I agree since I am now combing the comments to determine precisely which tester he is using to test the cables. Dave could have made it easier by including the name. Better yet, he could have included an affiliate link to the exact tester and made some money while educating.
@@jatlastit’s a Fluke networks DSX CableAnalyzer, copper cable certifier. Part of their Versiv family of products. Model number is DSX2-5000, cost around $13,000. The DSX2-8000 = $15,000.
"Stars on RJ-45" I'll see myself out. Thanks for the spectest! Invaluable stuff, few of us ever get around to taking a hard look, too busy making do I guess.
Your shirt reminded me of my first real job in 1973, I was an auto parts delivery guy. I'd make the deliveries and at the last stop, I had to call the warehouse to let them know I was done and on my way back, in case I needed to do a pickup on my way in. I'd call and say, "This is Dave, I'm on my way back." And, of course I'd get the "Dave's not here" response every time.
another great video at the end of the 70s I was testing telephone cables we had a lorry full of equipment I think those two little boxes on your bench do the same job if the cables were loaded impedance matched there would be some improvement I remember your test with a kilometre long microphone cable 😊 There's a formula line resistance line inductance continuity Shunting The line capacitance shunting the line and the only practical thing you can do is increase the Capacitance Another thing with connections Is when the material changes the different energy levels in different conductors create something called the sea back Affect You could also put analog repeaters in the line
Shorter runs will have less issues. Yes, you can add switches or boosters to many protocols but they add latency and complexity and another thing to go wrong. Especially for festivals where they would need to be somewhere in the dirt between stage and delay clusters or FOH. Using high quality wide bandwidth cables and avoiding couplers with wide bandwidth signals is good Or switch to fiber. But fiber adds cost and is more fragile
Im wondering if the 10x10 meters has any influence... What would happen if it were to be 10 cables of different lengths? what the harmonic bumps be more spread out and less pronounced?
Interesting. Definitely can measure and see where couplers are and cable lengths in measurements. Some tests seem to indicate that having very short jumper cables right near the ends is especially undesirable and creates reflections and resonances where the signal level is highest.
Very interesting as usual. I have a small selection of lengths and a 5e coupler to give me more distance options (currently being used on the LAN until I make some cables up). Only running short distances compared to you guys, but might switch up to the 6.
Thanks for this. Question: is there any chance that using a Cat-5e coupler with the SuperCat 6 cable creates a mismatch? I know jack squat about these cables and how they affect signal loss/stability, but for me, as a touring FOH A1, I will always fight for a single cable from mixer to stage box without connectors. It’s amazing how many house techs fail to see the importance of reducing possible points of failure.
I didn't do it in the video but I did test the Cat5e couplers with Cat5e cable and the cat 6a couplers as well. The cat 6A couplers outperform the Cat5e couplers in every scenario. But since Cat5e is only rated up to 100 megahertz, both couplers perform good enough in that range. The sensitivity to couplers really happens at frequencies well above 100 megahertz where the cat 5e 100 megahertz coupler performs poorly and the cat6 a coupler before much better
@@Brody-Dolyniuk It's not linked to cable lenght. It is the data transmission speed that the CAT5e cable was designed for (100MHz and around 1Gbps of network traffic). CAT6a is designed for faster data links (10Gbps) and as such needs tighter tolerances and higher data transmission speed and frequency at 500MHz.
I tested the Neutrik cat 6A pass through panel mounts and there as good if not better than the cat 6A couplers. Because their ethercon on one side and RJ45 on the other they can actually get the connections even closer than the EtherCon to EtherCon couplers do The Cat5e pass throughs have a little circuit board and I haven't tested those but I'm guessing they show up somewhere between
Everything points to the higher frequency being carried by the wire the more impacted by the couplers it will be. Since analog audio only goes to 20K The coupler should have no impact whatsoever or no impact that matters. For AES that goes to 48 or 96K again we're not that high of a frequency compared to 100 megahertz being carried by network traffic. And the couplers really start to impact the signal when we get up into the 500 megahertz and beyond. So I wouldn't worry about couplers in most situations except if you're running very high frequencies and also running very long cables in the 80 m range or longer One exception to this is AES 50 which is extremely temperamental with cable lengths over 60 m unless the cable is very high quality so minimizing couplers on AES 50 is always wise when running longer lengths
Video Idea: PLEASE do this exact same style testing video with “hi-fi/audiophile” grade audio cables. I’d LOVE to see the actual difference if any in a $10,000 snake oil speaker cable and an $8 Hosa cable compared.
Oh my yeah the problem is getting a hold of the $10,000 audio cables. And yes it's snake oil I don't have the ability to measure and we won't hear a difference. That's it Total crap cables can be bad and that I'll dive into. I was thinking about doing a video on cheap mic cables versus quality mic cables and how they sound microphonic and shielding and such.
Hey Dave great video! A question that's totally off-topic: do you wire the ground wire to the shielding of your XLR cables? Thanks for the awesome videos!
Great question and That's going to be the topic of my next video. XLR Shell grounding, good or bad? Where I will show the advantages and disadvantages of grounding those XLR shells
I wonder how a cat6a keystone coupler would perform vs the cat6a coupler tested here? Probably exactly the same without the rubber on the outside, but they cost 1/8th of what the NE8FFX6-W does.
🤔 Serious food for thought... This is making me wonder how much signal Degradation is incurred, by adding a panel mount connectors and short patch cables in a couple of racks I'm currently designing.. Thanks Dave for yet another great insightful video.
The cat6A passthrough panel mount connectors are quite good and as good if not better than the cat6A couplers The Cat5e panel mount passthrough connectors are better than the Cat5e couplers but not as good although the Cat 6a couplers and pass through. For Rat, we have tested and standardized on 100 meter supercat XM cables plus a Cat6a passthrough on each end in the rack or console and jumper to the console or stage rack. And with this setup we can add up to 30 meters of supercat XM and still be good and stable.
@DaveRat Thanks for your reply. This is the exact scenario in wich I was already thinking. The only mayor difference would be the cable used as Ratsound Supercar Cable is as close to unobtainium as rocking horse kaka, at least here in The Netherlands. Therefore I'm currently investigating Klotz and Talisker cables wich can easily be obtained from various courses. Hope you have a Merry Christmas and all the best for 2025.
@@pauljanssen1325if you’re using Ethernet cable, any certified Cat 5E or higher cable is going to be fine. Then cables from audio companies are great, but unless you need them to be really rugged, they don’t perform any better for the price difference. Rule of thumb with every cabling system is to have as few terminations as possible in a system
Yup, important to test the runs in-situ as well, since things like wireless systems / RF cabling / power cables in close proximity to the network runs can cause interference as well. If using punched-down patch panels, the quality of the punch down work matters a lot too, esp if you actually need Cat6 spec.
