connectivity alongside a tactical/armored optical snake. (both AES50 "A" and "B" get sent down that pipe, W/ shielded cat5 as a backup at the push of a button, so bonus) After the misery last year I never want to deal with dropouts again. I'm not convinced that THIS issue was cat5 related, though, with all the other things I've been reading, seems like a good idea to just optically isolate the snake and board all together.
Aside from those upgrades, in hindsight, it does seem foolish the keep the DL32 in the back of the stage left amp rack, considering its a high speed network device, being subjected to all sorts of potential noise (both electrically and physically being beat up) but I'd still like to know if anyone has had issues quite like this? Any distance issues to consider? (how far should the DL32 be from sources of magnetic fields, etc)
-Any take would be greatly appreciated! I still can't decide if my new DL32 is defective (loose connections or IC pads actually vibrating around creating issues) or if this is just normal and par for the coarse with me making stupid life choices in regard to where I'm placing equipment. ;)
Thanks for the read and any advise anyone might have. I know this is somewhat of a wall-o-text post, but I wanted to be as concise as possible.
-Jason
-
My first thought is that you have a vibration issue; a physical problem not an electronic one.
Something in the chain doesn't like shaking at 50 hz.
-
My first thought is that you have a vibration issue; a physical problem not an electronic one.
Something in the chain doesn't like shaking at 50 hz.
Normally I'd completely agree however I could almost, with precision, estimate when a drop would occur depending on how much current was being drawn by the PA. (which I suppose could correlate to "shake" threshold) On the other hand the 80's sets and bands with WAY louder and punchier than the EDM sets were. -That's my only hesitation in this instance. I'm hoping to try and replicate the problem this summer/spring before any big events, I'm just having a location problem. -Never a convenient place to setup the entire rig and pound on it like that without some kind of complaint or issue ;)
If there is something going on with the DL32, I'm not sure how I'd even get them to warranty/test something like that. I feel really stuck at the moment.
-
Quick one to check: is the shield from the cat5 connected to the shells of both ethercons? What's the lenght of the cat5?
My DL32/16's have never skipped a beat. Last weekend one lived 4 days between racks of d&b D12, D20 and 1200's, and some dimmers, with EDM.
-
Normally I'd completely agree however I could almost, with precision, estimate when a drop would occur depending on how much current was being drawn by the PA. (which I suppose could correlate to "shake" threshold) On the other hand the 80's sets and bands with WAY louder and punchier than the EDM sets were. -That's my only hesitation in this instance. I'm hoping to try and replicate the problem this summer/spring before any big events, I'm just having a location problem. -Never a convenient place to setup the entire rig and pound on it like that without some kind of complaint or issue ;)
If there is something going on with the DL32, I'm not sure how I'd even get them to warranty/test something like that. I feel really stuck at the moment.
Is it possible that voltage sag was causing the issue?
Lee
-
Normally I'd completely agree however I could almost, with precision, estimate when a drop would occur depending on how much current was being drawn by the PA. (which I suppose could correlate to "shake" threshold) On the other hand the 80's sets and bands with WAY louder and punchier than the EDM sets were. -That's my only hesitation in this instance. I'm hoping to try and replicate the problem this summer/spring before any big events, I'm just having a location problem. -Never a convenient place to setup the entire rig and pound on it like that without some kind of complaint or issue ;)
If there is something going on with the DL32, I'm not sure how I'd even get them to warranty/test something like that. I feel really stuck at the moment.
The LABSub has an impedance minimum right around 40Hz, so the most current would be drawn on sustained notes at that frequency. EDM often has sustained continuous LF content, a surface resonance may be shaking the DL32 enough for the cord to loose continuity, while short "louder and punchier" peaks evidently aren't exciting the resonance enough to become problematic.
You could get an equivalent SPL of four with just one LAB Sub if you remove the grill and place the DL32 inside the mouth, do a slow sine sweep right around 40 Hz and see if the problem occurs, and determine at what specific frequency and SPL.
If it does occur, you should be able to eliminate the problem by applying pressure around the connector to dampen the resonance.
Good luck!
Art
-
Is it possible that voltage sag was causing the issue?
