The impressive thing is that the audio travelling through 9600 meters of cable has the evquivalent delay of around one centimeter through air! Assuming your FOH is at the end of the line, you‘d have the signal 28 seconds earlier on the console than your hear it. Not the best for tapping delays I guess…
@@DaveRat That's a whole lot of time to bleep out some expletives... or even just... you guys totally suck, I'ma just cue up an mp3 (or even go find a cd) here, eat an ice cream, and play that... before anyone else hears how bad you (they) are live :)
What about if you mic'ed and realtime broadcasted the center of a thunderstorm or a war. Seems nearby people six miles away may hear the broadcast and then the delayed acoustic sound
That is a really cool demo. I've never used a mic cable more than about 150 feet in total (snake + mic cable proper). Back in the mid-70's when I was being taught about pro audio, I was told that normal balanced low-z mic lines could be up to 1,000 feet with no ill effects, and if you wanted to go further, there were a lot of mics that could be switched to a really low source impedance (50 ohms, usually) and these could drive a mile of cable. Looks like the engineers who designed the system way back in the day knew their way around a slide rule!
Back in ‘94 I helped install 8 or 10 12 pair audio mults for an NFL playoff game telecast, the runs were about 1,000 feet from tv trucks to the stage inside the stadium. We also had 10 coax video feeds plus 8 triax camera cables. Today we would run a single tac12 bundle and have spare fibers. The audio guys never whined that I recall. They had to run mics from the stage back, RTS intercom & IFB’s back to the stage from the trucks.
I think this shows that for low and mid fidelity runs, long cables work for signal level. For high fidelity, I have done tests that show there are audible errticats in the HF at 300 meters or so begining to occur with audio snake. I have not tested 300 meter Cat5e with analog for high fidelity
And not to this extent but amlnalog backup lines to delay clusters and a large scale festival as well as a bunch of speakers for events like an air show or speakers along a marathon route
Good demo! Back in the old days Ma Bell sent voice hundreds of miles over small gauge unshielded twisted pair. You could get special equalized dedicated lines for audio that would meet specs to FM radio.
Hiya Dave, love these interesting experiments you are doing! Years ago, I was involved with an outdoor recording experiment that involved some condenser mics placed about 600-700 meters away from the preamp. It was in a semi-rural environment, with a broadcast TV and radio antenna located about 4 km away. We had plastic conduit buried in the ground, and the first attempt at this involved using some Belden foil shielded install grade mic cable. The result was huge amounts of 60Hz noise on top of the mic signal. We tired a few different approaches, using preamps at the mic end and sending the signal much hotter(lower Z) back to the recording location, but that didn't improve the situation much. What did however work, was that we changed the mic cable from the foil shielded cable, to a star quad braided shield cable. Once we did that, the problem completely disappeared. We had four distinct mic lines, all doing the same thing that the original foil shielded cable. In hindsight, seeing you do this with much longer runs, had me thinking what really was the source of the problem we had. The only difference I see is that you have these cables coiled up, in a very small area, inside a building space. Our situation was outdoors with a straight 600 meter run. I suppose it's possible the Belden cable wasn't as twisted as it should have been, or how your Sound Tools cable is twisted. Seems to me, that as long as you have a very consistent twist and a decent preamp with good CMRR, the cable doesn't even need a shield. After all, the telephone company ran much longer twisted pair lines to carry very noiseless audio, much longer distances, without any shielding and worked this way for almost 100 years. That's the one thing we didn't try... using telephone cable. I suspect that you will be able to run your mic cable setup to a usable distance that will be dependent on the actual impedance of the copper connection. The more distance you add, the cable impedance will start to overtake the output impedance of the mic, and you will start to have a level drop as you already measured, like inserting a pad, but you still should be able to recover it with gain on your preamp, just a little more added ambient noise will also be there. I wouldn't hesitate to guess you could probably get 25 miles of cable, with still useable audio that would only need a little gain and EQ to fix. Thanks again for the experiments, they're awesome!
Thank you. As far as the 60hz, that was not from the radio antenna, more likely there was a wall current flowing through the shield somehow. As foil or braid over twisted wire should be pretty good at filtering out such a low frequency.
@@DaveRat That was my first thought when reading this, assuming sonicfrog is in the USA - here in the UK we'd expect a 50Hz hum. Over those distances there's a very good chance that currents circulating in the ground from the electricity network could cause a significant difference in potential between the mic and mixer. If any of the XLR shells are grounded or the mic chassis were grounded (I think this was more common in days gone by?) and in contact with general earth then it's easy to see how there could be a ground loop. Although it's weird that it went away with another (better quality) shielded cable.