Cool cool @pauljanssen yeah if your cable runs aren't pushing 100 m or more and your 80 m or under then finding a cat 6A cable should not be too tough. That said I kind of assumed that finding a cat6a cable with 19 dB or less loss and 100 megahertz 100 m would not be overly difficult especially for install grade cable with solid conductors. Someone said they were looking at using the Super Cat XM Cat6a cable for an install where a 120 meter run would save them much grief and my response was that they should just find and buy a less expensive less robust, solid conductor and fire rated cable that would be better for the application. So I started looking and wow, try finding 100 m cable with less than 19 DB of loss at 100 Mhz And let me know what you come up with. When looking for cables for applications where you're going to push the 100 m limit or go beyond it you definitely will want to look at cables with minimal insertion loss at the frequency that you'll be using them. The spec for 100 megahertz insertion losses 24 dB maximum. So if you can find a cable that's 18 DB of loss there's a pretty good chance you can run it reliably up to 130 m for so. Or looking at it the other way You've got 30% more safety margin for degradation and reliability then you would with a cable of 24 dB of loss
I'm sure most of us would only use a coupler if our life depended on it. I'm glad to have an idea exactly what the impact looks like. I'd also be interested to know how one of these out of spec signals would affect Dante or other popular digital formats. For example would a run slightly out of spec cause audible issues, would they cause the connection to reset and lose everything? In general, what would you see/hear on an out of spec situation?
Here is a video I did on AES50 but have since tested Dante in a similar way. Will do a vid at some point. Dante tends to be more stable than aes50 and has dropouts rather than loud pops and crackles. But all in all, the results are similar. ruclips.net/video/Fn3OanfKcnc/видео.html
Much like Dave writes below, the stream formats, like AES50, MADI, AES3 et c will have bit errors that are audible directly whereas the Ethernet carried packet formats like AES67, ST2110, AVB, NDI, etc etc will cause silence. This is because the entire packet (typically 1ms of audio at 48kHz; the "ptime" value in the RTP stream description) is discarded for one bit error, and the conversion from RTP stream packet back to bitstream of samples (remember, there's a PTP time stamp on the packet telling us when it was sampled)) has no choice but to leave a hole because data for that time is missing. The error is the same, from a transmission perspective -- a 0 becomes a 1 or other way around, but the effects may vary depending how the data is chunked. Related, compressed video streams might have longer dropouts because the more data might be required to restart the playback.
Things that mess with the signal lack of twists, lack of shielding, and abrupt changes in things like wire size and amounts of metal in line. I'm guessing that Cat6a rated punch connectors will test somewhere like the cat 6A couplers and cat 5e rated punch connectors won't do as well
Good day Dave! Ofcourse your videos are amazing, been watching all of them. What network tester are you using for this video? Looks like some advance form of Fluke
Thanks for the excellent video! Insertion loss in data cables is something I've often wondered about, but didn't have the equipment to test. Question: If you had one of the out-of-spec runs creating errors, would it have an effect on any other devices connected to the same network switch? Thinking specifically about AVB or Dante applications.
Being out of spec for a cable run may or may not impact the signal. The probability of impacting the signal depend on whether the spec that you're testing to correlates to the signal being used. I hate to make guesses but I can tell you that for deployment at our sound company we use 100 m Super Cat XM cable that you see in the video plus a cat 6A panel mouth pass through on each end plus a 3 ft jumper cable from the pass through to the console or rack as our standard. We also tested and found that we can add up to 40 meters more Super Cat XM cable and three cat 6A couplers and still have reliable signal integrity.. Based on this testing, we have established at 120 meters Plus the pass-throughs and jumpers, is the maximum allowable lengths for our Dante and Milan applications.
While you can't test them to spec (2GHz) I'd love to see if the cat8 couples perform meaningfully different up to 500MHz. The ones i have are built like the 6a couplers so i imagine theyd be pretty much identical, but maybe theyll be better on some parts still.
Another super interesting video! With the large number of jumper results it almost looks like comb filtering, as if the jumper causes a reflection due to impedance mismatch at some frequencies but not others? Would it be possible to test the impact of one jumper at different points in the cable, to see if the frequencies that are affected move down the further away from the end?
great data to be sure! usually 100 meters is the spec limit i shoot for... as a rule of thumb i add 10 meters to the total length for each termination point so far side distro (10M) 60 m of run then the patch panel (10m) for a total of 80M. This quickie rule should keep you functional but the 6a couplers with a good solid end termination is the way to go. edit* i now need to go talk to the boss to get that fluke qualification meter.....
Hey Dave - will you look into what happens if you send analog and Aes signals down that line with 10 connectors. Could be interresting. Happy x-mas to you
Interesting thought in Merrry Christmas. All this is in the hundred megahertz plus range where all the stuff shows up. For analog and AES which are much much lower frequencies, The impact of these couplers is negligible. I did do a video testing AES 3 with lots and lots of the Cat5e couplers. I think I found it adding a cat 5e coupler equates to about adding two or three feet of cable
I would like to see a test where you modify the Cat 5E couplers by adding the proper amount of twist inside the coupler. I have a feeling this would improve its performance
Interesting idea. I will ponder that. Though I don't think the performance will ever get up to what the cat 6A couplers can do. Maybe someone else here knows more about this but I believe that going from a very thin wire to a thicker wire and back to a thinner wire also creates issues at very high frequencies. So I think you'd want to also change the wire gauge to match the cable wire gauge inside the coupler. The other issue is the Cat5e couplers have little circuit boards in them so there's yet another set of connections there. RJ45 to female ethercon wiper to circuit board to wire to circuit board to female ethicon wiper to RJ45 is the signal path for the Cat5e coupler The cat 6A coupler is RJ45 to ethercom wiper to circuit board to ethercon wiper to RJ45
Those ripples are probably because of reflections over reflections and the fact that all shorter cables have 10 meters. The Cat6a couplers seem much better. There must be impedance matching on those couplers.
Yeah having a bunch of identical cable lengths probably sets up some resonances and other crap. It's pretty much a worst case scenario which is part of the reason I did it this way. The other reason is we only make that red cable in three lengths. 10 m, 60 m and 100 m. So I had a pile of available 10 m pieces
IBM would say that connections are the source of most of their problem s from the 50 and 60s . history yes but it applies here the fewer the better . Have seen some of the cheap connectors cause all kinds of standing wave problems and what can be viewed as ghost signal . This is a very clear and proper coverage of the problem nice data there Dave. Funny Cheech and Chong reference.