Lee
OP stated:
Troubleshooting steps taken
* measured power at amp rack to check for voltage dip. None. Rock Solid.
* measured (mains) power at supply (150A 3 phase 208) -all phases during event. -Perfect
* measured power going directly TO the DL32 -rock solid
So low voltage is not the problem.
-
OP stated:
Troubleshooting steps taken
* measured power at amp rack to check for voltage dip. None. Rock Solid.
* measured (mains) power at supply (150A 3 phase 208) -all phases during event. -Perfect
* measured power going directly TO the DL32 -rock solid
So low voltage is not the problem.
Oops, Thanks Art, I missed that.
That's what I get for reading and replying at work while I'm doing other things. :-)
Lee
-
Is it possible that voltage sag was causing the issue?
Lee
I never saw any voltage sag whatsoever. (never dipped more then two volts at any given time during the dropout)
-
Quick one to check: is the shield from the cat5 connected to the shells of both ethercons? What's the lenght of the cat5?
My DL32/16's have never skipped a beat. Last weekend one lived 4 days between racks of d&b D12, D20 and 1200's, and some dimmers, with EDM.
It was 150 feet including patch cables. Ground tested on every cable. (barrel to barrel)
-J
-
OP stated:
Troubleshooting steps taken
* measured power at amp rack to check for voltage dip. None. Rock Solid.
* measured (mains) power at supply (150A 3 phase 208) -all phases during event. -Perfect
* measured power going directly TO the DL32 -rock solid
So low voltage is not the problem.
Yeah, we put a logging meter on the rack to watch all sorts of stuff. Nothing out of the norm, unfortunately.
-
The LABSub has an impedance minimum right around 40Hz, so the most current would be drawn on sustained notes at that frequency. EDM often has sustained continuous LF content, a surface resonance may be shaking the DL32 enough for the cord to loose continuity, while short "louder and punchier" peaks evidently aren't exciting the resonance enough to become problematic.
You could get an equivalent SPL of four with just one LAB Sub if you remove the grill and place the DL32 inside the mouth, do a slow sine sweep right around 40 Hz and see if the problem occurs, and determine at what specific frequency and SPL.
If it does occur, you should be able to eliminate the problem by applying pressure around the connector to dampen the resonance.
Good luck!
Art
That's actually an interesting idea! I'll likely give that a try and see what happens. At least then I can either verify or rule out physical issues with connectors are surface mount / board issues! Thanks!
-
That's actually an interesting idea! I'll likely give that a try and see what happens. At least then I can either verify or rule out physical issues with connectors are surface mount / board issues! Thanks!
Here's my guess - it's a vibration-caused intermittent connection of an RJ45 (or one of the 8 conductors thereof) inside its EtherCon. You might check that all of the RJ45 have the same amount of recess inside the shells, too.
"tis a puzzlement"
-
If it's not a power issue, it just has to be a vibration issue. Is there anywhere that a cable is laying on a surface (the stage, a sub cab, a weak floor, draped in the air) where it could vibrate? The only time I have ever lost sync was when cable was exposed to people walking or dancing on it. Maybe the console or stage box is being vibrated enough to have a problem or a defect revealed.
-
Here's my guess - it's a vibration-caused intermittent connection of an RJ45 (or one of the 8 conductors thereof) inside its EtherCon. You might check that all of the RJ45 have the same amount of recess inside the shells, too.
"tis a puzzlement"
He replaced the cable, and tried port a and b...
150 feet is not too much.
I have never opened up my stageboxes. Maybe there is a loose connector somewhere?
-
Where are your amp racks located in relation to your subwoofers?
If it is too close, I can see it getting some pretty solid vibrations. Maybe try isolating the dl32 from the rack rails and set it on top of the rack on some egg-crate foam?
Maybe placing the rack in a null might help. A 40Hz-50Hz wavelength equals about 22ft-27ft, so maybe a 1/2 wavelength between the subs and a wall?
-
He replaced the cable, and tried port a and b...
150 feet is not too much.
I have never opened up my stageboxes. Maybe there is a loose connector somewhere?
The problem could exist with more than 1 connector or cable. What the connection life of an RJ45? IIRC it's around 2000 cycles but I could be wrong.