Hi guys, thanks for the replies. It was in US(60Hz) and early 2000s time frame. We went through all kinds of things trying to figure out what was the source of the noise. We had condenser mics that needed phantom, and we tried even a small battery supply at the mics, coupled with some different line amplifiers, and using transformers with that original foil shielded install mic cable. There were lots of combinations of things tried to come to some solution. All of the attempts passed the audio, and allowed it to work, but there was such a strong 60Hz component there, it was unusable. Ground loop was the obvious first thought, but believe me that got rooted out on day one with the loads of experiments with power supplies, line amps and transformers. Our final thought was that the mic cable might just not be as twisted and uniform between the signal pairs as one would think. For say 50 meters maybe it was fine. But for the distance we had, it didn't work until we changed out the cable. So, the noise was actually in the signal lines, just not equally and hence, showed up as a nice 60Hz, from whatever was in the area inducing it. We had ourselves a 700m long guitar pickup with that original cable. :)
@@sonicfrog1 dare I ask what classified or unclassified project this was on, I mean for? I mean why would you did you need to go that far? I assume it was some sort of sensor monitoring type scenario. I recently went on a hike to an old Canal tunnel and thought it would be neat to put some microphones at both ends and make some noises and then realized I would need a third of a mile of cable which I don't have (yet). I still wanna try it although it'll probably be done with Wireless which is another distance question, how does it propagate underground how does it propagate in a tunnel....
This channel, this video, this guy. Just gold. Great content, very interesting to see what happens at such high distances. I also loved your comparison x32 vs. M32 a lot ! Very in depth but never boring. Thanks for sharing your knowledge with us Dave!
This is such great timing for me! I just ranover 500ft 8 channel snake cable, 350ft of speaker cable and 2000ft of cat5e cable for my studio. If 8 channel snake cable was a bit easier in the wallet, I would have ran a bit more. I never considered how well mic level could travel down cat5e; shocking really. Although, it being coiled as opposed to actually run at length may make a difference, but I ain't running 6 miles! Thanks Dave, always incredible information you share. Cheers!
Digital boards and audio over cat5 are the biggest game changers in the live audio field in my lifetime. Good timing too cause now I'm mid 50's and can't lug all the gear lol. Still have a 220 foot 32x8 splitter snake. It's heavier than I am! I don't think I've run my cat5 once without smiling thinking about how that beast of a snake is asleep in my basement hahaha. Amazing to see phantom work 6 miles away! I didn't expect that to work out. Great demo Dave. 👍
I've done a couple of things with 16 channel snakes and thought about other ways of doing it. When I built my own system out I decided on prioritizing portability and weight. I really do like the analog over CAT5 nature for simplicity and I pick myself up 4x 100ft foot cables which will probably be good for most of what I will ever do. Although I do have a idea for a multi CAT5e snake in my head using CAT5e 25 pair cable using AMP ends into a 6 Jack breakout at each end. That way the large ends can be removed for storage aiding the process of rolling the long 100-150ft cable up and shorter ( say 25' ) lengths can be substituted for smaller venues. I figure out of the 6 channels 3 channels would give me 12 channels of send and 1.5 channels would give me 6 channels of return, leaving a gigabit link open between stage and house and 2 pairs for maybe DMX. But it's just a crazy idea floating in my head, I wish somebody made a break out already, there is one from +studiohub that's more of a panel mount open design and I forget where it was they had a design lacking shielded connectors so no phantom power without modifications.
We (The rental side of Rat Sound) we're using multi Cat5e cables in an overall jacket but stopped doing so. A failure of on line kills the whole cable. We now user multiple Cat5e bundled and swap out lines if they get damaged. That way we have multiple lines than can be dicital or analog bundled with fiber as well
@@DaveRat I was wondering how the rental handle Booth -T- Stage, or if the cat system was just being used as breakouts on stage. Interesting so it was killing the whole cable more so than acting like a dead channel, hmmm. Yeah some way to bundle multiple individual line seems to be the way to go our big stuff like that. I was perusing the offerings and there aren't a lot of options for bundled CAT premade off the shelf. All I saw was 2 or 4 and assume anything past was custom.
It was not killing the whole car but carrying a 4xcat bundled cable with 1 dead cat line is not an option for use. 1 dead line means the cable is of no use to us
Given the amount of inductance in those reels RF problem will be reduced some role off I can see . Nice demo The Schoeps shoot gun mikes where sensitive to coiling 300ft cap vs not coiled . This was at a telarc recording the cable does make a big effect on sound . This cable looks very fine .
With tightly twisted pairs like what Cat5e has, the coiling inductance of large diameter couls should be minimal. I think the dominant issue is capacitance and as the length increases, the capacitance is rolling off the HF and we see the phase shift on the smaart screen.
