sundance-communications.com/forum Forums Wiring - Telecom Carriers - Trunks & Services T1's, CSU's and DSU's T1 Crosstalk testing

hey guys i'm having some issues with a t1 that bounces for a 5 seconds then comes up we test to the niu tests fine no errors. so my next thought is there is t1 cross talk on the extended dmarc. yes the tx and rx are shareing the same riser cable. is there any test set that can test for this cross talk and give it a pass? if not what are some other options that people use to extend the dmarc from the mpo to the end users suite when there is only a single riser cable to this floor without having to run riser any type of extenders? any thoughts thanks in advance.

The only cases of x-talk I've seen on a T1 were due to slips. I've never experienced x-talk because the trans and rec were in the same jacket. What do you mean by bounces for 5 seconds and comes back up? What errors are you seeing in any provided PM data?

A couple of questions:

- How are you testing to the NIU?

- If the tx & rx are on the same cable are they on the same binder? If there are multiple binders in the riser cable, move half the ckt to another binder, that might help.

- Having said that, we would always prefer to run two shielded cables or one piece of General Cable "T-1 Cable" (each pair shielded and there's an overall braid) to feed T-1s on a floor.

- Have the carrier Stress test the circuit. We used to do 10 minutes of all 1s, 10 minutes of all 0s (if it's B8ZS), 10 minutes of 3-24 or 2-23 and then 1 hour of QRS. If there was a real problem we would go "head-to-head" (a test set on each end).

Sam

Lots of good advice so far.

Stress test to the CSU. When you loop the NIU, it loops on the front end of the card and doesn't engage the CODECs on the card. I've had an HDSL-4 (smartjack was Adtran H4-TUR) with one of the 2 HDSL2 loops totally dead, but we could loop the NIU and run clean pattern to it. If you loop the CSU, the NIU has to encode/decode the signal and often reveals trouble.

Recheck or reterminate all of your ends. Make sure you aren't putting 8P8C male ends onto cable that it wasn't designed for, terminate cables to jacks and use manufactured patch cables.

Make a loopback plug and put it at the end of the ended portion of the circuit further past the smartjack and use Sam's suggestion to run various test patterns with the tberd. If you do that while you have the loopback plug in it will also test your extended wiring instead of just to the smartjack.

I also saw a very slick box at nxtcomm a few months ago that was a passive monitor for flaky T1 circuits. It was able to be looked at remotely via ethernet, serial, or cellular modem. I can't think of the manufacturer but it should be googleable.