I don't remember the pinout of AES50 but I'm pretty sure that sync/clock is on its own pair.
-
A dl32 can't be clock master. The M32 can be slave though...
OP? Did you check that? But if clocking is the problem it should happen always.
Edit: sorry Tim, read that wrong. You are talking about the cable pair.
-
The problem could exist with more than 1 connector or cable. What the connection life of an RJ45? IIRC it's around 2000 cycles but I could be wrong.
I don't remember the pinout of AES50 but I'm pretty sure that sync/clock is on its own pair.
Cable I replaced it with was a band new ProCo cable I keep on hand for emergencies (spare/backup) I'm 99% sure it wasn't a cabling issue this time around. Two different cables were used for testing. (one that was only a half a year old, and one that was brand new) dropout behavior was the same with whatever cable I used.
-
A dl32 or dl16 can't be clock master. The M32 can be slave though...
OP? Did you check that? But if clocking is the problem it should happen always.
Edit: sorry Tim, read that wrong. You are talking about the cable pair.
Yeah, clock was good. System ran top notch all four days except for the evening raves.
-
Quick one to check: is the shield from the cat5 connected to the shells of both ethercons? What's the lenght of the cat5?
My DL32/16's have never skipped a beat. Last weekend one lived 4 days between racks of d&b D12, D20 and 1200's, and some dimmers, with EDM.
It was between racks the racks, but what else with in with the DL32? Mine are physically in my MA2 rack. I'm thinking that having the DL32 on top of/directly next to the Digium is a bad deal. Still seems strange to me though. I can't imagine I'm the first person to have a configuration like this.
-
I think the dl32 is faulty.
Was removing the dl32, and run only the m32 an option?
-
I think the dl32 is faulty.
Was removing the dl32, and run only the m32 an option?
Unfortunately not. Of all the backup stuff I truck in, an analog snake wasn't one of them. >_<
-
It doesn’t sound like it from what you are describing, but are you aware that there is a known problem with some DBX Venue 360 units. It has to do with the power supply connector to the main board. I have one here that I just did the fix on yesterday for someone. I wonder if it is on the same power drop as the DL32 if it could cause a problem with the DL32, if it is plugged in but not even being used. From what the problem is it shouldn’t be an issue, but when it comes to digital things I am not so quick to write off any thing even when it looks like it couldn’t be the cause.
-
What are you using to monitor voltage, and what is its response time?
You say that the console/snake head connection indicator went from green to red, so I agree that you are losing sync. I don't see how some kind of outboard gear is causing that.
My money is on either a loose RJ45 connectiion on console or snake, or that you were getting a momentary voltage drop which caused the console to reboot. Voltage drop below something like 85 volts for more than half an AC sine wave will cause a reboot. You didn't mention a UPS in your list of gear.
Wait, how long were the dropouts? Reboot would be a while, certainly longer than the duration of a kick hit (or whatever).
If it wasn't rebooting then something was loose, and since you replaced the snake it likely wouldn't be in the snake. That leaves console and stage box.
Did you have a spare stage box? Why not? Spare console?
That's the only thing I can think of: have spares if/when something goes bad. Good that you had a spare snake, but it didn't solve the problem. This gear is cheap and lightweight enough, in historical comparison, that it doesn't make sense to not carry spares for everything. A Producer or X32Rack in SKB or whatever cases hardly weigh anything, and a spare S16 or S32 don't take up much rack space. I'm not familiar with the DL series.
Otherwise, you did a good job of sussing it out, or trying to. Not trying to bust your chops, because you were thorough and prepared in your troubleshooting. But spares are good.
-
Hey Folks,
First time poster here.
Please go to your profile and change the "Name" field to your real first and last name as required by the posting rules displayed in the header at the top of the section, and in the Site Rules and Suggestions (http://forums.prosoundweb.com/index.php/board,36.0.html) in the Forum Announcements section, and on the registration page when you registered.
Mac
admin
-
What are you using to monitor voltage, and what is its response time?
You say that the console/snake head connection indicator went from green to red, so I agree that you are losing sync. I don't see how some kind of outboard gear is causing that.