Merry Christmas and that would be great if they did. Since the a couplers are about an inch an long you need 12 of them per foot or about 33 per meter so about 330 of them for 100 m run. Plus 229 Cat6a male to male adapters. Would that be better than a cat 5e I don't know but it seems suspicious
@ I hoped you remembered me as the kind to make such a joke. Yes, a solid train of couplers, half with male ends, half with female ends. By stock in the coupler maker!
Possibly worth qualifying that the 100MHz bandwidth is acceptable for speeds up to 1Gbps - the 500MHz bandwidth is for 10Gbps+ networks - not throwing away my CAT5E couplers quite yet considering most network requirements! But future purchases will definitely be the CAT6A models...
This is very important information for if you have a stagebox in a rack with panel mount ethercon and one at FOH for the X32. This is very important. Allen and Heath may be a similar situation. It would be interesting to test a similar configuration with Digico. Theres a video series idea!!! Test a bunch of major consoles using several couplers between the surface and the stagerack. That would be cool. I have successfully mixed shows with my LV1 where I can run the Cisco switch at FOH into a jack panel (6 inch Cat6A), into a 10ft Cat6A extension with a cat5e coupler across a 100 meter cat5e snake with another 10ft Cat6a extension, with another Cat5e coupler, plugged into the jack panel on the stage rack plugged into the stage Cisco switch using a 3ft length. So I have successfully used 4 couplers in that long of a length. If I exceed that length connecting the Cisco switches it will fail. So I can do roughly 350ft between those switches. This can probably apply to using Cisco switches for Dante as well. I would never attempt this with AES50. AES50, do the Midas or Behringer. For those AES50 consoles, I insist on the AES50 port on the console into the AES50 port on the stagebox with no couplers between. With enough couplers between the desk and stagebox, you will hear a degradation in audio quality.
and ya always gotta remind people that when something is stating a rated bandwidth, that's at the 3db down point on the rolloff curve; that's one reason that empirical results might not match as close to expected as ... uh ... expected.
If you need to run near 300ft, or around 100 meters, then you should just use single mode fiber. The cost of 10gbe single mode sfp modules is pretty cheap now. 10gtek and qsfptek sell aftermarket transceivers for about $10 or so each now. And single mode is not that expensive.
Interesting and you kind of hit the nail on the head for the reason this video was created and these tests. We have about 30 drive systems all based on 100 m lengths that have fiber runs included on our main drive snakes. We have been running into issues with fiber breakage as well as long lead times for repairs and high costs for repairs and replacements. Finding a robust and reliable alternative to fiber is one of the reasons I designed this red cable and exactly why I'm doing this specific test. The goal is to make sure that we can run 100 m of this red cable plus two couplers plus two short jumpers, one on either end. The results for our application was that we can run up to 120 m + the couplers and jumpers and still have a real world reliability and function buffer of at least 20 m beyond that. We were able to reliably run 150 m cable plus couplers and jumpers with Dante and Milan In the past 6 to 8 months we have converted all of our system main snakes to have two of these red cables Cat6a runs on them and moved fiber to separate reels that we send as an additional option that can be run as an add-on should the technician or tour desire to do so, but increases load in and loadout time By making this change we've increased reliability, maintained ease of setup and increased redundancy, as now each tour has four long runs available to them, two cat 6A runs and two fiber runs should they choose to deploy them. But yes lots of people just run fiber and also throwing money at the high tech solution is sometimes worse.
How much time do you think I have! I do love testing and also I try and focus on test either solve issues that I'm working on and share the results. Or address industry confusion or heated debates and misunderstandings. Let me know tests that I can do that are related to hotly debated topics, or help solve practical real-world issues that people are facing
That frequency related ringing looks almost like an impedance mismatch. You see this in RF cables with SWR. Those Cat5e jumper leads would have terrible impedance at higher frequencies. The Super cat cable World have very good impedance and where they mate you get a mismatch and therefore reflections bad SWR. Now it just so happens at special frequencies you'll get great performance and also some terrible performance. That's a kind of laymans explanation of SWR and what your graphs look like to someone that has a RF background. 500MHz is well and truly in the radio spectrum. SWR is seen transmission lines. A rule of thumb is that if you can have more than one wavelength of a signal on the cable at any time you may have transmission line effects to consider. 100M with will have many wavelenghts even at 100Mhz. So the more i think about it i think you have transmission line effects at play here. Cool content as always Mr Rat.
Super cool and thank for the insight and info. And agreed. Seems the signal when it especially hits the Cat5e couplers the drastic change in wire gauge, circuit board, and connections all combine to create a reflection due to impedance mismatch. Or something down that line. The full understanding is outside my realm of expertise but I am good at testing and coming up with deployable methods that are reliable
one of the FIRST things you learn in communications is that the more data, the shorter the cable has to be. learned later on that if you oversaturate your data lines trying to pump in more data than the bandwidth you have to cover it, you will start having issues due to data losing its timing clocks (fix is to unplug users). wired > wireless but the same rules apply, the further away you are from the source, the weaker the signal is honestly, would love to find people who talked about 8P8C and how it is NOT RJ45 and wtf a registered jack actually is
And getting lots of data down long cables that are reliable remains an important quest. Balancing reliability versus data capability is a challenge indeed. The main drive to make this video is based on testing 100 m CAT6A cables add to all our sound systems as an alternative to the fiber reels cables. We've been having problems with the technicians breaking the fiber on our touring sound systems and high costs and long lead times for repairing or replacing the fiber. Finding a cat 6A solution that's capable of being reliable and carrying the data load at a lower cost with more robust durability as a desirable solution. I actually designed and we manufacture this red cable that allows running Dante, Milan, AES 50 and other formats at distances beyond their specified 100 m, reliably. This test came out of the goal to ensure that the 100 m long length plus two couplers plus two short jumpers, one on either end would meet spec and be reliable. It turns out we were able to run even longer lengths and can go up to 120 m with the jumpers and extensions and still have a significant buffer before we run into real world signal drop. And yes I agree with you length, plus I'm out of information which tends to equate to a higher frequency data rate and interference are all issues as well as the number in as well as the quantity and quality of connections in line, are all important factors to consider
(Single mode) Optical has the edge here, as long as it is maintained and handled correctly. Length ceases to be a problem, too. Think the longest run we have in our facility (large TV studio complex) is just short of a US mile, and as it is in wiring closets, it has none of the "roadie connecting stuff" issues. Our portable gear does see a lot of ferrule wear and some resultant reliability issues, of course. However, since even a "normal" 10G SFP+ is good for "10 km" of attenuation, having 6 connections on a 1km run (fairly normal if you go device > patch panel > intermediate wiring room > patch panel > device) is a no-brainer. As long as the couplings are clean, of course. I'd.be interested in the kind of penetration optical has in live sound, and what the issues are. Since I'm in TV, I have zero chance of escaping, because uncompressed video...