@@DaveRat The increased Capacitance will be harder to drive current and thus signal as it increases in a transmission line . Causing the impedance to change over it length since even though the cat5e is balanced the mic and micpre are not matched . What you see is the phase shift caused by the delay line you showed . I think it was a very good demo to show that in any real application Cat5e is a workable choose . Analog is effected in ways that simple transmission theory misses . This was a fine demo of what Heaviside work was in the 1890's . You are a fine teacher of some very advanced theory .
Well. The pink was not line level but it was a low impedance output. So what you see on smaart with pink is what line level will do and what you hear is what mics will do
Hmm, well the electro magnetic waves travel about 60% the speed of light or 300,000 meters per sec. So 9600 meters divided by 300,000 equals . 032 or approximately 3 ms wire latency. That sounds like a fun test.
Cable rolls are collecting not only shunting capacitance C (depends on cable's quality and length - reducing upper frequencies), and serial resistance R (depends on cable's quality and length - reducing signal level), but serial inductance L too (depends on cable ROLLS summary serial inductance). This inductance is affecting passive dynamic mics (just voice coils, or voice coils + transformers like SM58); transformer output active electronics dynamic and condenser mics; playing games with shunting cable's capacitance. And the next input's impedance parameters are taking part in this game too, all it is rather complex...
@@DaveRat Thank you very much, your videos are so simple and so educating! Forgot to add - all the losses depend on the microphone's (or any other transducer's or gear's) output impedance. Lower impedance reduces affects depth. 8 ohms, or 50 ohms, or 600 ohms, or 10 kohms, etc., output impedances (passive or active) mean a serious difference in the signal loss in cables + other affects, including EMI/RFI noise and some phase shifting coming from LC components. But in the real life, we rarely have so long cables, so no reason for panic... :)
definitely helps you visualize/appreciate what considerations those of us that design sports venues have when they jam the tv /radio folks up in the rafters of a 1.2 million square foot building.
Interesting, I would have expected a significant drop in phantom power voltage. Thanks for testing! Would have been interesting to measure how much of the 48V is still getting through ;)
There will be a significant drop in voltage over even the 'shorter' 1.5 mile lengths, but it's dc and and most mics really don't need 48V except for some super high end stuff, like perhaps the Nuemann, but all that mean is its gonna lose some dynamic range and noise floor, hardly significant compared to the frequency rolloff's and cable induced noise at the lengths being talked about. Take away is... if you can put up with the sound degradation the phantom power can still fuel your condensers (unless you started with some crap deck that only sends out 12V and pretends that is good enough).
Even for a short cable the voltage can drop significantly from 48 Volts with a mic connected. That's because the phantom power standard is actually 48V fed onto pins 2 and 3 of the XLR through 6.8K resistors. So the resistance of the 6 miles of cable has a resistance far smaller than the resistors in the preamp feeding the power into the cable in the first place. It would be about 820 ohms added to the existing 6800, assuming 24 gauge copper mic cable. So the amount the voltage drops below 48V when the mic is connected would only be about 12% more with the 6 miles of cable than with a short cable. In reality this drops would be a bit more because I havent calculated the additional drop from the shield resistance, but its likely to be small compared to the twisted pairs. With one the Cat5e conductors serving as the GND wire rather than a proper shield, the drop would be more significant maybe closer to 35% more than without a long cable.
Dave, it would be very interesting to see this test with a digital console sending a high amount of tracks down the different distances to see if there's any digital degradation once you start adding tracks. If I had that much cable I would try it myself...
Might want to talk with old radio engineers that used telephone dry lines that are cat3 for miles for broadcast feeds or studio to transmitter links. Yes, they had to have a device at the end to compensate for EQ.
Very cool. Yeah I was getting questions asking how far they can run analog down the Cat5e cable I designed. Mainly looking at the 500 to 1000 meter lengths. But one customer wanted longer as well. Well this was to show that down a mile or so without any additional stuff, it will work fairly well just plug and play
My understanding is usually that couplers is where you usually lose your most power (especially with stranded, so check your 5e. Solid is more annoying to terminate, but would be better for insertion loss), so I'd be interested to see if you could push it further with just a longer cable instead of multiple connected spools.
I think that couplers do have an impact on data frequencies of 100k, 1 meg or higher, but this is analog over cat5. So the relative simple 20 to 20k is a lot more resilient and not impacted much.
For analog over Cat5/6 do you need to use cable with individually shielded pairs or is an overall foil/braid shield good enough? I'm putting together a fixed installation wtih about 150 ft of Category cable runs from one end of a room to another. It may will run either in a soffit past lights or on the ground next to electric heaters.