My money is on either a loose RJ45 connectiion on console or snake, or that you were getting a momentary voltage drop which caused the console to reboot. Voltage drop below something like 85 volts for more than half an AC sine wave will cause a reboot. You didn't mention a UPS in your list of gear.
Wait, how long were the dropouts? Reboot would be a while, certainly longer than the duration of a kick hit (or whatever).
If it wasn't rebooting then something was loose, and since you replaced the snake it likely wouldn't be in the snake. That leaves console and stage box.
Did you have a spare stage box? Why not? Spare console?
That's the only thing I can think of: have spares if/when something goes bad. Good that you had a spare snake, but it didn't solve the problem. This gear is cheap and lightweight enough, in historical comparison, that it doesn't make sense to not carry spares for everything. A Producer or X32Rack in SKB or whatever cases hardly weigh anything, and a spare S16 or S32 don't take up much rack space. I'm not familiar with the DL series.
Otherwise, you did a good job of sussing it out, or trying to. Not trying to bust your chops, because you were thorough and prepared in your troubleshooting. But spares are good.
I was using both a Fluke 289 for testing (0.025 ms, 0.25 ms, 2.5 ms) and my system also has a full time E-gage in the power distribution system for monitoring power quality (voltage, current, entire rig power usage, etc) (https://www.egauge.net/) -If there was even a small blip on the mains side, I'd know for sure.
My reasoning for why I thought it would be electromagnetic was due to the fact that the DL32 was sitting right on top of the Digium (which is a very high power class D amplifier) The switching power supply in it can produce HUGE amounts of magnetic fields, and high frequency switching noise under load that I assume can make its way into a digital data stream. -Like I said though, I am drawing at straws at this point. I'm hoping to do some testing this weekend and will report back with my findings. I'll be checking both shake test, which was mentioned above, and I will also be looking at exactly how much RFI/magnetic radiation is on top of that digital amp. With the AES50 signal being a low voltage signal as it is, I don't think it would be out of the realm of impossibility that high current devices could interfere with the processing of those signal. especially if they use coil filters/etc. But, we'll see! I honestly have no idea what I'm dealing with yet! Hoping to also get that DL16 before this weekend for testing.
And yeah, I totally agree with having spares! My main issue was that I upgraded RIGHT before my event and admittedly, I wasn't prepared. I usually even carry an analog snake in the truck as well. Not this time. (oversight) I'm hoping to actually outfit the rig with at least a spare dl16 for exactly this reason, though. Having the DL32 box go out would be a show stopper for sure! (I have spares for every other critical system in place already)
Thanks for the time and advice! -J
-
Please go to your profile and change the "Name" field to your real first and last name as required by the posting rules displayed in the header at the top of the section, and in the Site Rules and Suggestions (http://forums.prosoundweb.com/index.php/board,36.0.html) in the Forum Announcements section, and on the registration page when you registered.
Mac
admin
Apologies for that, I did read and see that mention, however I guess it wasn't clear at the time that the registration username was also the Forum name -That said, hindsight is 20/20 as you guys even put in quote "This is the displayed name that people will see" and I didn't put it together. -Can't win`em all I guess ;)
Got it fixed up.
-
It was 150 feet including patch cables. Ground tested on every cable. (barrel to barrel)
-J
Just to be clear, and I may have missed this. You mention patch cables. Did you have the Cat5 patched directly between the M32 and DL32, or hooked into a rack panel, and then patched to the DL? I've been told that Midas advises a direct connection between devices.
-
You mentioned patch cables. Did you check the patching adapters?
Sent from my iPhone using Tapatalk
-
Just to be clear, and I may have missed this. You mention patch cables. Did you have the Cat5 patched directly between the M32 and DL32, or hooked into a rack panel, and then patched to the DL? I've been told that Midas advises a direct connection between devices.
Yeah, I wasn't very clear about that part. So I have a "master" snake that has two fiber (10gig Ethernet connections) SDI video cables, DMX cables, and four AES50 cables. The first night I was using the master snake. 2nd night I ran an independent direct connection between the M32 and the DL32. Behavior was the same when tested both ways. -good catch, I hadn't mentioned that!
