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 => Installed Sound/Contracting => Topic started by: Chris Clark on January 18, 2014, 06:47:58 PM
-
I am working on a plan to link a new area into a distributed system (between multiple buildings), but at this time I do not have the ability to bury a dedicated audio line or the budget to purchase a fiber interface (which is how we are getting to other areas currently, borrowing fiber lines from the IT department). I can secure a dry-pair from the telecom department that can be patched from source to destination, which by mine and the telecom supervisor's estimations will be between 1350 to 1500 feet (1350 ft is based on flat-ground point-to-point measurement, not taking into account elevation changes or curvature of the buried lines around obstacles, hence the 1500 "high" estimate).
I know it is theoretically possible to do this, as my college's radio station maintained a few dry-pair audio links, one as a backup between the studio and the transmitter in case the microwave STL went down, and a set from the campus hockey rink press box for broadcasting games. What I don't remember is whether they had any interfacing equipment or whether they patched the balanced audio directly into and out of the lines.
The equipment to be installed will be balanced on both sides so I'm not worried about baluns, but I'm wondering if a line driver (or a booster?) or isolation of some kind is still necessary, especially at this distance? Also would the lightning protectors (as much as they can actually provide...) at each building entrance be adequate do you think? The lines are underground except where in buildings or pedestals so I would presume that would create some inherent protection. The 25/50/100/what-have-you pair shield is also grounded at each point of emergence from the ground (building entrances, pedestals). I know lightning is unpredictable enough it would take out everything anyway if it wanted to, but anything that would help as a just-in-case...
Any other thoughts on this idea? I hope it to only last a year or so as I plan on putting in some major upgrades next year, budget permitting (either more fiber convertors or more hopefully conversion to a networked system based on, say, Cobranet), but they want the sound there this year.
PS, sound quality of the resultant background music/paging doesn't need to be of the utmost quality, the speakers in the area are of the "mushroom" landscaping type and therefore lack high-end anyway. In fact the end system has been in place for a while but the goal is to tie it in so its audio source comes from the main rack and can therefore be controlled centrally and include paging.
-
If you don't mind an experiment, try a cat5 audio balun, one with screw terminals since you will only have one pair. It may not work, but it just might...
-
That brings up memories of 111 (one-elevens) repeater transformers, from 40 years ago.
-
If you don't mind an experiment, try a cat5 audio balun, one with screw terminals since you will only have one pair. It may not work, but it just might...
No balun required. A twisted pair is all that's needed for balanced audio, either analogue or digital. That said, it would be best to isolate the buildings by using a transformer at one end of each line for an analogue run.
Lee
-
Let 'er rip! I work in remote broadcast audio, and we used to run audio over telco wire at many outdoor events. The Detroit Grand Prix comes to mind, when they used to run through downtown (late 80s?). We used 25 pair telco cable in 500' lengths with amphenol connectors. This was all temporary, and just laid on the ground just outside the race fence. The race circuit was a 2 mile loop, and the TV compound was outside the circuit at one end. The runs around either way came to well over 5000' feet each. A mic with battery powered line amp was placed where needed, and tied into the 25 pair cables, one pair per mic. We would have 20-40 efx mics around any given race track, and maybe some RTS intercom in the wire as well. The other end of the cables would breakout and patch into the console or intercom as needed. We used to wire golf courses like this too until fiber came along...
-
Let 'er rip!
Yup. Assuming the twisted pair is in good condition, it should work fine. We also use these in remote broadcast in older football stadiums to get from the truck to the pressbox. Both RTS and line level audio.
We had something like this:
http://m.markertek.com/product/detail/http%3A%7C%7Cwww.markertek.com%7CMobile%7CConnectors-Adapters%7CAudio-Adapters%7CAudio-Binding-Posts%7CSescom%7CSES-MKP-23.xhtml
- Tony Williams
-
Awesome news! I won't even need the XLR to wire conversion honestly, as the equipment at both ends will have euroblocks on it anyway. I think my budget can support an isolation transformer as Lee suggested (just in case). I was just hoping I didn't have to find any special line amplifiers or something, it looks like I don't!