Hey Dave! Awesome demo! I love this type of direct comparison. Two SMALL points: 1) The base-line Frequency response of your mixer seems a bit outta wack (i.e. larger deviations from 0 dB Transfer function than one would expect), and I think the changes in the response would be CLEARER (and easier to quantify) with a smaller vertical scale on the transfer function. It looks like you have ~+24dB to -15dB, or an ~ 39dB range. This range could be reduced to +/- 15 dB making the changes easier to see (A vertical scale of +10dB to -20dB or -25dB might be best to show the roll-off at the longer lengths) THANK YOU for making these videos.... informative and FUN!!!! - Sam.
Phantom power is very similar the voltage present on a copper telephone line. The same thing happens to telco lines the farther you go out the more attenuation in the high end but in the telco world we employ what is called a passive load coil that helps to make up for the loss. Since you are using what is essentially telephone wire to run this test I'm not sure if the cables coiled up are effectively acting like a load coil and affecting your results positively or not. You could try to unroll them and repeat the test and place one roll up cable reel in the middle switching it in and out to verify weather or not it effects your high end.
Hmmm, comparing the impact of coiling of high bandwidth precision twisted pair Cat5e to telephone line may not be fully on point Are you sure the twisted pairs of Cat5e would be impacted by coiling the same as telephone line? Anyway, the test was to see how far audio and phantom would go down Cat5e and it well exceeded expectations. So you think that if I uncoil it will go even farther than what is already too far for nearly all applications?
@@DaveRat You are correct they are not exactly the same telco twisted copper is is generally Cat3. I was just curious if there was any effect having them all coiled up.
My understanding is that perfectly twisted pairs should be immune to coiling inductance. That said, nothing is perfect. And I am not curious enough to uncoil 6 miles of cable and find out. At least not yet
@@DaveRat Yes I agree don't do that, make an intern or someone do it. LOL. just kidding. Love the channel by the way keep it up. en.wikipedia.org/wiki/Loading_coil
I was being cocky about Cat-snakes (worried about audio quality using let's say 10m stage Cat-snakes, not 3 miles😁). Worries; XLR cabling (designed for audio fidelity for analog transfer), Ethernet cabling (designed for data integrity for digital transfer). Different cable specs for different applications. With Cat-snakes we are going to use cabling designed for transmission of digital signals for transmission of analogue signals. Without researching it, my technical brain said that it will negatively impact audio quality. You proved me wrong by showing it in practice! Brilliant! Thanks. I don't have to worry about using 10m Cat-snakes👍🏻🤟🏻😃
Good stuff and to be honest I knew it would work well but I did not realize that it would work that well better than mic cable and AES cable. I did other videos testing it as well as a 10 mi test
Great video! I wonder how much of that roll off is due to the wires being coiled. They are acting like air core inductors. What would the difference be if the wire was straight vs coiled?
Dave you make very cool videos and I have learned a lot from you. I have a question concerning this video. Did you consider the fact that coiled cable reels may create millihenry choke coils which will inherently create a low pass filter in your experiment?
I've been in large recording studio complexes with several studios where the audio from say, studio c's live room would get patched to studio a's control room, and the total signal path could be hundreds or even thousands of feet.
The idea for this video came from a customer buying Cat5e cable to put speakers up for a 1/2 marathon route 8n an area without cell service or WiFi The question was "how far can I run analog over Cat5e? So I tested it and made a vid. The 1/2 marathon had the drive unit in the middle. So 6 miles each way would do it and with shorter runs because there was a generator and the ability to refresh the signal at each speaker area evenly spaced on the route.
Wonder what the voltage drop is. I know most condenser mics will run in 12v even when the spec is +48v, but I have a couple that are more sensitive to voltage. Also be aware, that much wire will delay the signal a bit. Old osciliscopes used to use long coils of wire exaclty for that. Not enough to hear, but phase issues can arise. Have you done this with a dynamic lol? Now i just need to wire my house in cat5 and add preamps to every room. :D
Hmmm, math calcs. As far as delay, since sound travels about 200,000,000 meters per second in wire, the delay would be around .05 ms or about the equivalent of 1/2". In other words, the delay in the wire is less than moving the mic an inch further from lips As far as the voltage drop, that is an easy calc as well, Look up the resistance per 1000ft of 26ga wire, divide by 2, as there is 2 wires, then double it as it travels there and back. Then look up the current draw of the mic. Then apply ohms law. I believe I did the calc in another comment here on the vid got don't recall the result, I think it was less than a 12 volt drop. But the cald will get ya close
I'd have a question kinda related: could you use analog over cat5 for four (4) different headphone cue feeds in studio situation? i.e. four different unbalanced stereo-signals powered by headphone amps from the send-side. Or is unbalanced analog generally a total no-go with these cat5 based systems? To be clear just regular indoor cable lengths, not massive distances like on the video.
That works, we have several customers doing that and happy with the setup. You don't really want headphone caples too too long. Sending 4nstereo line level signals to headphone amps or sending stereo headphones directly is all good.