-
You mentioned patch cables. Did you check the patching adapters?
Sent from my iPhone using Tapatalk
Just replied up above about a similar question, but in short, yes. console to DL32 was checked. All pins and ground. (on all AES50 cables) -Then on the 2nd day I used a independent cable that was connected directly between the M32 and DL32.
-
Yeah, I wasn't very clear about that part. So I have a "master" snake that has two fiber (10gig Ethernet connections) SDI video cables, DMX cables, and four AES50 cables. The first night I was using the master snake. 2nd night I ran an independent direct connection between the M32 and the DL32. Behavior was the same when tested both ways. -good catch, I hadn't mentioned that!
This snake sounds pretty awesome! If you feel like posting pictures, I'd like to see it.
One question: if the master snake has 4 AES50 cables already, why didn't you swap over to one of them? Or were they already in use? Why the need for the totally separate single run snake?
Oh, and is there an AC run in that master snake? Preemptory comment: Having an AC cable in a bundle with AES 50 cables will not have an effect on digital transmission, as long as all cables have twisted-pair construction. I'm just curious how you get your AC to FOH.
-
This snake sounds pretty awesome! If you feel like posting pictures, I'd like to see it.
One question: if the master snake has 4 AES50 cables already, why didn't you swap over to one of them? Or were they already in use? Why the need for the totally separate single run snake?
Oh, and is there an AC run in that master snake? Preemptory comment: Having an AC cable in a bundle with AES 50 cables will not have an effect on digital transmission, as long as all cables have twisted-pair construction. I'm just curious how you get your AC to FOH.
The "master" snake has been a work in progress for a bit. Having a custom cable made would of been insanely expensive so I end up building it myself. I've been adding lines to it as needed but Its finally hit its maximum. I don't think I could squeeze one more line into it. -I was looking though my phone and found a pic of the non-completed version of it that I sent as a text message to someone a few years back while building it. Not the best quality pic, and not its current state, but it'll get the idea across. Its since had its cover pate installed and as been finished with proper connectors, more lines, etc Its about 150ish feet on a rolling cable reel. (see pic below) Perfect for the indoor events I've been doing. Its also since been finished off with a hubble connector on the FOH side that has everything laid out nicely for patching. This snake doe NOT have any power running in it. Its completely low voltage. FOH gets its power from a five wire cable that has 30A of three phase on it. (enough for lighting world, video world, audio, etc) It works out to be about 30 amps per department, off the sub distro. Which is complete overkill generally but I'd rather have more than enough for breathing room. (the extra has come in handy for spotlights, long throw video projectors, and other random things)
As for the other AES50 lines, they were in use at the time for some other task so I was unable to leverage them for testing. When I ran into trouble, I wanted to run directly between the M32 & DL32 which I would not of been able to do going though the snake. When I ran the spare it had no patches or connectors in the middle so I was able to rule out a few things on that front. It seemed like a good place to start while in a panic during the event ;)
(http://webstopcentral.com/snakeV1.jpeg)
-
The "master" snake has been a work in progress for a bit. Having a custom cable made would of been insanely expensive so I end up building it myself. I've been adding lines to it as needed but Its finally hit its maximum. I don't think I could squeeze one more line into it. -I was looking though my phone and found a pic of the non-completed version of it that I sent as a text message to someone a few years back while building it. Not the best quality pic, and not its current state, but it'll get the idea across. Its since had its cover pate installed and as been finished with proper connectors, more lines, etc Its about 150ish feet on a rolling cable reel. (see pic below) Perfect for the indoor events I've been doing. Its also since been finished off with a hubble connector on the FOH side that has everything laid out nicely for patching. This snake doe NOT have any power running in it. Its completely low voltage. FOH gets its power from a five wire cable that has 30A of three phase on it. (enough for lighting world, video world, audio, etc) It works out to be about 30 amps per department, off the sub distro. Which is complete overkill generally but I'd rather have more than enough for breathing room. (the extra has come in handy for spotlights, long throw video projectors, and other random things)
As for the other AES50 lines, they were in use at the time for some other task so I was unable to leverage them for testing. When I ran into trouble, I wanted to run directly between the M32 & DL32 which I would not of been able to do going though the snake. When I ran the spare it had no patches or connectors in the middle so I was able to rule out a few things on that front. It seemed like a good place to start while in a panic during the event ;)
(http://webstopcentral.com/snakeV1.jpeg)
Thanks for posting the pic. It looks like a nice project.
