Tek-Tips is the largest IT community on the Internet today!

Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!

  • Congratulations Mike Lewis on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!

Interconnect vs Cross connect

Status
Not open for further replies.

chewbacca

Technical User
Dec 22, 2002
3
0
0
SG
Interconnect vs Cross connect
I am more familiar with Interconnect and was thus surprised to see a request for Cross connect.


Cross connect is more costly and uses more rack space.
SO why does some people still want a Cross connect system?
Any one can advise?
 
Well it depends what you are talking about. Voice? Data?

Routinely for data, we terminate the wiring on a patch panel, use patch cords to the switch. However for voice, we normally terminate the cable on 110 type blocks, and use cross connect jumpers to the 110's on the phone system.

So...what are you talking about?


It is only my opinion, based on my experience and education...I am always willing to learn, educate me!
Daron J. Wilson, RCDD
daron.wilson@lhmorris.com
 
Thanks Daron, glad to have a RCDD advising the issue.

THe scenario is for data.
I was told that in an Inter-connect system, outlets are terminated to a patchpanel. The patchpanel is then patched directly to a switch. Most cabling systems I came across uses this design.

I recently encountered what is I was told is a Cross connect design.
The outlets are similiarly terminated into a patchpanel (patchpanel 1). HOwever, so are the switches (patchpanel 2). Connnection is made by patching patchpanel 1 to patchpanel 2.

This effectively doubles the number of patchpanels needed. I recommended an Interconnect design but was told that the Cross connect is preferred as it increases the reliability of the system. I still do not understand why.

greatly appreaciate all advise.

MERRY CHRISTMAS TO ONE & ALL.
MAY YOU AND YOUR FAMILY ENJOY HAPPINESS, HEALTH & GOOD FORTUNE IN THE NEW YEAR.


 
You're right to be concerned Chewbacca. Throwing in the extra patch panels to connect to switches doesn't make any sense to me either. I also don't understand why the person stated it makes it more reliable. My impression is that it would be less reliable. Every connection added is another physical point of failure to worry about.
 
I'd have to lay this out on paper just to be sure, but I can't think of any way you can comply with the EIA/TIA standards for data cabling by having two patch panels like that. Unless there is some critical piece we are missing, I can see no advantage to doing your crossconnect option, and I can see many reasons for NOT doing that. You may want to ask your vendor for some EIA/TIA design standard that shows this. Also, most vendors can offer you a structured cabling system with a 15 or 20 year warranty. It's a bunch of hooey in my opinion, but we do it too, some customers like the sound of it. Anyway, ask if they do offer that, then check with the product manufacturer and see if they agree with the installation.

Basically any other connection point weakens the system, and as you start testing for compliance thru that weak point, you can often even see it on the test results.

Either they didn't accurately describe what they are offering, or you are likely being given the wrong information.

Hope that helps. Seems to me I remember in the early early days, there were some products (hubs) that terminated on something other than modular jacks. So you could pigtail them out and put a patch panel on the end of a 25 pair cable or something, that would be understandable, but that type of thing is long gone.

Good Luck! It is only my opinion, based on my experience and education...I am always willing to learn, educate me!
Daron J. Wilson, RCDD
daron.wilson@lhmorris.com
 
At one time HP had some switches that had 25 pair connectors instead of rj45's. You used cat5 25pair cable to a patch panelon the wall or in the rack.
 
There are some high density switches with amphenol-type connectors at each blade, which then plug into a harness connected to RJ45 patch panels (though I don't think you can get CAT5E by this method). Patching then becomes like a cross-connect field.

Some new designs I've been building provide for future VOIP: horizontal voice cabling is CAT6 or 5E terminated into patch panels, same as the data. Voice backbone feeder pairs also terminate at an additional bank of 48-port RJ45 patch panels on pins 4/5 (skip the 25th pair of each bundle), with patch cords dedicating the voice ports to specific feeder pairs. Mainframe 110 block cross-connects in the MDF hardwire the ports to the PBX, which actually functions as a programmable switch.
 
I have never seen this setup either. I'm currious as to what kind of cable they are using on the second patch panel and how this cable is being patched ito the switch? I don't see how stranded or solid cable can be used on this without violation EIA/TIA standards. Sounds like a out of spec install no matter what is said.
 

Thanks all for inputs.

I am not familiar with this section of the standards. How does this violate the standards? I would sure like to convince the user that this cross connect design does not meeet the standards.


NO special 25 pairs connectors are used in this case.
For the 2nd patchpanel, a stranded cord is used.
This cord has a RJ45 plug on one end and is open on the other.

The open end is terminated to the patchpanel using a punchdown tool while the RJ45 plug is connected to the switch like a patchcord.

Does this violates the standard?
Incidently, this type of cable (for cat 6 )is not available. I have to cut a patchcord into 2 halves to make 2 cords.

Thanks and regards

 
Well it violates in many ways. But first there are a couple technical issues. Cat6 patch cords are certainly available. If you are taking a Cat5e patchcord and cutting it in half, then trying to punch it down, you will likely have some problem. Patch cables are stranded wire, designed to have more flexibility. As such, they are limited in length by the standard because they are MUCH higher in attenuation (loss) than solid wire. While it is possible to use somewhat lengthy patch cables within the standards, the patchrack is quite likely designed with 110 style IDC connections which are for solid wire. The connector has two metal blades that slice through the insulation and make a connection with the solid wire of a specific gage.

For the standards, there are some very good overviews of the standards at several structured wiring manufacturers web pages. If you look at Leviton, they have a book (also available as PDF files) that is just an explanation of the standards and how it relates to horizontal wiring.

In short, the standard requires a permanent wiring link up to 90 meters in length, and allows for patch cables at both ends, with the total not exceeding 100 meters. There are provisions for breaking that permanent length with either a consolidation point (CP) or a MUTOA. Both of these are designed for in the field connections, and make no provision for adding a patch panel and cable in the telecommunications room (TR).

Sorry it is 6 am and pre-coffee, I'd have to look through the standards again and see if there is some other way this may be considered compliant. However, my first response is that it is a needless addition of patch cables and patch panels with no real benefit.

If you'll go look at Leviton, Panduit, Hubbell, AMP, etc. you can find some PDF files quickly that show how the system goes together. This should give you some basic direction.

Good Luck!

It is only my opinion, based on my experience and education...I am always willing to learn, educate me!
Daron J. Wilson, RCDD
daron.wilson@lhmorris.com
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor

Back
Top