sundance-communications.com/forum Forums Wiring - Telecom Carriers - Trunks & Services T1's, CSU's and DSU's ALL 1's, ALL 0's, and 1 in 8 Failures

Hey Folks, Hoping to get some insight into this problem I had with an Adtran HDSL circuit. The sucker has got a new doubler, new smartjack, and no apparent cable issues. Not even any bridge tap (I NC'ed it to cable and the cable tech found it had already been cut off). Anywho.....it always passes QRSS yet it will fail 1 in 8, all0's and all1's. BUT it seems to pass one of these 3 patterns, then fails the next intermittently.I gave up and patched to the spare. I'm new to specials (4 months) but I've never seen this before....any ideas?

All 1's will stress-test a repeater. So it could be bad if you get failures with it.

QRSS isn't a very stressful pattern like the other ones are, so seeing no errors when testing with it isn't unusual.

Failing All 0's and/or 2^20 could indicate something is optioned AMI rather than B8ZS.

like I said, new doubler, new smartjack. Could it be the office repeater, even though it test ok?

You said you patched to a spare. Is it good on the patch? If so something in your original circuit is bad.

Like Dex said failing all 1's indicates a bad repeater. The only thing all zero's should be used for is confirming B8ZS. Change the repeater and test again.

I don't know what the 1:8 pattern is for.

QRSS was used to simulate real T-1 traffic and was low stress as Dex pointed out.

Where's it failing at...r-card, doubler, or c-card? Not sure how your circuit is designed but get a tester, or if you have test access, start by looping the c-card from the dacs or closest dsx, and stress test the circuit with pattern T1 Daily or 55 Octet. If it passes, loop the doubler and if that passes go for the r-card. This should give you an idea about what section of the span is bad.

Also use the PM data. The Adtran cards pretty much record everything and if you can decipher the info it will give you a clue as to where to look. If it is a model that has the Tscan feature it should have a bad slice detector under the troubleshooting menu... it will give you the distance to the bad splice. Sometimes it is right on, other times you need to give or take a few hundred feet.

We used QRSS to stress test the PHYSICAL portion of the job (jacks, cable connections, patch cords, etc.). All '1's verifies that the circuit will pass AMI and as Dex and Bill said should stress the repeater and all '0's is used to prove B8ZS.

1 in 8, 3 in 24, 2-20 are all miscellaneous stress tests. It sounds to me like the circuit has some of it's repeaters or circuit boards wired wrong. Maybe some parts wired for AMI and others for B8ZS, maybe the LBO is wrong.

You're going to have to break the circuit down into pieces and check each part. Use standard troubleshooting techniques- go to the midpoint of the circuit and test back to the source. If that works/fails then keep going to the next midpoint till you find the bad segment.

ANY circuit should pass all '1's. If it won't pass that don't bother with anything else till you can get that part working.

Sam

I think that the "all 1's stresses a repeater" rule only applies to a real T1 span using 239 style repeaters (which were essentially analog dc op-amps). The OP said it is provisioned over HDSL, which is completely different - all 1's produces a signal on the span that is really no different electrically than any other data stream.

Just my 2 cents... [sorry, haven't run my jaw in a while]

Jim
**************************************************
From a bunker in the hills of Pennsylvania

What is a doubler?

Doubler-Hdsl Repeater

When swapping out parts were you able to switch shelf/slot positions in the C.O; repeater housing and cust-prem housing? Backplanes and breakout wiring can do be failing too.