I work at a high school and use Wall-Cats all over the place. The district is much better at installing Cat cable rather than multiple mic lines. My question is can you run DMX and audio over the same Cat line?
That is one hell of a lot of inductor coils! I wish you would had measured the total resistance of 6 miles of wire. And also the inductance. Curious to know that.
Very interesting demo would it be possible for you to make a vid on how foh desks are switched in and out at festivals for all the guest engineers that would be a great help for me thank you I love your videos ❤️
@@DaveRat yep :) its just ideal to show people that are not from the field of pro audio :) that expensive cables 1000$ and up (for one meter of cable having a line signal.....) are completely useless.. we already knew... but yeah
Especially the power cords after 50 miles of ac lines on cables and 75 feet or 12Gauge solid conductor wire in the house walls, some kbuclehead buys a $5000 IEC cable thinking it will make a difference. Makes me smile.
Great/interesting demo. The HF roll-off of the pink noise generator over extremely long cable lengths (causing increased inductance, resistance, capacitance) might be attributable to generator's output impedance. However, if the output of the pink noise generator is coupled directly from its opamps, and does not have any series resistance or capacitance, the long cable length roll-off might be negligible. If its output is transformer-coupled, then that would also introduce losses. (All of this applies to microphone impedances as well.) Thank you for such a great demo!!!
Yes! And this is why your test is so cool, for it deals with "real world" values, typically 150 ohms. In theory, as the source (mic, pink noise, etc.) impedance goes down, the effects of cable losses (resistance, capacitance) would go down as well. Soooo, this test reveals that the resistance of the cables affects volume, while the capacitance of the cables affects high frequencies. If the source impedances were reduced from, say, 150 ohms down to an ideal zero ohms, these cable lengths would have no effect at all. Thanks again, Dave, for such a great test! Your efforts are much appreciated!
@@DaveRat Dave, like your other viewers, I really appreciate this valuable testing you're doing...AND THEN...generously sharing it with all of us on RU-vid. I'm very grateful. Sure, theory is nice, but to see this stuff in action is so cool. As a studio tech, I've always been fascinated with the reasons "why" things sound the way they do. For instance, what makes a 414 and a U47 sound different, or an SSL and a Neve sound different, or a Marshall and a Fender sound different, etc. It reminds me of music. Yeah, it's great to understand the theory behind different musicians' styles but, hey, nothing compares with actually hearing it, right? (BTW, you also get into the "why" of things when you ask and answer important questions like "Why crowded shows sound dull...", etc., and that's really cool too.) One really important point you made during your testing is, despite any losses created with 6 miles of cable, you can then just reach up and tweak a tone control and get it all back again! Wow! Thanks again for everything you're doing.
I have no idea what any of this means except the fact that you can get a fairly clear signal that far without boosters or repeaters whatever they are called. Also, what is phantom power and what is it doing in this situation.
Would I be correct in saying that these effects will differ depending on the output impedance of the source device? If so, would adding resistors at the output of the console cause it to replicate the performance of a dynamic mic that would have a (relatively) higher output impedance, for the purpose of the analyser measurement?
Yes, impedance will 8mpact it and the analyzer is reading pretty much what a mixer would do sending signal and would not be accurate for the mic. Probably the best way to truly test the mic would be to mic a pink noise source, then compensate it to flat with a short wire in the measurement software. Then change cable lengths.
Did you measure the voltage drop, I believe the phantom mics are specified 18-48V. Video idea, at what voltage will phantom mics not work. Nice video, thanks!