Fiber is great and very common in live sound. Llive sound is a broad rangee that includes fixed installations . And touring and mobile setups of many sizes and magnitudes. Rel Does fiber have the edge? We use fiber on delay clusters where the lengths exceed 100 meters or whatever the fiber is in the ground like for Coachella festival For 100 meter runs, fiber is common but using a tool that has capabilities that exceed the need of the application while being less robust and more costly actually a wise choice? Yay fiber, let's use it everywhere! But be very careful and run it separate ly and it will take extra time to deploy and cost more and break more often and is harder to fix and the in house tools and training cost a fortune and you'll need several different types Or Hey, here is a CAT6a cable that is less expensive, more reliable, faster to deploy as it can be taped to to the main snake, easy to fix, easy to repair, works for multiple applications and as long as you don't need to extend beyond 120 meters, works perfectly. Digico consoles have built in MAdI with fiber connections as the best interconnect for them. So for digico, not using fiber makes things tough. But, I also realize the allure of pretty light is powerful, and most or many people would rather have higher cost and less robust solutions because they have capabilities well beyond what is needed.
@@DaveRat Yeah, I get exactly why you have come to the conclusion you have, and it makes sense in your application, hands down, no discussion. Copper is, as long as you can stay within a manageable operation envelope, much, much easier to deal with. Field repairs, for starters, are practical, and you don't need to get 3 kilobucks of microscope and another 3 kilobucks of light source and calibrated level meter to just verify the system. (And then you sort of don't have bothered with the OTDR or the optical spectrum analyser, both 30K instruments with annual calibration.) So, staying on copper when possible makes sense. I can't, and once we take the plunge, it's much better to learn and do it 100% right. I've seen so many people shun fiber because they haven't understood what it takes, and your attitude to the copper cable science shows you have the chops to take on optical too -- but perhaps not all the people you rely on have the time and motivation to do it properly. So kudos to you for the practical considered solution, well put together.
Fully agreed and yeah it's all about using the proper tool for the job and not being emotionally biased. For us we actually shifted All of our main snakes include fiber and a redundant fiber as well. Costs and failures and maintenance were annoying but what really got us back to copper was two things. First was lead time on ordering additional fiber along with repair time frames for the cables we had to send out. We just couldn't keep enough functioning fiber in stock to deploy during the busy season. Parallel with that I was developing that red cable for the sound tools division to solve long runs of AES50 and create a high quality durable cable capable of exceeding the 100 m limits of Dante, Milan, and AES 50 along with other formats. So with the pressure on us to find a solution to our fiber challenges combined with the availability of a cable that can carry our signals up to 150 m, or carry the signal for 120 m with a 20% safety margin, he scenario was right for making the switch. We now send fiber on separate reels that the tours can run if they wish but we've standardized on CAT6A as the baseline deployment. And we've had zero failures! It's all about looking for and using the best tool for the job! And thank you
I think if you're using that many couplers, you really need to examine your cabling choices. Of course, that's not the point of this. I think in general, this is a great example of why one should avoid any couplers whenever possible. It does also show a strong reason why an improved product is the better choice.
Yeah I do a lot of testing where in order to clearly identify and illuminate an issue I overdo it so that we can know what to look for When testing digital consoles and comparing them to analog consoles, The difference is in sound and noise are not readily clearly audible comparing one channel of digital to one channel one channel of analog. But put 10 channels in series with 10 d to a and 10 a to D converters and compare it to an analog console going through 10 input channels and then the differences become more pronounced and we know what to look for and listen for. I used this methodology in this test as well of using too many couplers. I see it as a similar method to having a magnifying glass we were able to magnify the issues such that we can hear or see them better
@@DaveRat I'm with the other RF interested people here, that is RF transmission line stuff happening, mismatches and all. No surprises, and a very nice illustration of the problems.
Yeah splicing the wire is definitely an option. But even there there can be issues. Maintaining the twists all the way through and not having straight wire and making sure the twisted pairs have a shield that's continuous are important factors. The distance that the wires in the Cat6a coupler are not twisted is extremely short and the shielding is maintained. I would bet the splicewire rivals the cat 6A coupler depending on how much time and care you put into that splice
Well, before making statements or claims, maybe we should test and confirm the statements or claims are credible? And if you are wondering if the tester shows any differential in test results between a coiled and non coiled cables, let me know as I have tested that as well.
You just inadvertently created the greatest commercial for these Fluke testers. Making next year's holiday wishlist already, haha!
Those things are great but they're definitely not cheap
@@DaveRat I'm actually a little surprised you have the Versiv 2. We use that at my work to certify structure cabling. It's extreme.
Yeah, I bought to test and make sure the Sound Tools cables we manufactured meet specs and to help test the prototype cables as we design new cables
I agree since I am now combing the comments to determine precisely which tester he is using to test the cables. Dave could have made it easier by including the name. Better yet, he could have included an affiliate link to the exact tester and made some money while educating.
@@jatlastit’s a Fluke networks DSX CableAnalyzer, copper cable certifier. Part of their Versiv family of products. Model number is DSX2-5000, cost around $13,000. The DSX2-8000 = $15,000.
Thanks
Wow and thank you Olly and much appreciated!
Use 6a couplers (no more than 2 per run) got it 💪🏻 thanks Dave.
Exactly, unless your cables are shorter in which case you can be less strict
@@DaveRatIs this the model you tested?
Neutrik NE8FFX6-W
Yes, the two models with part numbers are shown at 00:11
"Stars on RJ-45"
I'll see myself out.
Thanks for the spectest! Invaluable stuff, few of us ever get around to taking a hard look, too busy making do I guess.
🤙👍🤙
Your shirt reminded me of my first real job in 1973, I was an auto parts delivery guy. I'd make the deliveries and at the last stop, I had to call the warehouse to let them know I was done and on my way back, in case I needed to do a pickup on my way in. I'd call and say, "This is Dave, I'm on my way back." And, of course I'd get the "Dave's not here" response every time.
Perfect and fun!
Great and extremely useful information again. Thanks!
🤙👍🤙
Thanks again, Dave, you are the best! 🙂
Cool cool Edwin
another great video at the end of the 70s I was testing telephone cables we had a lorry full of equipment I think those two little boxes on your bench do the same job if the cables were loaded impedance matched there would be some improvement I remember your test with a kilometre long microphone cable 😊
There's a formula line resistance line inductance continuity Shunting The line capacitance shunting the line and the only practical thing you can do is increase the Capacitance
Another thing with connections Is when the material changes the different energy levels in different conductors create something called the sea back Affect
You could also put analog repeaters in the line
Awesome stuff Thanks for sharing Dave
Cool cool thank you Hennie!
