Pure Bare Copper Wire

Discussion in 'Accessories' started by botics, Mar 13, 2019.

Share This Page

  1. botics

    botics Getting comfortable

    Joined:
    Feb 9, 2019
    Messages:
    133
    Likes Received:
    50
    Location:
    Coral Gables, Florida
    What is your take on Pure Bare Copper Wire Patch Cables attached from a patch panel to a PoE Switch?

    Good or Bad idea?

    Also, I have a female Cat6 Connector on my Cat6 runs to my cameras... What about connecting Pure Bare Copper Wire Patch Cables for flexibility to my cameras?

    Good or Bad idea?

    Thanks to all!
     
  2. mat200

    mat200 IPCT Contributor

    Joined:
    Jan 17, 2017
    Messages:
    4,150
    Likes Received:
    2,196
    Bare wires? NO, not a good idea imho.

    Remember, the twist in the cat5e/6 cable is there for a good purpose - to help cancel EMF which create issues on the wires and reduces data flow.
     
  3. botics

    botics Getting comfortable

    Joined:
    Feb 9, 2019
    Messages:
    133
    Likes Received:
    50
    Location:
    Coral Gables, Florida
  4. HelloAgain

    HelloAgain Young grasshopper

    Joined:
    Jan 26, 2019
    Messages:
    30
    Likes Received:
    19
    Location:
    USA
    When I first read this, I thought you were talking about truly bare/exposed copper cable... IE: basically trying to run your network over speaker wire. If you're asking if your network cables should be pure copper as opposed to aluminium (copper coated), then the answer is yes. Any cables that truly comply with CAT5e/Cat6 specs should already be this way.

    Now, as for the super thin cables that you provided a link for, that's a new one. I've used a lot of monoprice stuff and have always been pleased. It'd be worth the experiment, they aren't that much. My main concern would come with pushing PoE current over those wires for long distance runs or if they'd be more susceptible to cross-talk interference when in a large bundle. The PoE aspect is the most concerning.
     
    botics, J Sigmo and mat200 like this.
  5. mat200

    mat200 IPCT Contributor

    Joined:
    Jan 17, 2017
    Messages:
    4,150
    Likes Received:
    2,196
  6. J Sigmo

    J Sigmo Known around here

    Joined:
    Feb 5, 2018
    Messages:
    653
    Likes Received:
    616
    Good point. POE supplies power via the ethernet cable. You want solid copper, as opposed to copper-clad aluminum, and heavy gauge to carry the current. Thin cables will be smaller gauge, and have higher resistance, so there will be more voltage drop through the cable and more heating due to that voltage drop.

    The wire needs to be low resistance, and your connections also need to be low resistance for good POE performance and safety.

    I, too, was confused by the wording "bare". That usually means uninsulated, which would be unworkable.

    The insulation thickness and material sets its dielectric characteristics, and thus, the capacitance value per foot. And the twists per foot set the inductance (and to some degree, capacitance as well), and all of that that sets the impedance of the "transmission line". The cable impedance needs to match the impedance of the devices at both ends to eliminate reflections (keep the SWR - standing wave ratio as close to 1:1 as possible). This maximizes power transfer (for the data signal) and keeps those pesky reflections from causing data loss (think of it as interference).

    People need to realize that at the extremely high data rates we are getting out of these ethernet connections, this really is RF engineering. As someone coming from old-school radio work, it amazes me how well all of this works considering how easy the connectors are to install.

    But this also points out that you really do need to pay careful attention to the correct installation of the connectors, etc. You're making high frequency RF connections. And in the case of POE, you are also making power connections at the same time. It's pretty miraculous to someone who is used to being very careful with radio connections and cabling down into the sub megahertz range!
     
    Last edited: Mar 14, 2019
    looney2ns, mat200 and botics like this.
  7. botics

    botics Getting comfortable

    Joined:
    Feb 9, 2019
    Messages:
    133
    Likes Received:
    50
    Location:
    Coral Gables, Florida
    Just a thought... a bad one! These are patch panel cables.
     
  8. alastairstevenson

    alastairstevenson Staff Member

    Joined:
    Oct 28, 2014
    Messages:
    10,454
    Likes Received:
    3,196
    Location:
    Scotland
    Just to throw a slight disagreement into the discussion:

    An important aspect of a patch cable can be flexibility, especially when it's coming out of a patch panel for a short distance to a switch or device and has to bend around a bit, for example.
    If a patch cable isn't going to be tied or clipped down, the stiffness of a 24AWG CAT6 cable can be less than ideal for hooking up ports which are a metre or 2 apart.