Is there an Ethercon cable we're not seeing? I'm not recognizing any of those connectors as being an Ethercon, but I just might not be seeing it from the side like that.
If there's not one, it could have been that the DJ had some kind of static electricity discharge (ESD) that was causing your lack of sync periodically. Sparking to anything connected to your snake head (like a mic or DI) can result in loss of sync when you don't have Ethercons and shielded cable throughout. Even though Midas officially says you don't need STP, I've caused sync disruption in a Midas console (Pro 3) in tests using UTP.
Just one more possibility.
-
There are a few quirks with AES50 that not everyone realises because its not very obvious.
On AES50 the word clock travels on the extra 2 pairs of the RJ45 Cable. One pair will do send and the other will do receive. The data travels on the other two pairs inside a the network cable. Thats why AES50 needs all 8 pins on a network cable. However, which pair does receive and and which one does send is determined by the system as it starts up. It is a bit random. Should now one of the pairs be damaged the fault will only become obvious when the direction on clock and the damage coincide. ( Hope this makes sense so far ). Thats why often people will report that it was working fine one day and not fine the other, and then go back to be fine the next job. So that is your first thing you could check. A regular CAT5 tester will often not do, as it will only check the inner two pairs as that is all that is required by ethernet.
The next thing to really look out for is that when you use none Ethercon connectors it is possible to bend the outside pins of a socket on either the console or the stage box by inserting the RJ45 not 100% straight. Should that happen the socket needs to be replaced. And this will manifest itself with word clock issues like you describe.
-
A regular CAT5 tester will often not do, as it will only check the inner two pairs as that is all that is required by ethernet.
A slight correction: Gigabit Ethernet uses all four pairs in a Cat5e (or above) cable. 100MB Ethernet only uses two pairs, but it's the second inner pair and one of the side pairs (i.e. pins 1, 2, 3, and 6).
In my mind, the reason that most inexpensive Ethernet testers don't work well is because they typically only run a continuity test (i.e. testing whether the cable will pass D.C.), and there are lots of things you can do to a cable that will make it pass D.C. flawlessly but not pass a high frequency signal (like Ethernet or AES50) well (or at all).
-Russ
-
In my mind, the reason that most inexpensive Ethernet testers don't work well is because they typically only run a continuity test (i.e. testing whether the cable will pass D.C.), and there are lots of things you can do to a cable that will make it pass D.C. flawlessly but not pass a high frequency signal (like Ethernet or AES50) well (or at all).
-Russ
Thx. Do you have a recommendation for a good tester, a good value?
-
It is a wonder that they ever incorporated RJ45 connectors in pro audio. Durability is not their strong suit. Not very easy to repair or replace connections/connectors when compared to a XLR or phone plug. It is nice to be able to repair a cable in the field with basic tools and skills.
Troubleshooting a bad cable also seems to be problematic.
-
Do you have a recommendation for a good tester, a good value?
A while ago I picked up an older device called the "Real World Certifier" (which is really more of a verifier than a certifier), although I haven't had much chance to use it. The impression I got at the time (although this might have changed) is that anything more capable than this product had the word "Fluke" on the front and a 4-digit price-tag on the back.
That said, there are many people on these boards with more experience than me who can give you a better answer.
-Russ
-
The problem is those expensive testers are all Ethernet centric. That really is not what you want.
All you need is a basic pair scanner with a TDR for fault location.
Sent from my VS995 using Tapatalk
-
The problem is those expensive testers are all Ethernet centric. That really is not what you want.
All you need is a basic pair scanner with a TDR for fault location.
I was hoping you'd chime in! What do you recommend along these lines?
-Russ
-
I was hoping you'd chime in! What do you recommend along these lines?
-Russ
The Tripplet LVPRO20 has cable map, fault location and full PoE testing for $200. It also tests Coax !