Is there any device like a booster or filter/fixer that could be added somewhere in the run that could help with the quality? Thanks
Shorter runs will have less issues. Yes, you can add switches or boosters to many protocols but they add latency and complexity and another thing to go wrong. Especially for festivals where they would need to be somewhere in the dirt between stage and delay clusters or FOH.
Using high quality wide bandwidth cables and avoiding couplers with wide bandwidth signals is good
Or switch to fiber. But fiber adds cost and is more fragile
@ It all makes sense. Don’t be short is the answer. Thank You 🙏
Im wondering if the 10x10 meters has any influence... What would happen if it were to be 10 cables of different lengths? what the harmonic bumps be more spread out and less pronounced?
Interesting. Definitely can measure and see where couplers are and cable lengths in measurements.
Some tests seem to indicate that having very short jumper cables right near the ends is especially undesirable and creates reflections and resonances where the signal level is highest.
Very interesting as usual. I have a small selection of lengths and a 5e coupler to give me more distance options (currently being used on the LAN until I make some cables up). Only running short distances compared to you guys, but might switch up to the 6.
At 7:05
That is great advice for all electronic tests!
👍😁👍
Thanks for this. Question: is there any chance that using a Cat-5e coupler with the SuperCat 6 cable creates a mismatch? I know jack squat about these cables and how they affect signal loss/stability, but for me, as a touring FOH A1, I will always fight for a single cable from mixer to stage box without connectors. It’s amazing how many house techs fail to see the importance of reducing possible points of failure.
I didn't do it in the video but I did test the Cat5e couplers with Cat5e cable and the cat 6a couplers as well. The cat 6A couplers outperform the Cat5e couplers in every scenario.
But since Cat5e is only rated up to 100 megahertz, both couplers perform good enough in that range.
The sensitivity to couplers really happens at frequencies well above 100 megahertz where the cat 5e 100 megahertz coupler performs poorly and the cat6 a coupler before much better
@ is 100 mHz synonymous with 100 m length?
@@Brody-Dolyniuk It's not linked to cable lenght. It is the data transmission speed that the CAT5e cable was designed for (100MHz and around 1Gbps of network traffic). CAT6a is designed for faster data links (10Gbps) and as such needs tighter tolerances and higher data transmission speed and frequency at 500MHz.
👍🤙👍
I wonder how wall terminations would compare to these couplers?
I tested the Neutrik cat 6A pass through panel mounts and there as good if not better than the cat 6A couplers. Because their ethercon on one side and RJ45 on the other they can actually get the connections even closer than the EtherCon to EtherCon couplers do
The Cat5e pass throughs have a little circuit board and I haven't tested those but I'm guessing they show up somewhere between
Out of curiosity, what is the impact on signal if you are using it to carry analog and/or aes pairs?
Everything points to the higher frequency being carried by the wire the more impacted by the couplers it will be.
Since analog audio only goes to 20K The coupler should have no impact whatsoever or no impact that matters.
For AES that goes to 48 or 96K again we're not that high of a frequency compared to 100 megahertz being carried by network traffic. And the couplers really start to impact the signal when we get up into the 500 megahertz and beyond.
So I wouldn't worry about couplers in most situations except if you're running very high frequencies and also running very long cables in the 80 m range or longer
One exception to this is AES 50 which is extremely temperamental with cable lengths over 60 m unless the cable is very high quality so minimizing couplers on AES 50 is always wise when running longer lengths
@DaveRat Thank you Dave...
👍🤙👍
Video Idea: PLEASE do this exact same style testing video with “hi-fi/audiophile” grade audio cables. I’d LOVE to see the actual difference if any in a $10,000 snake oil speaker cable and an $8 Hosa cable compared.
Oh my yeah the problem is getting a hold of the $10,000 audio cables. And yes it's snake oil I don't have the ability to measure and we won't hear a difference.
That's it Total crap cables can be bad and that I'll dive into.
I was thinking about doing a video on cheap mic cables versus quality mic cables and how they sound microphonic and shielding and such.
@ that’ll work! Finally settle the debate on why you shouldn’t use XLR cables to run 3 pin DMX 😂
Hey Dave great video! A question that's totally off-topic: do you wire the ground wire to the shielding of your XLR cables? Thanks for the awesome videos!
Great question and That's going to be the topic of my next video.
XLR Shell grounding, good or bad?
Where I will show the advantages and disadvantages of grounding those XLR shells
I was literally looking this up today , also would like to know.
Cool cool I'm on it and it's interesting
@@DaveRatGood to know! Will be looking forward to this one. Thanks a lot for doing these kind of experiments!
Cool cool just finished the video it will go up on the member side of the channel first and then move to the public side probably late January or Feb
I wonder how a cat6a keystone coupler would perform vs the cat6a coupler tested here? Probably exactly the same without the rubber on the outside, but they cost 1/8th of what the NE8FFX6-W does.
Yeah, probably very close to the same and if your application does not need ethercon, that may be a good way to go
🤔 Serious food for thought... This is making me wonder how much signal Degradation is incurred, by adding a panel mount connectors and short patch cables in a couple of racks I'm currently designing.. Thanks Dave for yet another great insightful video.
The cat6A passthrough panel mount connectors are quite good and as good if not better than the cat6A couplers
The Cat5e panel mount passthrough connectors are better than the Cat5e couplers but not as good although the Cat 6a couplers and pass through.
For Rat, we have tested and standardized on 100 meter supercat XM cables plus a Cat6a passthrough on each end in the rack or console and jumper to the console or stage rack.
And with this setup we can add up to 30 meters of supercat XM and still be good and stable.
@DaveRat Thanks for your reply. This is the exact scenario in wich I was already thinking. The only mayor difference would be the cable used as Ratsound Supercar Cable is as close to unobtainium as rocking horse kaka, at least here in The Netherlands. Therefore I'm currently investigating Klotz and Talisker cables wich can easily be obtained from various courses. Hope you have a Merry Christmas and all the best for 2025.
@@pauljanssen1325if you’re using Ethernet cable, any certified Cat 5E or higher cable is going to be fine. Then cables from audio companies are great, but unless you need them to be really rugged, they don’t perform any better for the price difference. Rule of thumb with every cabling system is to have as few terminations as possible in a system
Yup, important to test the runs in-situ as well, since things like wireless systems / RF cabling / power cables in close proximity to the network runs can cause interference as well. If using punched-down patch panels, the quality of the punch down work matters a lot too, esp if you actually need Cat6 spec.
Cool cool @pauljanssen yeah if your cable runs aren't pushing 100 m or more and your 80 m or under then finding a cat 6A cable should not be too tough.
That said I kind of assumed that finding a cat6a cable with 19 dB or less loss and 100 megahertz 100 m would not be overly difficult especially for install grade cable with solid conductors.