    Despite the slightly higher resistance and slightly higher signal attenuation of either a thinner gauge, or stranded cable, a few metres of patch cable will have an insignificant effect on PoE voltage drops, or signal to noise ratios.

    In other words - sometimes the mechanical attributes of patch cabling outweigh the electrical attributes that are important for lengthy runs on the main cable.

    In my view.
     
    c hris527 and J Sigmo like this.
  9. mat200

    mat200 IPCT Contributor

    Joined:
    Jan 17, 2017
    Messages:
    4,150
    Likes Received:
    2,196
    Hi @alastairstevenson

    I've never had issues with 5e/6 patch ( which have stranded wires ) cables not being flexible enough for a 24 port switch to an adjacent patch panel.

    Thus, I see no need to skip on the quality of the cables to increase flexibility even for short patch cables.

    Also note, many users / consumers will go out and buy 100 foot long "cat6A" cables and wonder why they're having issues with PoE cameras at the end of the run...
     
  10. Whoaru99

    Whoaru99 Pulling my weight

    Joined:
    Dec 22, 2018
    Messages:
    423
    Likes Received:
    159
    Location:
    Here
    I use a couple of those thin cables at the office to connect the VoiP phone and computer docking station to the network. Dont recall... maybe 3m from wall jack to VoIP phone and 1m from VoIP phone to computer dock. Seems to work just fine for that, but I don't know how much draw the VoIP phone presents.

    Do share concern in cases that draw material amount of power, particularly if long(er) as length further compounds any issue with thin wire, current, and voltage drop.
     
  11. alastairstevenson

    alastairstevenson Staff Member

    Joined:
    Oct 28, 2014
    Messages:
    10,454
    Likes Received:
    3,196
    Location:
    Scotland
    I did say stranded is fine - my point was that patch cables don't need to be solid copper of thick gauge like the main run if it makes it difficult to route that last small leg.
     
    J Sigmo likes this.
  12. awsum140

    awsum140 Known around here

    Joined:
    Nov 14, 2017
    Messages:
    1,283
    Likes Received:
    1,113
    Location:
    Southern NJ
    Personally, I'd stick to 22 or 23 gauge, stranded, if it's really a tight, crowded, installation. Introducing a different resistance in the mix isn't "good practice" in my mind. I've done some major installs, as in over 100 connections per network closet, and never had a problem using full sized cable. It's all in how you patch it all together.
     
    KT88s, J Sigmo and mat200 like this.
  13. mat200

    mat200 IPCT Contributor

    Joined:
    Jan 17, 2017
    Messages:
    4,150
    Likes Received:
    2,196
    KT88s likes this.
  14. J Sigmo

    J Sigmo Known around here

    Joined:
    Feb 5, 2018
    Messages:
    653
    Likes Received:
    616
    This is a good example of where terminology can get us all confused.

    In this discussion, the word "solid" is being used for two meanings, and that's confusing.

    In some places, we're using "solid copper" to mean "pure", or "not plated", as in: "Not copper clad aluminim".

    Then, in other places, we're thinking of "solid" as meaning "not stranded".

    I don't think there is any problem with good quality pure-copper stranded wire. As long as the connectors are the correct type for stranded wire, and installed properly, and the stranded wire is of sufficient gauge, there's no reason you couldn't use it even for long POE cable runs.

    And as you point out above, short patch cables won't have too much effect on the total resistance of a run even if they're made of lighter-gauge cable.

    And further, as you point out, it is good to keep mechanical stress off of the connectors, so a more flexible patch cable may well improve the reliability of a system if it reduces the stresses on the connections.

    Just be sure, for POE systems, that the patch cables are rated for the POE current of the circuit. Some of the thin, super-flexible patch cables will overheat, especially right at the connectors, when used with high-power POE loads.

    I'm sure you understand this all, but for the sake of others stumbling onto this thread, I'm just trying to clarify some of this.

    A fine point of any wiring is the difference between a voltage drop issue and an overheating issue. Both of these issues need to be carefully considered.

    You can have a system where the wire and connections are all safe from an overheating perspective, yet create too much voltage drop, and end up causing unreliable operation. This can happen when the cable run is too long even though the gauge is large enough to not overheat.

    You can also have a situation where the voltage drop is acceptable, yet the wire or connections will overheat. This can happen when the wire gauge is too light, or the connections are poor, yet the run is short enough that the voltage drops in these connections and short cables are not enough to adversely affect the voltage at the load.

    And, of course, you can have all sorts of combinations of these issues.

    People need to consider all of these points and make sure that every connection and cable in their systems is safe at the current it will be required to carry. And they must also consider the total resistance of each complete circuit to be sure they don't end up with too much voltage loss.
     