-
The Tripplet LVPRO20 has cable map, fault location and full PoE testing for $200. It also tests Coax !
I have the Byte Brothers version. Originally bought the 20, upgraded it to 30 for the extra network stuff. Worth every penny! Was just on a gig where one of the flown DSMs did not work. Tests out as a short 20 ft from the end. Sure enough, right at 20 ft was a kink in the cable. It would also save me hours and miles of walking during large network deployments. The simple distance to fault and what pr its on is the best trouble shooting feature I have found in the field.
-
I thought I'd do a quick update here since I have figured this out and have tested and verified the issue over the course of the summer. I can comfortably say without a DOUBT That the DL32 can NOT be subjected to high magnetic fields without having sync issues. My suggestion for anyone who has a huge amount of power amplification is to definitely keep their AES50 snakes away from such devices and MOST CERTAINLY do not put them in are around such racks. I have tested this with two separate DL32's and two different m32 boards. Same results. Once I put my production DL32 into the wireless mic rack all issues went away without any further sync issues.
I had brought a second mixer and DL32 on a few events and played around with it trying to get it to fail in a similar manor. The wheeled extra DL32 I brought had constant sync issues when wheeled near or around the mains. Problems went away at around three feet from the SIDE of the labs or 1 foot from the amp racks at program level. (these amp racks are VERY densely packed)
Anyway, there it is. I'm hoping in the future I can get solid Gauss measurements and come up with an actual threshold but for now I'll just chalk it up to "Keep your sensitive data devices away from crazy high magnetic fields" (Duh!) ::)
A Huge Thanks to all who brainstormed on this tricky one, lots of good insight!
-Jason
-
I thought I'd do a quick update here since I have figured this out and have tested and verified the issue over the course of the summer. I can comfortably say without a DOUBT That the DL32 can NOT be subjected to high magnetic fields without having sync issues. My suggestion for anyone who has a huge amount of power amplification is to definitely keep their AES50 snakes away from such devices and MOST CERTAINLY do not put them in are around such racks. I have tested this with two separate DL32's and two different m32 boards. Same results. Once I put my production DL32 into the wireless mic rack all issues went away without any further sync issues.
I had brought a second mixer and DL32 on a few events and played around with it trying to get it to fail in a similar manor. The wheeled extra DL32 I brought had constant sync issues when wheeled near or around the mains. Problems went away at around three feet from the SIDE of the labs or 1 foot from the amp racks at program level. (these amp racks are VERY densely packed)
Anyway, there it is. I'm hoping in the future I can get solid Gauss measurements and come up with an actual threshold but for now I'll just chalk it up to "Keep your sensitive data devices away from crazy high magnetic fields" (Duh!) ::)
A Huge Thanks to all who brainstormed on this tricky one, lots of good insight!
-Jason
Nice job, and thank you! You've really advanced the state of our knowledge of how this new stuff works.
After reading through the thread again, I'd like to ask you to revisit the difference in performance between the Midas gear and the Behringer X32/S16 combination, which you said performed flawlessly compared to the Midas.
In your testing and verification, did you confirm again that the S16 etc. had no problem?
Did you try it with the M32 or only all Midas vs. all Behringer?
Also, you say to keep both the DL32 AND the snake away from high magnetic field devices, but you only describe problems when the DL32 itself is in the magnetic fields. What did you experience when the device was away from the magnetic field but the cable was in it? Are the fields too great for the common mode rejection from twisting and balancing transformers (assuming they're there like they are in Ethernet) to handle?
Thanks again!
-
The Tripplet LVPRO20 has cable map, fault location and full PoE testing for $200. It also tests Coax !
TDR is key to successful tests of twisted pair cable. Cables that appear good using continuity test devices are often not capable of much more than finding breaks in the cable or poorly attached connectors. A TDR based test set (Time Domain Reflectometer) will pinpoint anomalies at very specific distances.
Substituting the cable with a like cable from the same manufacturer is not always a good test. Quality is the same and results will be similar. Use higher quality cable from American manufacturers (Belden).
Bandwidth may be diminished at lower frequencies.
Put your stage boxes in their own case away from magnetic fields, possibly with a foam cushion under them if you feel vibration is the issue here.