Super interesting video! I've hooked up a measurement mic with 100m of cable before and compared it to the direct response, with negligible error. Cool to see that even 10x that distance doesn't have a huge effect! It might be worth noting that the error also depends on the input device's input impedance. Some ADCs have really low ii (
Is it the input impedance or the output 8mpedance of the mic that controls the overall line impedance? My understanding is that lower mic output impedance allows longer cable runs with less hf roll off
@@DaveRat I consulted with the guys on my Audio Discord and worked out a long answer: There are three main electrical properies of the cable, which cause signal degredation. These are the cable's resistance, capacitance and inductance. These three components form a passive circuit along with the two devices (mic and preamp), similar to a crossover circuit inside a passive speaker. tl;dr: High preamp input impedance reduces overall signal loss Low mic output impedance reduces HF roll-off Thicker cable=less overall loss, but more HF loss, relatively speaking Long version: Let's talk about resistance first, since that's easy: First of all, we can say that at analog audio frequencies, cable resistance is the same for all frequencies (no need for silver plated audiophile snake oil). The mic's impedance is small compared to the preamp's, so it has almost no effect here (similar to damping factor in amps). The cable resistance causes a flat loss at all frequencies. Looking at the lowest frequencies in your 6 mile test, we see a loss of about 7dB, which plugged into a calculator looks like the wire is 27AWG. The Xenyx's input impedance is 2.6kΩ. The X32's input impedance for example is 5kΩ, so you'd see a 4dB reduction instead, compared to the 7dB on the Xenyx. Time to talk about cable capacitance: This is what causes the HF roll-off. It's basically a parallel capacitor, which is a 1st order LPF. Preamp input impedance has close to no effect on this, it is indeed the source device's output impedance, like you said. The Xenyx's is 240Ω, which I what we see in SMAART, I suppose? You can generalize, that a device with half the output impedance will roughtly have half the HF loss. SM58 is 150Ω, which would be better than the Xenyx already. KM184 is 50Ω, which would reduce that HF loss to only about a fifth, for example. Now comes the wierd part, we don't really have to worry about, inductance: This depends on cable layout. We know having a power cable on a drum/coiled up heats it up quickly because it's a)bunched up and can't transfer its heat to its surroundings well and b)a coil is an inductor, which forms a magnetic field. Since you're sending a differential (symmetric) signal through twisted pairs of cable, the two magnetic fields cancel each other out and not much is happening. If that signal was an unbalanced signal however, you'd have a series inductor in the loop, which forms a 2nd order LPF with the cable capacitance, making HF loss even worse. By how much I don't know, someone would have to test that. Note that this is mostly an issue when the cable is still on the drums and you (probably?) won't have to worry about it when it's laid out in a straight line. Of course, sending an unbalanced signal through miles of cables will cause a lot of other issues, but that's what we have DI boxes for. And for the cable properties: A thicker cable will have a lower resistance, but higher capacitance. This means lower loss overall, but more HF-loss.
With regards to speed of signal transmission: the speed of Electromagnetic signals running through a copper cable is approx. 200,000 km / second. Very fast indeed! Meaning that you would need to have 200,000 km of cable in order to introduce a delay of one (1) second. That’s a lot of cable. A typical single pair mike cable weighs 65g/meter. 200,000 km of cable would weigh 13,000 tons (13,000,000 kg). As said above: that’s a lot of cable!
Wouldn’t the cable drums act as induction coils? Would we see different results if the cable run was reeled out? Reason I ask is that cat5 sendin AES50 usually suffers voltage drops around 100ft and a repeater psu is required to keep them from dropping in and out.
Hmmm, well there's lots of variables that could be explored but the goal of the test was to see how far I could run phantom and a mic and what it did to pink noise. People are asking how far they can run signal down our super cat cable and sound tools products, so I did a test
@@DaveRat ah fair play buddy. I enjoyed the video and was overly surprised that a dynamic mic like a 58 could still be sent over 6miles of cable. Just that question i asked loomed in my mind so thought I'd ask :). Also loved the figure 8 speaker cable test you done. Will definitely be doing that from now on
Cool cool and it will have a minimal effect on speaker cable but for AC power feeder with separate hot, nuetral and ground, it will have the most effect of reducing coiling inductance
Hi Dave. Two questions. 1: How on earth did you come up with this challenge? 2: You have now concluded that it is possible to run an analog microphone signal through several miles of cable with only a minor impact to the quality of the signal. But do you know why that is?
Well, before there was digital and fiber, analog was run everywhere. I believe analog was designed to run long distances. Older designed mics like the RE20 have selectable impedances going down to 50 ohms That should allow it to run 2x to 3x farther with the same HF rolloff vs the 150 or 200 it is set to when sold.
Though the series resistance will be 500 ohms, phantom current is quite low, maybe 20mA at the highest, so voltage drop should be 10v max. Many, but not all, 48v microphones would be ok with this.
Cat5/6 has an inherent characteristic impedance of 110ohms - so it will handle 50 to 200 fairly well. You'll get some worse frequency rolloff on a 600ohm mic... and trying this with a guitar pickup... well, just lets not :)
Next try different consoles? See if different consoles can push 48v that far. Would also be interesting to measure heat on that pre circuit. Would also be interesting to see how much resistance was in that lenght of cable. As always ty Dave!
hello dave would you be kind enough to make a video with a dsp for 2 subwoofers 15 or 18 inches with frequency cuts and equalization I myself have a dsp xilica 206 but my md218 master audio boxes do not sound like I want or even a default preset that will allow you to start off on the right foot thanking you in advance musically
I wonder what the voltage drop is from 48V at that distance. Also, does the coils of wire indroduce an inductance that would chance the input impediance ( 250ohms ) being seen at the board ?
I wasn't about to uncoil all those wires but I figured that the coils if they had any effect at all would make things worse than straight wire, so this seemed to be a good test of worst case scenario
@@DaveRat Thank you for conducting these tests in your videos! It sure puts a guys mind to ease about having to have everything short as possible in fear of signal loss ect. 🙂
Pretty sure capacitance is the primary issue. Inductance is pretty well cancelled out by twisted pairs. The capacitance though is cumulative and will show as a phase shift in HF as we see. Basically the wires create a capacitor that is shorting the HF out.