Someone said they were looking at using the Super Cat XM Cat6a cable for an install where a 120 meter run would save them much grief and my response was that they should just find and buy a less expensive less robust, solid conductor and fire rated cable that would be better for the application. So I started looking and wow, try finding 100 m cable with less than 19 DB of loss at 100 Mhz And let me know what you come up with.
When looking for cables for applications where you're going to push the 100 m limit or go beyond it you definitely will want to look at cables with minimal insertion loss at the frequency that you'll be using them.
The spec for 100 megahertz insertion losses 24 dB maximum. So if you can find a cable that's 18 DB of loss there's a pretty good chance you can run it reliably up to 130 m for so.
Or looking at it the other way You've got 30% more safety margin for degradation and reliability then you would with a cable of 24 dB of loss
I'm sure most of us would only use a coupler if our life depended on it. I'm glad to have an idea exactly what the impact looks like.
I'd also be interested to know how one of these out of spec signals would affect Dante or other popular digital formats. For example would a run slightly out of spec cause audible issues, would they cause the connection to reset and lose everything? In general, what would you see/hear on an out of spec situation?
Here is a video I did on AES50 but have since tested Dante in a similar way. Will do a vid at some point.
Dante tends to be more stable than aes50 and has dropouts rather than loud pops and crackles.
But all in all, the results are similar.
ruclips.net/video/Fn3OanfKcnc/видео.html
@@DaveRat Cheers!
🤙👍🤙
Much like Dave writes below, the stream formats, like AES50, MADI, AES3 et c will have bit errors that are audible directly whereas the Ethernet carried packet formats like AES67, ST2110, AVB, NDI, etc etc will cause silence. This is because the entire packet (typically 1ms of audio at 48kHz; the "ptime" value in the RTP stream description) is discarded for one bit error, and the conversion from RTP stream packet back to bitstream of samples (remember, there's a PTP time stamp on the packet telling us when it was sampled)) has no choice but to leave a hole because data for that time is missing.
The error is the same, from a transmission perspective -- a 0 becomes a 1 or other way around, but the effects may vary depending how the data is chunked. Related, compressed video streams might have longer dropouts because the more data might be required to restart the playback.
Cool and thank you for the better explanation!
I wonder if punch down couplers would perform better than the rj45 connections.
Things that mess with the signal lack of twists, lack of shielding, and abrupt changes in things like wire size and amounts of metal in line.
I'm guessing that Cat6a rated punch connectors will test somewhere like the cat 6A couplers and cat 5e rated punch connectors won't do as well
Doing the lord's work!
🤙👍🤙
Good day Dave! Ofcourse your videos are amazing, been watching all of them.
What network tester are you using for this video? Looks like some advance form of Fluke
Fluke
Versiv Dsx-5000
Haaa. Love the T Shirt !!! Only us old school dudes know
🤙👍🤙
Love the shirt maaaaann! 🤣🤣🤣🥰 Edit: wonder what a switch in the mix does?
👍🤙👍😁
Thanks for the excellent video! Insertion loss in data cables is something I've often wondered about, but didn't have the equipment to test. Question: If you had one of the out-of-spec runs creating errors, would it have an effect on any other devices connected to the same network switch? Thinking specifically about AVB or Dante applications.
Being out of spec for a cable run may or may not impact the signal. The probability of impacting the signal depend on whether the spec that you're testing to correlates to the signal being used.
I hate to make guesses but I can tell you that for deployment at our sound company we use 100 m Super Cat XM cable that you see in the video plus a cat 6A panel mouth pass through on each end plus a 3 ft jumper cable from the pass through to the console or rack as our standard.
We also tested and found that we can add up to 40 meters more Super Cat XM cable and three cat 6A couplers and still have reliable signal integrity..
Based on this testing, we have established at 120 meters Plus the pass-throughs and jumpers, is the maximum allowable lengths for our Dante and Milan applications.
While you can't test them to spec (2GHz) I'd love to see if the cat8 couples perform meaningfully different up to 500MHz. The ones i have are built like the 6a couplers so i imagine theyd be pretty much identical, but maybe theyll be better on some parts still.
I don't know of a cat8 ethercon coupler.
I know there's RJ45 Cat8 versions out there. Are you aware of any cat8 ethercon couplers?
@DaveRat sadly not any ethercon ones either, only normal rj45.
Another super interesting video! With the large number of jumper results it almost looks like comb filtering, as if the jumper causes a reflection due to impedance mismatch at some frequencies but not others? Would it be possible to test the impact of one jumper at different points in the cable, to see if the frequencies that are affected move down the further away from the end?
Check the video around 13:18 where I do 2x 60m
@@DaveRat so you did, guess I need to be more observant! Thanks!
All good.
Man I’m old I remember being worried about a whirlwind 24 ch. snake with monitor split.
great video. very well outlined.
👍🤙👍
great data to be sure! usually 100 meters is the spec limit i shoot for... as a rule of thumb i add 10 meters to the total length for each termination point so far side distro (10M) 60 m of run then the patch panel (10m) for a total of 80M. This quickie rule should keep you functional but the 6a couplers with a good solid end termination is the way to go. edit* i now need to go talk to the boss to get that fluke qualification meter.....
🤙👍🤙 those flukes are expensive!
Hey Dave - will you look into what happens if you send analog and Aes signals down that line with 10 connectors. Could be interresting. Happy x-mas to you
Interesting thought in Merrry Christmas.
All this is in the hundred megahertz plus range where all the stuff shows up.
For analog and AES which are much much lower frequencies, The impact of these couplers is negligible.
I did do a video testing AES 3 with lots and lots of the Cat5e couplers.
I think I found it adding a cat 5e coupler equates to about adding two or three feet of cable
I would like to see a test where you modify the Cat 5E couplers by adding the proper amount of twist inside the coupler. I have a feeling this would improve its performance
Interesting idea. I will ponder that. Though I don't think the performance will ever get up to what the cat 6A couplers can do. Maybe someone else here knows more about this but I believe that going from a very thin wire to a thicker wire and back to a thinner wire also creates issues at very high frequencies.
So I think you'd want to also change the wire gauge to match the cable wire gauge inside the coupler.
The other issue is the Cat5e couplers have little circuit boards in them so there's yet another set of connections there.
RJ45 to female ethercon wiper to circuit board to wire to circuit board to female ethicon wiper to RJ45 is the signal path for the Cat5e coupler
The cat 6A coupler is RJ45 to ethercom wiper to circuit board to ethercon wiper to RJ45
Thanks for this. Super helpful. I've always wondered how much those things were affecting the signal.
Thank you Nick!
I will go ahead and leave a comment here
Those ripples are probably because of reflections over reflections and the fact that all shorter cables have 10 meters. The Cat6a couplers seem much better. There must be impedance matching on those couplers.