Try running the cable directly from the stage box to the interface, bypassing the spool.
Good luck.
-
Thanks Dan! Yeah, its been an interesting ride on this one! It tested my sanity for sure!
In your testing and verification, did you confirm again that the S16 etc. had no problem?
Did you try it with the M32 or only all Midas vs. all Behringer?
Interestingly, I never did have an issue with the S16. Unfortunately I didn't have access to any Behringer gear at the time of testing, but I would certainly love to revisit if the opportunity arises. I was sitting here just now trying to rack my brain on the differences between the setups at the time and I still have no idea what they would of been. Now that I have a good "Control" baseline I could easily test Behringer stuff and see if it has similer issues. I'd also love to also try a DL16 and see if it has the same vulnerability. It could come down to design of the 16 vs 32.
Also, you say to keep both the DL32 AND the snake away from high magnetic field devices, but you only describe problems when the DL32 itself is in the magnetic fields. What did you experience when the device was away from the magnetic field but the cable was in it? Are the fields too great for the common mode rejection from twisting and balancing transformers (assuming they're there like they are in Ethernet) to handle?
So on that one I should extrapolate a little bit. -admittedly I made the post really quick between doing few other things.
I completely misspoke when I said that. At the time I was thinking of the stage box when I said "snake" -Not the actual cable, so I didn't communicate that properly. That said, I do spool my snake and when there's extra I leave it on the spool. Since that last incident I have been keeping it clear of the racks and stacks just as a good practice measure. I didn't test against that however. -I'm also a little suspect of the panel mount Ethercon connector that was also mounted in the amp rack since their traces arn't shielded on the PCB or otherwise twisted, but that might just be me nitpicking at this point.
I'm hoping as time goes on we'll all learn more and more of the limitations of these systems and with enough collective experience we'll nail it out of the park with a list of general "Do's and Don'ts" OR maybe better put: a list of "Best Practice" Each event is always a little different but I think if we're all aware of the possibilities we can best mitigate a show from hell ;)
-J
-
Interestingly, I never did have an issue with the S16. Unfortunately I didn't have access to any Behringer gear at the time of testing, but I would certainly love to revisit if the opportunity arises. I was sitting here just now trying to rack my brain on the differences between the setups at the time and I still have no idea what they would of been. Now that I have a good "Control" baseline I could easily test Behringer stuff and see if it has similer issues. I'd also love to also try a DL16 and see if it has the same vulnerability. It could come down to design of the 16 vs 32.
If you're good at sussing out board-level components, take a look in the product reviews section, someone did a teardown of the S16 and DL16 and found quite a few differences in the components used in each. It seemed that most of those differences were in the preamp boards, though, and the analog and digital I/O boards seemed almost identical (at least to my eye). If I were a betting man, I'd say that the 32-channel versions would have the same issue unless the additional preamp boards provided some kind of unintentional insulation for EMF.
-
Now that I have a good "Control" baseline I could easily test Behringer stuff and see if it has similer issues. I'd also love to also try a DL16 and see if it has the same vulnerability. It could come down to design of the 16 vs 32.
If you want, you could PM me your location. I'm a Behringer dealer and am in pretty good with the local Rep and I bet he knows Reps around the country, so we should be able to get you some Behringer gear to test. If you want.
Based only on the heft of them, I believe the S16 and 32 have steel chassis. I can't find anything about the construction of the DL series; do they have any of the cool looking and expensive anodized and brushed aluminum forming one or more sides of the chassis? I would think they'd be the same as the S series with that ugly old painted steel, but thanks to this thread I'm now aware that the B and M boxes are not identical construction and so chassis construction cannot be taken for granted, I guess.
-I'm also a little suspect of the panel mount Ethercon connector that was also mounted in the amp rack since their traces arn't shielded on the PCB or otherwise twisted, but that might just be me nitpicking at this point.
This could have significance, depending on what you mean by that "Ethercon mounted in the amp rack".
Do you mean a connector that you put there to avoid climbing in the back of the rack each time you connect/disconnect the snake?
Or do you mean the Ethercons on the DL boxes which are its input and jumper?