Sound tools mic swapper. It allows me to swap in the long versus short cable. soundtools.com/microphone-accessories-page-swapper.html ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-zaZ7lRwV9Do.html
What can be interresting, when you plug mic to 1input, and try to draw gignal from remaining chanel lines in Cat cable. How strong crosstalk analog induction in cat cables is, when working balanced/unbalanced.
Hey dave , another classic vid , obviously u wud,nt run those cables spooled but but for instance if they were layed out how do they deal w/ heat dispersion, a mile though, absolutely amazing ,...
Great experiment! The roll-off is imputed only by the length of the cable or because is in the shape of an inductor? I think the pops you are hearing are mostly caused by the inductor shape of th drums
I think it is the capacitance the causes the pops and rolloff from the long wires in close proximity. I think the twists cancel out the inductance but further testing would be needed to verify
Yeah, would be interested to see someone do that. For me, I was mainly interested in Will it work and how far before issues are audible. Real world stuff. I will disecting the exact why to someone else
Also in the real world we rarely put analog down 6 miles Cat5e. I do like the practical perspective and this test was to answer Elbow far can I run analog down your super cat and super cat sound cable. So I set up 1.5 miles on reels, as a worst case. It worked better than expected so I looped it back to get 3 and 6 miles. Thought is would be an easy, fun and interesting.thing to do and a chance to fly the drone around
Interesting! I use a program called ShareX to record screens to video, it will also record the laptop sound making it easy to align the clips in the editor. This allows me to avoid pointing a camera at the screen .
Please test and compare it to 100 x 100m mic leads all connected sequentially. And also 10,000m of Monster cable. And 10,000m of unrolled cable. LOL! What if it was unbalanced - does it pickup AM radio? Can you use an EQ to restore and flatten it out? The 48v Phantom should go for several km because that's what old analog PSTN phones ran from sent from the telephone exchange.
My prediction before watching. 48V will work fine, the resistance of the cable won't be that significant in comparison to the 6.8K resistors the phantom is already being supplied via, though it will add some capacitance. The 48V should drop when it sees the uncharged capacitance, then come back up again as it charges, and work fine from then on out.
A ton of cheap audio gear will put out more like 12-24v on phantom. It's amazing the number of preamps that tolerate it. I do have a few that want the full +48v however. Test first!
Twisted pairs should not be impacted by coiling but... If the pairs are not perfectly twisted, the coils could create inductance which should rolloff the HF. So... It is possible that if they we uncoiled we could run the cables even longer!! But probably not enough to make a difference we could hear. I have done measurements on coiled versus not coiled Cat5e cables and the impact is very minor at high frequencies like a megahertz. The impact it 20K should be non-existent
@@DaveRat as i posted realised that those are probably shielded as well which should mitigate that. Any way, glad have found your channel, helped a lot with insight in to live sound and solutions to problems that occur.
I would say that if it has any effect at all, it would make it worse, not better. Also running signal both directions down the cable could make it worse as well. So figure a straight run will be the same or slightly better but doubtful
I don't think so I think it is due to capacitance. Because tightly twisted pairs cancel out electromagnetic fields and inductance needs an electromagnetic field to induct
True but for Cat5e it should not be a major issue. Though if we want max fidelity over 6 miles of Cat5e, having less coiled would probably be better than the extra water coil lengths
Well we’ll well! Hahaha. This is even more Brilliant!!! Wonder what a cloudlifter or a line driver on one end “pushing” the audio rather than pulling the audio. Just thinking if you would actually need this for simple monitoring this is great. Not for recording for any Grammy awards but hey. I bet someone would buy it. LOL.
Thanks for this. No idea why but i just had this weird assumption that too long of a cable would cause a crazy delay. You've obviously debunked that for me so thank you!
Well, the theory says that tightly twisted pairs cancel out electromagnetic fields. And inductance in coiled cables is dependant on there being an electromagnetic field. If it did not require uncoiling 6 miles of cable I would test to see if there is an effect and how much longer the cables could be with less impact
I am actually really impressed by the fact that anything even remotly related to the original signal would come out of a cablerun like that! However i see a slight potential "issue" with the testing of the phantom power as i asume either the XLR splitter, the XLR switcher or maybe both have the grounds of all connectors bridged. What that would mean is that the sheild of the long cable would be in parallel to the sheild of the short cable run, almost eliminating the impact of it's electrical resistance. In terms of signal strength there shouldn't be much of a difference with a balanced signal, however i do think that the impact on phantom power and maybe also sensitivity to EMI could be affected by this test setup. Either way, this was a great video and an impressive test regardless.
