Great video, thank you!! Could you also do a little testing like that with some excessive untwist prior to punch-down blocks for wall connectors? Not sure if it would make the same difference from RJ45s but I'm curious to find out. I am aware that the specs apply to both, but it doesn't look like a lot of electricians are aware of that... Quite common to remove wallplates and find something like 4 inches of sheathing/cable jacket removed and pairs with more than an inch of untwisting before being punched into their IDC.
Hello and thanks for your compliment and comment. Indeed, testing excessive untwist at a IDC termination is also worthy of investigation. I would think things would end up the same, but who knows? We will add that to the list of videos to do!
Nice. Wish people were throwing money at me so I could get a certifier, but that's not the case. Will just have to be more mindful of this. Can you try comparing the pass through ends with the standard crimp ends with the same untwist? The idea of bare copper hanging out the ends makes me really leery of using them, but they would make it easier to push your twists in further. Would like to see if that bare copper does indeed cause any problems or whether it's fine. Thanks.
Hello! You are in luck. Although not on video, I have extensively tested untwist for performance with passthrough and non-passthrough connectors. As it turns out, the biggest factor is fitment of the cable and insulated conductors to the plug itself. The TIA standard accounts for 1/2" untwist at each connector, so achieving tighter untwist at the connector does not get you much if any more performance and can actually cause issues in some edge cases. In other words, just achieve 1/2" or slightly less untwist and you will be fine. As to the difference between passthrough and non-passthrough connectors there is no performance difference **assuming** the cable to plug fitment is good in both cases. I will say that passthrough connectors are a bit more sensitive to ill-fitting insulated conductors, but that is more a feeling generated by experience. I have used passthrough plugs for many outdoor WiFi AP and camera installations with no ill effect, and those terminations are still going strong 5+ years later. It really all comes down to cable to plug fitment and meeting the 1/2" rule. Hope this helps!
Good question Cash! No, that would only affect that single termination. Now, there could be an edge case scenario where two Cat6A terminations adjacent to each other, if they are BOTH poor terminations and BOTH terminations are being driven to 500 MHz, may impart ANEXT onto each other. In this case you would experience high NEXT on the individual terminations and ANEXT on top of it. That would be a weird scenario, but it is in the realm of possibility.
Nice video. As an engineer who has been making a server rack doing my own patch cables and Keystone Jack connections it has concerned me untwisting the wires. For the Plugs I am using your Standard RJ45 Connectors. Just looked at one and it seems under 1/2 inch and more like 1/4 inch or just over 1/4. I would like to see this test with keystone Jacks. Also what happens in the chain if I have 4 keystone jacks compared to if I have 5 or 6 keystone jacks? Is the problem additive? For me real world I ran what leaves the Rack to a patch panel with a keystone jack because it looked cleaner but that means one more connection. I would like to see a cable chain with varying number of connection combinations and test to see performance. Thanks for video
Hello Kim! You are definitely thinking correctly and have your head in the right place. First, I recommend against using solid copper Ethernet to create patch cords in the field unless you have a Fluke DSX-8000 around with PCA adapters to test each and every one. There are way too many things that can go wrong. Get matching Category patch cords that have been factory terminated and tested. Untwist at the termination is cumulative in performance degradation. Any electromagnetic disruption (a termination of any kind for example) will eventually add up to a poorly working cable run. Generally, the maximum amount of terminations per TIA 568 is going to be 6 per Channel. A Channel is from powered device to powered device. It is the full run, including patch cords. In this scenario, SIX terminations leaves you 2 x rack to jack style terminations (keystone to keystone or patch panel to keystone) and then two factory made patch cords. The quality of the terminations counts as well. Keystone or patch panel terminations are far better than hand terminated RJ45s. If you use hand terminated RJ45s you should not even exceed ONE in the Channel per the standard. The example would be keystone or patch panel to RJ45 to a PoE end point device and is known as a MPTL (modular plug terminated link). Now, it you want to use use male plugs and sort of "defeat" the TIA 568 standard in a way, then use field termination plugs. You are still limited to six, but field termination plugs perform just as well as keystone jacks! As for more testing and videos...that is on the radar!