    Last edited: Mar 16, 2019
    TonyR, KT88s and alastairstevenson like this.
  15. IAmATeaf

    IAmATeaf Pulling my weight

    Joined:
    Jan 13, 2019
    Messages:
    283
    Likes Received:
    159
    Location:
    United Kingdom
    I didn’t know that the connecters were different between stranded and solid copper, never really gave it a thought but now that it’s been mentioned above it makes sense.
     
    KT88s likes this.
  16. awsum140

    awsum140 Known around here

    Joined:
    Nov 14, 2017
    Messages:
    1,283
    Likes Received:
    1,113
    Location:
    Southern NJ
    My objection to changing gauges comes from the "old days" of Thick and Thin Net, basically high end RG-8, thick, RG-58(thin). We always took a TDR trace of every installation and you could see the connectors, "T"s ( thin net), taps (on thick net) and terminators quite plainly. Later, when twisted pair came along, we also took a TDR trace of each drop. Even a kink or tight radius bend stood out on TP. Changing gauges will introduce an impedance mismatch that on a TDR would look like a mountain. We are talking signals in the RF spectrum and impedance, and resulting standing wave, can be a factor to consider, even though the signal is actually digitally based. Anyhow, I just feel it's not a "best practice".
     
    KT88s and looney2ns like this.
  17. J Sigmo

    J Sigmo Known around here

    Joined:
    Feb 5, 2018
    Messages:
    653
    Likes Received:
    616
    Honestly, from an RF perspective, I'm amazed that any of this Ethernet stuff works at all!

    From a background of carefully terminating Heliax using connectors that cost, in the '70s, upwards of 125 bucks each, and cutting them off, and throwing them away if they didn't pass testing, it blows my mind to think that gigahertz signals successfully pass through 50 cent crimp-on insulation displacement connectors plugged into crappy RJ receptacles, etc.!!!

    And yet, it seems to work!

    Edit to add:

    I think the relative ease and low cost of these connections hides the actual complexity of what's happening in these systems. We take it for granted and don't consider how amazing (and critical) these interconnects really are.
     
    awsum140 likes this.
  18. awsum140

    awsum140 Known around here

    Joined:
    Nov 14, 2017
    Messages:
    1,283
    Likes Received:
    1,113
    Location:
    Southern NJ
    I've done my share of coax and heliax connectors as well. I've always wondered about how the impedance manages to stay relatively consistent through an RJ45. Boggles the mind, but it does, somehow. Maybe the gold plating on them helps a bit in terms of actual conductivity as well as at 100MHz. Beats the heck out of me.
     
  19. tangent

    tangent IPCT Contributor

    Joined:
    May 12, 2016
    Messages:
    3,832
    Likes Received:
    2,479
    Easy mistake to make, but 1000Base-T Ethernet doesn't use GHz signals. The data is split over multiple pairs and gigabit uses a more complex signaling.
    Cable frequency ratings (not the actual signal frequencies):
    Cat-5: 100MHz
    Cat-5e: 100MHz
    Cat-6: 250MHz
    Cat-6a: 500MHz
    Cat-7: 600MHz only shielded cables
    Cat-7a 1GHz only shielded cables

    It is pretty impressive what you can do with differential signaling over twisted pairs of wire. For UTP cable to work correctly the twists in the cable to have to be fairly particular and there's a lot of cable sold that doesn't meet the standards. As you get into the higher speed specifications actually getting everything to work gets at full speed is harder. Part of what happens is the link auto negotiates a slower speed and some people don't even notice. I read recently that compared to fiber optic, using copper for 10Gig and 40Gig ethernet uses as somewhere around 10W more electricity per connection because of all the filtering and signal processing required to get a viable signal.

    As to how twisted pairs work, here's some reading material:
    https://www.highfrequencyelectronics.com/Nov02/HFE1102_Lao.pdf
    https://m.eet.com/media/1141050/19611-pdf_only.pdf
    http://www.ece.tamu.edu/~spalermo/ecen689/lecture3_ee689_tlines.pdf
    http://www.w0qe.com/Papers/TransmissionLinesPresentation2.pdf
     
    mat200, J Sigmo, awsum140 and 2 others like this.
  20. KT88s

    KT88s Getting comfortable

    Joined:
    Apr 11, 2019
    Messages:
    29
    Likes Received:
    19
    Location:
    Here
    Just saw this thread, thought a quick read through before bed for this aging engineer might do me well.
    THAT WAS TWO HOURS AGO!
    After these Links posted i need a drink.
    I do feel good with the fact i can follow every reply here, and know what is being stated. cool.
     
    tangent and TonyR like this.