Lee or others, any preference on which side the isolation transformer goes on? I'm personally thinking closer to the source side since the router (a Digitool MX32) is a tad bit more expensive than the amplifier on the other side and in my brain it seems this would provide another step of electrical protection from the outside world, is this a correct thought process or am I just spinning wheels and it really doesn't make a lick of difference?
Thanks guys!
-
Awesome news! I won't even need the XLR to wire conversion honestly, as the equipment at both ends will have euroblocks on it anyway. I think my budget can support an isolation transformer as Lee suggested (just in case). I was just hoping I didn't have to find any special line amplifiers or something, it looks like I don't!
Lee or others, any preference on which side the isolation transformer goes on? I'm personally thinking closer to the source side since the router (a Digitool MX32) is a tad bit more expensive than the amplifier on the other side and in my brain it seems this would provide another step of electrical protection from the outside world, is this a correct thought process or am I just spinning wheels and it really doesn't make a lick of difference?
Thanks guys!
Here's a link to the Jensen application notes page.
http://www.jensen-transformers.com/apps_sc.html
Depending on the specifics you may want to use something like a DIN mounted solution for either input or output or you may want to build something that has some transient suppression capability (if the lines are run outdoors between buildings for instance) or you may just want a preterminated isolator in a box. This gives you many options. Do take note that the impedance wants to be low on the driver side and high on the receiver side.
There are many solutions outside of Jensen as well.
Lee
-
Commonly available passive Di boxes will work well for this application.
Have the low Z "facing" the copper line and the high Z to the equipment.
-
Commonly available passive Di boxes will work well for this application.
Have the low Z "facing" the copper line and the high Z to the equipment.
Maybe, maybe not. Try it. See if you can test a model that you plan to use before paying for it and finding a problem.
The potential problem comes from the input impedance of the isolation DI at the receiving end. You don't want to drive an impedance that is lower than or even equal to your driven impedance or you may well have current limiting issues that create clipping at the driven end.
Lee
-
Commonly available passive Di boxes will work well for this application.
Have the low Z "facing" the copper line and the high Z to the equipment.
In addition to the problem Lee mentioned, many passive DIs reduce the level on the balanced side to match XLR mic inputs on consoles. You want to maintain the high level signal through the unshielded twisted pair. If the source equipment is a good balanced line driver there is no need for a transformer on the output, but the destination end should have a good input transformer on it. This will maximize the CMRR and provide electrical isolation. Transformers by their nature are perfectly balanced, while a non-transformer balanced input may not be, reducing the CMRR.
There are specific transformers designed to be input transformers, and ones designed as output transformers. For the best results they are not interchangeable.
For a one time gig you will probably be fine with any good quality 1:1 isolation transformer, if this is a circumstance that will happen often it may be worth buying the right transformer.
Mac
-
The outputs of the MX32 are listed as servo-balanced. (Is this the same thing as "electronic balancing"?) The existing amp (currently running off a local CD player, which is what we are looking to eliminate) is an older ComTech (honestly I don't remember if it is a x00 or x10 series, and won't have a chance to look for a few weeks but I don't think it makes a huge difference) using the basic terminal block PIP which if memory serves correctly is simply electronically balanced.
So if I were to get an Input transformer it would go closer to the amplifier, and if it was an output transformer it would go closer to the digitool, am I understanding that correctly?
I'd have to double check inventories, but I may have a Biamp CMA kicking around, these have a switch on input 1 for "600 ohm telephone" - would this serve the purpose I'm looking for with transformer isolation? If not I'll keep researching for isolation transformers.
-
Just to add to what others have said.
The freq response of a typical telephone connection is limited to 300-3Khz.
This is NOT because of the wire-but because of the equipment on either end. This was done to keep the bandwidth within the parameters of the mic and ear pieces and works fine.
So as long as you dealing with just wire and no telco gear you will be fine.