@@trueCABLE That was helpful but left me with questions and ideas. I will check out your field termination plugs. I have thought for some cases of buying a factory cable and cutting off one end so that I have a good plug. Then putting on a Keystone jack on the other end I cut off. Still clarifying in my head the connection add up. Is the output of a switch to a cable plug 2 connections? Also a Cable plug to a Keystone jack would be considered 2 connections? I would be cool for someone like me trying to picture if you had a diagram showing what you explained as the six connections. Using your standard RJ45 connectors looks pretty fool proof. I am able to slide the yellow piece over the wires and get about 1/8 of an inch from the jacket, Trim it, insert it and crimp. Looks like a really short length. I guess this shows my lack of knowledge on the subject. Would I have access to the ANSI TIA 568 spec to read?
Does exists a Windows software for testing a network speed and shows the number of packets dropped? I know that the standard PC's NIC is maximum 1 Gbps, but I'm more interested/curious about the # of dropped packets.
Damn, you built a studio and it is crap sound? Rebuild, re-mic. It's ironic that tech guys, whom It trust, BTW, dropped the ball on the most important aspect of video: sound.
Hello and thanks for your comment. Technically, quite correct. That said, copper twisted pair Category cable with eight conductors has become defacto "Ethernet cable".
Great job on the new video! Thanks for being a thought partner on network cabling and sticking to quantifiable differences.
Thank you, Mike! We are thrilled you find our content useful :)
Great demonstration. Thank you!
Thank you! 😀
Great video, thank you!! Could you also do a little testing like that with some excessive untwist prior to punch-down blocks for wall connectors? Not sure if it would make the same difference from RJ45s but I'm curious to find out. I am aware that the specs apply to both, but it doesn't look like a lot of electricians are aware of that...
Quite common to remove wallplates and find something like 4 inches of sheathing/cable jacket removed and pairs with more than an inch of untwisting before being punched into their IDC.
Hello and thanks for your compliment and comment. Indeed, testing excessive untwist at a IDC termination is also worthy of investigation. I would think things would end up the same, but who knows? We will add that to the list of videos to do!
Good video sir ! Proper termination sure makes a difference :)
Nice. Wish people were throwing money at me so I could get a certifier, but that's not the case. Will just have to be more mindful of this. Can you try comparing the pass through ends with the standard crimp ends with the same untwist? The idea of bare copper hanging out the ends makes me really leery of using them, but they would make it easier to push your twists in further. Would like to see if that bare copper does indeed cause any problems or whether it's fine. Thanks.
Hello! You are in luck. Although not on video, I have extensively tested untwist for performance with passthrough and non-passthrough connectors. As it turns out, the biggest factor is fitment of the cable and insulated conductors to the plug itself. The TIA standard accounts for 1/2" untwist at each connector, so achieving tighter untwist at the connector does not get you much if any more performance and can actually cause issues in some edge cases. In other words, just achieve 1/2" or slightly less untwist and you will be fine. As to the difference between passthrough and non-passthrough connectors there is no performance difference **assuming** the cable to plug fitment is good in both cases. I will say that passthrough connectors are a bit more sensitive to ill-fitting insulated conductors, but that is more a feeling generated by experience. I have used passthrough plugs for many outdoor WiFi AP and camera installations with no ill effect, and those terminations are still going strong 5+ years later. It really all comes down to cable to plug fitment and meeting the 1/2" rule. Hope this helps!
@@trueCABLE Thank you.
Does the poor cable termination impact adjacent connections in the rack? Crosstalk, etc.
Good question Cash! No, that would only affect that single termination. Now, there could be an edge case scenario where two Cat6A terminations adjacent to each other, if they are BOTH poor terminations and BOTH terminations are being driven to 500 MHz, may impart ANEXT onto each other. In this case you would experience high NEXT on the individual terminations and ANEXT on top of it. That would be a weird scenario, but it is in the realm of possibility.