Yeah having a bunch of identical cable lengths probably sets up some resonances and other crap. It's pretty much a worst case scenario which is part of the reason I did it this way.
The other reason is we only make that red cable in three lengths. 10 m, 60 m and 100 m. So I had a pile of available 10 m pieces
IBM would say that connections are the source of most of their problem s from the 50 and 60s . history yes but it applies here the fewer the better . Have seen some of the cheap connectors cause all kinds of standing wave problems and what can be viewed as ghost signal . This is a very clear and proper coverage of the problem nice data there Dave. Funny Cheech and Chong reference.
Very cool and thank you John!
Exactly my kind of test
🔊🤙🤙
My question is, do A couplers make E cable work better? Merry christmas, Dave!
Merry Christmas and that would be great if they did. Since the a couplers are about an inch an long you need 12 of them per foot or about 33 per meter so about 330 of them for 100 m run. Plus 229 Cat6a male to male adapters.
Would that be better than a cat 5e I don't know but it seems suspicious
@ I hoped you remembered me as the kind to make such a joke. Yes, a solid train of couplers, half with male ends, half with female ends. By stock in the coupler maker!
👍😁👍
Possibly worth qualifying that the 100MHz bandwidth is acceptable for speeds up to 1Gbps - the 500MHz bandwidth is for 10Gbps+ networks - not throwing away my CAT5E couplers quite yet considering most network requirements! But future purchases will definitely be the CAT6A models...
Agreed and same and thank you for the clarification and useful info.
This is very important information for if you have a stagebox in a rack with panel mount ethercon and one at FOH for the X32.
This is very important. Allen and Heath may be a similar situation.
It would be interesting to test a similar configuration with Digico.
Theres a video series idea!!! Test a bunch of major consoles using several couplers between the surface and the stagerack. That would be cool.
I have successfully mixed shows with my LV1 where I can run the Cisco switch at FOH into a jack panel (6 inch Cat6A), into a 10ft Cat6A extension with a cat5e coupler across a 100 meter cat5e snake with another 10ft Cat6a extension, with another Cat5e coupler, plugged into the jack panel on the stage rack plugged into the stage Cisco switch using a 3ft length. So I have successfully used 4 couplers in that long of a length.
If I exceed that length connecting the Cisco switches it will fail. So I can do roughly 350ft between those switches.
This can probably apply to using Cisco switches for Dante as well.
I would never attempt this with AES50. AES50, do the Midas or Behringer. For those AES50 consoles, I insist on the AES50 port on the console into the AES50 port on the stagebox with no couplers between. With enough couplers between the desk and stagebox, you will hear a degradation in audio quality.
and ya always gotta remind people that when something is stating a rated bandwidth, that's at the 3db down point on the rolloff curve; that's one reason that empirical results might not match as close to expected as ... uh ... expected.
👍🤙🤙
What were they thinking! Not even twisting the cables what is this cat3 cabling!
Right? They could have twisted the cables they used
If you need to run near 300ft, or around 100 meters, then you should just use single mode fiber. The cost of 10gbe single mode sfp modules is pretty cheap now. 10gtek and qsfptek sell aftermarket transceivers for about $10 or so each now. And single mode is not that expensive.
Interesting and you kind of hit the nail on the head for the reason this video was created and these tests.
We have about 30 drive systems all based on 100 m lengths that have fiber runs included on our main drive snakes.
We have been running into issues with fiber breakage as well as long lead times for repairs and high costs for repairs and replacements.
Finding a robust and reliable alternative to fiber is one of the reasons I designed this red cable and exactly why I'm doing this specific test.
The goal is to make sure that we can run 100 m of this red cable plus two couplers plus two short jumpers, one on either end.
The results for our application was that we can run up to 120 m + the couplers and jumpers and still have a real world reliability and function buffer of at least 20 m beyond that. We were able to reliably run 150 m cable plus couplers and jumpers with Dante and Milan
In the past 6 to 8 months we have converted all of our system main snakes to have two of these red cables Cat6a runs on them and moved fiber to separate reels that we send as an additional option that can be run as an add-on should the technician or tour desire to do so, but increases load in and loadout time
By making this change we've increased reliability, maintained ease of setup and increased redundancy, as now each tour has four long runs available to them, two cat 6A runs and two fiber runs should they choose to deploy them.
But yes lots of people just run fiber and also throwing money at the high tech solution is sometimes worse.
now do the finger twist joints of cat5e snd test them
Don't want to cut up All this nice cat6a cable though.
while here, do test 30awg cca cables, steel and whatever they have. cut those and get more tests in same go
How much time do you think I have!
I do love testing and also I try and focus on test either solve issues that I'm working on and share the results. Or address industry confusion or heated debates and misunderstandings.
Let me know tests that I can do that are related to hotly debated topics, or help solve practical real-world issues that people are facing
That frequency related ringing looks almost like an impedance mismatch. You see this in RF cables with SWR. Those Cat5e jumper leads would have terrible impedance at higher frequencies. The Super cat cable World have very good impedance and where they mate you get a mismatch and therefore reflections bad SWR. Now it just so happens at special frequencies you'll get great performance and also some terrible performance. That's a kind of laymans explanation of SWR and what your graphs look like to someone that has a RF background. 500MHz is well and truly in the radio spectrum. SWR is seen transmission lines. A rule of thumb is that if you can have more than one wavelength of a signal on the cable at any time you may have transmission line effects to consider. 100M with will have many wavelenghts even at 100Mhz. So the more i think about it i think you have transmission line effects at play here. Cool content as always Mr Rat.
Super cool and thank for the insight and info. And agreed. Seems the signal when it especially hits the Cat5e couplers the drastic change in wire gauge, circuit board, and connections all combine to create a reflection due to impedance mismatch. Or something down that line.
The full understanding is outside my realm of expertise but I am good at testing and coming up with deployable methods that are reliable
@DaveRat no worries. You share so such much freely, feel I need to pay it back.
Thank you and appreciated
one of the FIRST things you learn in communications is that the more data, the shorter the cable has to be. learned later on that if you oversaturate your data lines trying to pump in more data than the bandwidth you have to cover it, you will start having issues due to data losing its timing clocks (fix is to unplug users). wired > wireless but the same rules apply, the further away you are from the source, the weaker the signal is
honestly, would love to find people who talked about 8P8C and how it is NOT RJ45 and wtf a registered jack actually is
And getting lots of data down long cables that are reliable remains an important quest.
Balancing reliability versus data capability is a challenge indeed.
The main drive to make this video is based on testing 100 m CAT6A cables add to all our sound systems as an alternative to the fiber reels cables.