Use line level-NOT normal instrument level- Di boxes. and I would make sure they are 1:1 so you get the maximum voltage transfer. Of course most line level boxes are 1:1-since they are used for isolation and not so concerned with impedance loading.
-
The outputs of the MX32 are listed as servo-balanced. (Is this the same thing as "electronic balancing"?) The existing amp (currently running off a local CD player, which is what we are looking to eliminate) is an older ComTech (honestly I don't remember if it is a x00 or x10 series, and won't have a chance to look for a few weeks but I don't think it makes a huge difference) using the basic terminal block PIP which if memory serves correctly is simply electronically balanced.
Servo balanced is a method of keeping levels the same when the negative side is grounded (when using a 2 conductor plug).
A transformer at each end will give you your best chance for hum free and RFI free operation.
So if I were to get an Input transformer it would go closer to the amplifier, and if it was an output transformer it would go closer to the digitool, am I understanding that correctly?
This is nit-picking. you will have a difficult time sourcing anything marketed as an "input transformer" for this application.
I'd have to double check inventories, but I may have a Biamp CMA kicking around, these have a switch on input 1 for "600 ohm telephone" - would this serve the purpose I'm looking for with transformer isolation? If not I'll keep researching for isolation transformers.
When using 600 ohm systems terminating it with a 600 ohm load helps with impedance matching. see: http://en.wikipedia.org/wiki/Impedance_matching. In this case you are bridging a very low Z output... probably 50 or 100 ohms (whatever series resistor is used for protection in the unit)
Personally I'd probably use Barix ehernet solution http://www.barix.com/ or Studiohub http://www.studiohub.com/matchjackDAs.php
Contact me if you want WE11C repeat coils. I have some I can spare.
-
It keeps nagging at me every time I come across this thread, so I gotta ask; I keep thinking in my mind that this is older wire, although the OP says nothing about it. If this were old 100 pr cat 3 or less, with a lot less of a twist to each pair compared to new cat 5/6, wouldn't that make a huge difference in isolation and cross talk via induction given that this would be a shared with an existing operating telco line?
-
A transformer at each end will give you your best chance for hum free and RFI free operation.
The CMRR noise rejection happens at the input. Servo balanced, impedance balanced or transformer balanced at the output will all have the same CMRR with a transformer input
This is nit-picking. you will have a difficult time sourcing anything marketed as an "input transformer" for this application.
HERE (http://www.jensen-transformers.com/datashts/pi2xx.pdf), HERE (http://www.jensen-transformers.com/ln_in.html), HERE (http://www.lundahl.se/our-products/line-input/), and HERE (http://www.sowter.co.uk/acatalog/SOWTER_TRANSFORMERS_MIC__LINE_INPUT_3.html) for a start.
When using 600 ohm systems terminating it with a 600 ohm load helps with impedance matching. see: http://en.wikipedia.org/wiki/Impedance_matching. In this case you are bridging a very low Z output... probably 50 or 100 ohms (whatever series resistor is used for protection in the unit)
Most modern audio gear is low output impedance and relatively high input impedance (bridging), there is no real need to match impedance.
Mac
-
Another question in the problem is the wiring direct building to building, or to the telco exchange?
We had to special request to telco to make pairs building to building in the Past.
If it goes through an exchange would the frequency response be limited to Phone requirements,
say 300 to 12K ???
-
No we are not requesting a dry-pair from the local exchange. The infrastructure this would be on is all internal, a pair cross-connected from the source to destination and not patched into the PBX and, as far as the local telco is concerned, basically non-existant. The concern I had was what equipment would be necessary to reliably drive the ~1500ft twisted pair (yes, it is probably CAT3).
Because I'm rather limited on where I can order from (setting up a purchasing account sometimes takes an act of congress around here...) one of the companies we have an account with has a Rane LT22 (http://www.rane.com/lt22.html) available, seems to be one of the few isolation units I can find through any of them. Would a pair of these (one at each end) be sufficient, do you think?
-
Another question in the problem is the wiring direct building to building, or to the telco exchange?
We had to special request to telco to make pairs building to building in the Past.