Nice video. As an engineer who has been making a server rack doing my own patch cables and Keystone Jack connections it has concerned me untwisting the wires. For the Plugs I am using your Standard RJ45 Connectors. Just looked at one and it seems under 1/2 inch and more like 1/4 inch or just over 1/4. I would like to see this test with keystone Jacks. Also what happens in the chain if I have 4 keystone jacks compared to if I have 5 or 6 keystone jacks? Is the problem additive? For me real world I ran what leaves the Rack to a patch panel with a keystone jack because it looked cleaner but that means one more connection. I would like to see a cable chain with varying number of connection combinations and test to see performance. Thanks for video
Hello Kim! You are definitely thinking correctly and have your head in the right place. First, I recommend against using solid copper Ethernet to create patch cords in the field unless you have a Fluke DSX-8000 around with PCA adapters to test each and every one. There are way too many things that can go wrong. Get matching Category patch cords that have been factory terminated and tested. Untwist at the termination is cumulative in performance degradation. Any electromagnetic disruption (a termination of any kind for example) will eventually add up to a poorly working cable run. Generally, the maximum amount of terminations per TIA 568 is going to be 6 per Channel. A Channel is from powered device to powered device. It is the full run, including patch cords. In this scenario, SIX terminations leaves you 2 x rack to jack style terminations (keystone to keystone or patch panel to keystone) and then two factory made patch cords. The quality of the terminations counts as well. Keystone or patch panel terminations are far better than hand terminated RJ45s. If you use hand terminated RJ45s you should not even exceed ONE in the Channel per the standard. The example would be keystone or patch panel to RJ45 to a PoE end point device and is known as a MPTL (modular plug terminated link). Now, it you want to use use male plugs and sort of "defeat" the TIA 568 standard in a way, then use field termination plugs. You are still limited to six, but field termination plugs perform just as well as keystone jacks! As for more testing and videos...that is on the radar!
@@trueCABLE That was helpful but left me with questions and ideas. I will check out your field termination plugs. I have thought for some cases of buying a factory cable and cutting off one end so that I have a good plug. Then putting on a Keystone jack on the other end I cut off. Still clarifying in my head the connection add up. Is the output of a switch to a cable plug 2 connections? Also a Cable plug to a Keystone jack would be considered 2 connections? I would be cool for someone like me trying to picture if you had a diagram showing what you explained as the six connections. Using your standard RJ45 connectors looks pretty fool proof. I am able to slide the yellow piece over the wires and get about 1/8 of an inch from the jacket, Trim it, insert it and crimp. Looks like a really short length. I guess this shows my lack of knowledge on the subject. Would I have access to the ANSI TIA 568 spec to read?
Nicely done!!!
Could you please tell me what Ethernet instrument are you using to test those cables?
Greetings!
Hello! Thanks for watching our video and your questions. The device used for testing is a Fluke DSX-8000.
@@trueCABLE Thank you, Sr.
@@trueCABLE I searched it, and Jesus Christ, it's a really expensive toy!
Very interesting! Thanks! BTW, your audio was not up to par.
Glad you liked the content, RP! We appreciate your feedback and are working to get the audio fixed for the future :)
Does exists a Windows software for testing a network speed and shows the number of packets dropped?
I know that the standard PC's NIC is maximum 1 Gbps, but I'm more interested/curious about the # of dropped packets.
There is free software called "The Dude"
Did the lapel mike fail?
Damn, you built a studio and it is crap sound? Rebuild, re-mic. It's ironic that tech guys, whom It trust, BTW, dropped the ball on the most important aspect of video: sound.
Say it with me: "There is NO SUCH THING as 'ethernet' cable!!"
Hello and thanks for your comment. Technically, quite correct. That said, copper twisted pair Category cable with eight conductors has become defacto "Ethernet cable".
@@trueCABLE - Swap "defacto" with "colloquial" and you've got a deal. 😜
Say it with me, pedantic pricks thrive on the internet.