We've been having problems with the technicians breaking the fiber on our touring sound systems and high costs and long lead times for repairing or replacing the fiber.
Finding a cat 6A solution that's capable of being reliable and carrying the data load at a lower cost with more robust durability as a desirable solution.
I actually designed and we manufacture this red cable that allows running Dante, Milan, AES 50 and other formats at distances beyond their specified 100 m, reliably.
This test came out of the goal to ensure that the 100 m long length plus two couplers plus two short jumpers, one on either end would meet spec and be reliable.
It turns out we were able to run even longer lengths and can go up to 120 m with the jumpers and extensions and still have a significant buffer before we run into real world signal drop.
And yes I agree with you length, plus I'm out of information which tends to equate to a higher frequency data rate and interference are all issues as well as the number in as well as the quantity and quality of connections in line, are all important factors to consider
(Single mode) Optical has the edge here, as long as it is maintained and handled correctly. Length ceases to be a problem, too. Think the longest run we have in our facility (large TV studio complex) is just short of a US mile, and as it is in wiring closets, it has none of the "roadie connecting stuff" issues. Our portable gear does see a lot of ferrule wear and some resultant reliability issues, of course. However, since even a "normal" 10G SFP+ is good for "10 km" of attenuation, having 6 connections on a 1km run (fairly normal if you go device > patch panel > intermediate wiring room > patch panel > device) is a no-brainer. As long as the couplings are clean, of course.
I'd.be interested in the kind of penetration optical has in live sound, and what the issues are. Since I'm in TV, I have zero chance of escaping, because uncompressed video...
Fiber is great and very common in live sound. Llive sound is a broad rangee that includes fixed installations . And touring and mobile setups of many sizes and magnitudes. Rel
Does fiber have the edge?
We use fiber on delay clusters where the lengths exceed 100 meters or whatever the fiber is in the ground like for Coachella festival
For 100 meter runs, fiber is common but using a tool that has capabilities that exceed the need of the application while being less robust and more costly actually a wise choice?
Yay fiber, let's use it everywhere!
But be very careful and run it separate ly and it will take extra time to deploy and cost more and break more often and is harder to fix and the in house tools and training cost a fortune and you'll need several different types
Or
Hey, here is a CAT6a cable that is less expensive, more reliable, faster to deploy as it can be taped to to the main snake, easy to fix, easy to repair, works for multiple applications and as long as you don't need to extend beyond 120 meters, works perfectly.
Digico consoles have built in MAdI with fiber connections as the best interconnect for them. So for digico, not using fiber makes things tough.
But, I also realize the allure of pretty light is powerful, and most or many people would rather have higher cost and less robust solutions because they have capabilities well beyond what is needed.
@@DaveRat Yeah, I get exactly why you have come to the conclusion you have, and it makes sense in your application, hands down, no discussion. Copper is, as long as you can stay within a manageable operation envelope, much, much easier to deal with. Field repairs, for starters, are practical, and you don't need to get 3 kilobucks of microscope and another 3 kilobucks of light source and calibrated level meter to just verify the system. (And then you sort of don't have bothered with the OTDR or the optical spectrum analyser, both 30K instruments with annual calibration.) So, staying on copper when possible makes sense. I can't, and once we take the plunge, it's much better to learn and do it 100% right. I've seen so many people shun fiber because they haven't understood what it takes, and your attitude to the copper cable science shows you have the chops to take on optical too -- but perhaps not all the people you rely on have the time and motivation to do it properly. So kudos to you for the practical considered solution, well put together.
Fully agreed and yeah it's all about using the proper tool for the job and not being emotionally biased.
For us we actually shifted All of our main snakes include fiber and a redundant fiber as well.
Costs and failures and maintenance were annoying but what really got us back to copper was two things.
First was lead time on ordering additional fiber along with repair time frames for the cables we had to send out. We just couldn't keep enough functioning fiber in stock to deploy during the busy season.
Parallel with that I was developing that red cable for the sound tools division to solve long runs of AES50 and create a high quality durable cable capable of exceeding the 100 m limits of Dante, Milan, and AES 50 along with other formats.
So with the pressure on us to find a solution to our fiber challenges combined with the availability of a cable that can carry our signals up to 150 m, or carry the signal for 120 m with a 20% safety margin, he scenario was right for making the switch.
We now send fiber on separate reels that the tours can run if they wish but we've standardized on CAT6A as the baseline deployment.
And we've had zero failures!
It's all about looking for and using the best tool for the job! And thank you
I think if you're using that many couplers, you really need to examine your cabling choices. Of course, that's not the point of this. I think in general, this is a great example of why one should avoid any couplers whenever possible. It does also show a strong reason why an improved product is the better choice.
Yeah I do a lot of testing where in order to clearly identify and illuminate an issue I overdo it so that we can know what to look for
When testing digital consoles and comparing them to analog consoles, The difference is in sound and noise are not readily clearly audible comparing one channel of digital to one channel one channel of analog.
But put 10 channels in series with 10 d to a and 10 a to D converters and compare it to an analog console going through 10 input channels and then the differences become more pronounced and we know what to look for and listen for.
I used this methodology in this test as well of using too many couplers.
I see it as a similar method to having a magnifying glass we were able to magnify the issues such that we can hear or see them better
And this is with Neutrik made connectors. Imagine how terrible it'd be with low end couplers.
🤙👍🤙
Adaptor? I barely knew her!
🤙👍😁😁😁
As an RF engineer, I expected the 5e to degrade transmission line performance.
Very cool and curious. Trying to sort the tone of the statement.
So did it do as expected? Or worse or better?
And thank you!
@@DaveRat I'm with the other RF interested people here, that is RF transmission line stuff happening, mismatches and all. No surprises, and a very nice illustration of the problems.
Cool
👍🤙👍
We need RatGPT
Fun and I guess I am the human version of Rat Gpt
@@DaveRat And the only human allowed to train it.
👍🤙👍
❤
Dave? Dont do it.
But the temptation to use a whole bunch of little tiny cables for a very long run is so overwhelmingly tempting!
@DaveRat indeed!
🤙👍🤙
I Would Just Splice The Wire! But I Own The Network! It Would be Better Than Either Of These Couplers
Yeah splicing the wire is definitely an option. But even there there can be issues. Maintaining the twists all the way through and not having straight wire and making sure the twisted pairs have a shield that's continuous are important factors.
The distance that the wires in the Cat6a coupler are not twisted is extremely short and the shielding is maintained.
I would bet the splicewire rivals the cat 6A coupler depending on how much time and care you put into that splice
No cables should be coiled when testing
Well, before making statements or claims, maybe we should test and confirm the statements or claims are credible?
And if you are wondering if the tester shows any differential in test results between a coiled and non coiled cables, let me know as I have tested that as well.