If it goes through an exchange would the frequency response be limited to Phone requirements,
say 300 to 12K ???
Back when I was doing telephone work-the response was limited to 300-3K. Maybe it has changed since then. That was 30+ years ago.
-
. The concern I had was what equipment would be necessary to reliably drive the ~1500ft twisted pair (yes, it is probably CAT3).
1500 ft is nothing. Until you get past 4000 feet, I would not worry about it.
You will not have enough loss to worry about-especially at the impedances of modern gear.
Speaker lines would be a different issue.
-
Because I'm rather limited on where I can order from (setting up a purchasing account sometimes takes an act of congress around here...) one of the companies we have an account with has a Rane LT22 (http://www.rane.com/lt22.html) available, seems to be one of the few isolation units I can find through any of them. Would a pair of these (one at each end) be sufficient, do you think?
You really only need one of them at the destination end. The router is providing a balanced output. You want a transformer at the input of the destination to maximize CMRR (Common Mode Rejection Ratio) by providing a perfectly balanced input. The transformer will also provide electrical isolation. Be aware that the maximum level for that transformer is +4, while the Jensen I linked to is greater than +20. If the transformer saturates from too much low frequency level there will be increased distortion and less high frequency response.
HERE (http://www.jensen-transformers.com/an/Audio%20Transformers%20Chapter.pdf) is a chapter from Glen Ballou's book relating to transformers, written by Bill Whitlock.
Mac
-
Thank you for the help, all! I will keep looking at what other transformers might be available from our vendors, but the maximum level for the Comnet fiber transmitters we use on the other legs of this system is +6 so it shouldn't be a stretch to keep them all below +4. The playback systems in place from the central rack are also already high-passed at 100Hz as well so that the 70V transformers throughout the system are at less risk of saturating too, so I think that will help as well.
Thanks again!
-
Thank you for the help, all! I will keep looking at what other transformers might be available from our vendors, but the maximum level for the Comnet fiber transmitters we use on the other legs of this system is +6 so it shouldn't be a stretch to keep them all below +4. The playback systems in place from the central rack are also already high-passed at 100Hz as well so that the 70V transformers throughout the system are at less risk of saturating too, so I think that will help as well.
Thanks again!
Be sure to understand what the "max" ratings actually mean.
Typically if you read a level on a VU meter (say +4), that means that the peaks are way above that-say 20dB.
But if the max input is +6 on a digital system, then that would be the max-so the "average" that would be read on a typical VU meter may be 20dB or so BELOW that.
I don't know the particulars in this case-just pointing out some things to be aware of that could possibly cause distortion.
-
I have seen this done at football stadiums, using a passive balancer to send analog audio around the stadium (via underground conduit, manhole-to-manhole) to the visitor's side amp rack, about 1400-1600 feet.
The recurring problem, however, was lightning-induced spikes getting into the long copper line and knocking out the DSP on either end. Same thing happened with a power sequencing control line (sends a contact closure); lightning surge killed sequencing relays. They had me replace the audio line with fiber optic after twice replacing DSPs.
-
I agree, I worry about lightning induced issues, even though the vast majority of the infrastructure is underground. I can tell you that, as I mentioned, every point where the trunk lines emerge from the ground has the cable shield tied to ground, as well as gas-tube protection on the pairs where they enter buildings. In the run we'd be using, we're looking at protectors at the exit point of the building housing the source audio and the entrance point of the building housing the amplifier, as well as at least 6 points where the cable shield is tied to ground (based on my knowledge of where the cross-connect points would be for this circuit).
The building housing the source audio is also a building where one of our three PBX hosts are so I would think there's a fairly heavy lightning protection presence at the patch point into the main trunk line. I'm severely hoping this is enough protection for the digitool since to my knowledge we haven't had any losses to the PBX equipment due to lightning and we tend to take a lot of strikes.
Also like I said I'm hoping this will only be in place for a year or two before I have the budget to switch over to fiber (or better yet, I'd love to convert the whole thing to, say, Cobranet, but that's not happening for the foreseeable future).