Yes, on one hand the ground length is 25% of the miles but also the ground braid is much thicker than the twisted pairs and has a lower resistance. So phantom travels at least 6 miles down twisted pairs of the 12 total it would travel, but the 6 miles of ground that it is not travelling is low resistance So maybe the real world equiv of phantom down 4 miles or so
He llo Sono!! Interesting. Yes etherc9m couplers do cause 8ssues for cat5s running data as the frequency is very high. For analog over cat5e. Those ethercon couplers actually have much thicker wire than the Cat5e cable so they should have minimal impact. I believe the main causes of the drops are resistance from the long thin wire and capacitance from two wires that are very long being next to each other 8n the twisted pair
Any connection will slightly degrade the signal, high speed data is especially impacted like 100k, 1 meg or above. Analog audio is like 20k so it's a very low frequency and lower frequencies are less impacted. There are 2 versions of neuteik barrel as well. One is way better for data than the other. For analog it makes no real difference which one.
I wonder what would happen if that cable wasn't coiled in a small space, but in a straight line round trip, I think there would probably be a lot more interference
Hmmm, the customer that bought our Soundtools Supercat cable to run signal to speakers along a sporting event route said it still works great with no issues
Hello Dave I totally understand however rather to do anything I think it would be interesting to nevertheless have a credits in order to start from a base and it is up to me to consider your professional advice
There is some really early & interesting science on pumping _analog_ audio long distances. The focus was on extending telephone lines _before_ the invention of the triode vacuum tube. Three really inventive folks came up with versions of the "loading coil". ¶ If curious about this, please see Wikipedia for...: #1 Oliver Heaviside #2 George Campbell #3 Michael Pupin === AT&T bought Pupin's patent to stay out of legal battle, making him a very wealthy man. A physics professor at Columbia, the Department's building is named after him. A fascinating person, he was an inventor in several area and Edwin Armstrong was his student. _..._ /Mike
Great stuff and my goal was to see how the gear we have readily available to us responds to using it for long runs. Taking all the theory and ideas based on gear we don't have and seeing how the gear we do have works!
A fellow senior grumpy engineer friend of mine who's too pretentious to bother commenting here himself because of fear of "noob relation" mentioned: "Interesting test with the cable runs. I'm surprised that no one mentioned that the cables are in coils which will produce impedance even though the cables are shielded. The common practice with extra lengths when running cables, is to lay it out in a figure 8 to avoid the problem."
Hmmm, I think there is some misunderstanding of shielded vs twisted pair. Twisted pair is inductance cancelling, Shielded is not. Either way, if indeed the coils were to create issues, then we could go even longer, but I ran out of cables to use and the whole purpose of the test was to see how far phantom would power a mic. Is your friend saying that uncoiling would achieve my goal of phantom not powering the mic?
@@DaveRat he made no mention of phantom power being lost over that distance. I think he was just grumpy to see the coils and overlooked the mention of it being twisted pairs. Cheers for your quick reply Dave! Long live coffee cat :3
Ha and awesome! Yeah, people like to spot things and spit out info as if they know what they are talking about. Best thing to do is ask him for ducumentaion and a video of the testing he did to gain that knowledge.
@@DaveRat I think your education is more valid as you are far more relevant compared to an old grump who hasn't had a gig in two decades (nor embraced the digital era for that matter), I doubt running audio over twisted pairs is a concept he's even familiar with. If we're to do a similar installation but not use your specific branded cable, as long as we're using Cat 5e twisted, we should be fine for most installations (say for tie-lines in a studio where the runs would be no more than a few hundred meters?). I'm trying to help another peer of mine design and build a new studio space but classic analog cabling (mogami or the sort) gets pricey. I suggested some cat boxes and running the audio over twisted pair instead. Also does Sound Tools provide any solutions for the wall box that is more than 8 channels? Could we present a custom design and have that built out? say for 24 ch or more? Thanks for sharing your knowledge with us freely and publicly :-)
A coaxial cable would theoretically be the perfect fit for this. That white stuff around the wire is there to make sure there's always a constant charge that stays the same whether it's one meter long or one kilometer
No it would not... The miracle of cat5/6 is the incredible well balanced (and different) twist density on each pair. If you tried to do this on unbalanced co-ax you would hear nothing but two cities worth of hum before the mic signal hit your preamps.
Coax does deal with high frequencies well as it has less capacitance but but the twisted pairs are better at rejecting noise. But the test was to see how far phantom power would work and with coax, that would be easy, 0 feet.
Cool and also maybe check out Soundtools.com catbox, cat tails,, wallcat, and cat rack for aluminum ones that won't rust or have the paint chip. Plus Soundtools has much nicer tails and versions no one else makes like the supercat tails