Ahh, the wonderful dB/dBU/dBm measurements... One of those things I truthfully have never fully comprehended. I understand the logarithmic scale, the fact that dB is inherently a ratio, 3dB is essentially 2x or 1/2x the power, etc... But comparing the different forms (dB, dBu, dBm, dBx :P ) usually gets the better of me.
The Comnet Fiber tx/rx pairs already in use on the other legs are listed as "4.4V Peak to peak (+6dBm)" (I understand the 4Vpp measurement in itself... in that if ran a pure sine wave to it and measured with a DMM I could crank the voltages as high as 4Vpp, or 2V Peak, or roughly 1.4V RMS, etc..), and the Rane LT22 I was planning to get lists the 40Hz max level as +24dBu (also states "Limit 0.5dB"), and the Digitool lists the max output also as +24dBu (adjustable from -24 to +24). I hate to continue burdening you guys, but any insight as to what it all means and how it will interact with each other?
-
The building housing the source audio is also a building where one of our three PBX hosts are so I would think there's a fairly heavy lightning protection presence at the patch point into the main trunk line.
Typically the protector clamp voltage for telco lines is in the range of 240-600 volts. Certainly much too high for an audio pair. The thing to do is to add additional protection to that pair at each end that clamps at maybe 25 volts. They should be located at the building entrances and the grounding is important. Hopefully the ground for the existing protectors was done properly back to the service entrance electrical ground.
-Hal
-
I have seen this done at football stadiums, using a passive balancer to send analog audio around the stadium (via underground conduit, manhole-to-manhole) to the visitor's side amp rack, about 1400-1600 feet.
The recurring problem, however, was lightning-induced spikes getting into the long copper line and knocking out the DSP on either end. Same thing happened with a power sequencing control line (sends a contact closure); lightning surge killed sequencing relays. They had me replace the audio line with fiber optic after twice replacing DSPs.
Refer back to my original link
http://www.jensen-transformers.com/apps_sc.html
and look at the isolation application for \long runs with lightning protection.
Here's a direct link to the wiring for lightning protection.
http://www.jensen-transformers.com/as/as028.pdf
Lee
-
Chris,
You have received lots of great guidance on the technical aspects of dry pairs. I can only supply a story.
Some years ago (in the '90s) I hung out at an FM station that normally got their signal from the studio to the approximately 2 mile away mountain-top transmitter by an FM radio link (950 MHz, I think). They leased a pair of dry pairs from the local phone company as a backup, but preferred the radio link, which worked most of the time, for its better sound quality.
At the time (maybe still) the whole area suffered from a pair shortage. I had a phone tech confess to me that there were actually fewer pairs than subscribers and that at any given time someone was doing without. Sort of time-division multiplexing paced by service ticket response time.
Anyway, at the station it became clear that if there was no signal on a pair then sooner or later some phone guy hunting around with his buttset would appropriate it for another customer. Running a good loud test tone on the pairs at all times greatly improved the situation. Maybe not a bad thing to keep in mind if you are paying for pairs that sit idle some of the time. Perhaps a recorded message along the lines of "Don't f with this pair. It belongs to...".
--Frank
-
Running a good loud test tone on the pairs at all times greatly improved the situation. Maybe not a bad thing to keep in mind if you are paying for pairs that sit idle some of the time. Perhaps a recorded message along the lines of "Don't f with this pair. It belongs to...".
Yes, tone/program is your friend!
Running tone or program down the line at all times will let the tech who is hunting know that this pair IS being used and to not mess with it.
If it is silent, they may just up and use it for something else.
-
Ha, that's a good idea, although I doubt it is necessary here, nor would I want to blast test-tone in that area all night. As I said our lines are all internal with the exception of the incoming pairs on a separate cable by our telco for DID (direct inward dial) into the PBX. I've already verified that our telecom supervisor has patched and marked out the the pair we'll be using on both the 66 blocks and the wirelists, and during operating hours there will be constant background music on them. (If it seems odd that I know all this it is because I actually went to school for telecom engineering, but chose sound as my career lol)