NI2 PRI

[cilcomm]

I have two NI2 PRI for the DID/DOD coming into an Option 11C 25.40 switch. The second PRI ( ch 25-47 with secodary D ch) seems to be having tons of error messages, but mainly on the weekends. The first PRI has had no LOSFA or Slipped frames ever. We have had Verizon do a stree test on it and perform hadr loop tests to see if there is a problem with it but found nothing. After their tests last week, I had no trouble and no errors until today(saturday). The same thing happened last time we had tests run on this span. I have checked the programming on both spans and checked them against each other. I have replaced the extenal Adtran CSU, cabling and TMDI card. Even moved the TMDI to another slot and reprogrammed it but the same thing keeps happening. Has anyone else had this problem or anything likeit?, if so did you find a way to correct it or at least locate the problem?
I have also reset ans INI'ed the system to eliminate a
'ghost in the system'.
I'm at my wits end with this. The errors and slipped frames are causing the TMDI to go into disable and Ihave to manually rest the card and dchannel.

 

[acewarlock]

If this was my switch, I would swap the RJ45 plugs at the smart jack after hours and see what happens,

 

[cilcomm]

srmega41,
The span comes in on seperate pairs form the CO and I dont think they run through any external MUX before entering the building. The DS1 card in the Smart jack even has an B8Zs led on it and it is lit, Would this be lit if it was an AMI?

acewarlock,
I have replaced the RJ-45s at the samrt jack and after the CSU.

Johnpoole,
The smartjack is 3 feet from the switch. When I requested the PRI's for this site, I asked that the 1st PRI have the primary DCH and the 2nd PRI have the backup dch. The 2nd PRi is the one I am having trouble with. Could the problem be that they set the PRI's up with both PRIs using their own DCH ad the Primary DCH. in other words, the 2nd PRI is using the 2nd DCH as its primary DCh instead of the1st PRI dch?

Confused yet?.... me too!

 

[acewarlock]

I didn't mean replace thr RJ45, I meant just swap one end of the RJ45 just at the smart jack so PRI 1 is now PRI 2 and PRI 2 is now PRI 1, If the problem follows the swap then the Telco has the problem.

 

[johnpoole]

no it's usually better to run nfas, a d per pri. it adds to your redudant ability. the span is b8zs or it's not ever going to come up. if your more then a mile from the provider, then it's not hard copper all the way to the co. if you are able to access the cable head, here's a test i would run. what type of protection is on the entry cable? usually carbon modules that plug into a green or white field. i would open those pair at the telco demark, then measure the voltage of each conductor to ground. each side of a pair HAVE to read the same voltage. you may find one of then has no voltage. the reason it must be open between the cable and the smart jack is that in the smart jack each pair is connected with a tip and ring short. we use a cable pair as a single conductor when it gets to our smart jack. you need to see about 700 volts on each side. a pri has been known to work 3 conductor. it usually has sparatic problems. that is a last resort test, and if your not comfortable with it. request a vendor meet. a user is not by law allowed to alter connections before the demark. another thing i would have done way before now is request a rebuild. new pri, co installed, new pair, new smart jack,new ckt number. if your that sure the problem is not yours, make the provider PROVE that it is..

john poole
bellsouth business
columbia,sc

 

[cilcomm]

Acewarlock,
If I swap them wont the Primary DCh be on the wrong circuit? If the Primary Dch is now on the 2nd PRI wonthe dch messages be coming in on the wrong DCH to the switch?
I was thinking that all the dch messages for both PRI's come in on the Primary DCH and not the backup DCH. Therefore the switch is looking for DCH messages on the 1st PRI and not the 2nd PRI.

Johnpoole,
I will check that voltage next time I visit the site. The site is about 3/4 of a mile from the CO and the demarc in the bulding is only 10 feet away form the smart jacks and the smart jacks are only 3 feet from the switch. The only thing between the smartjacks are the Adtran CSU ACE's. I have swapped the Adtrans before when I was in the process of swapping out all the other components in the chain on at a time to see if anything made a difference. The only thing that seemed to make a difference was swapping the DS1 cards in the smart jack housing. I swapped cards with an adjacent circuit i the same samrtjack housing from another circuit that is only used once year for some video conferencing. After I swapped the cards, I had no errors for 2 weeks and the PRI never went down or lost its DCH. Then the errors started up again and the span goes down about every 2 or 3 days unless I RCNT in LD 60 before the threshold is met.
I will try to start the Round Robin test request today but it will probably take a couple of weeks before I get to talk to the right agencies to make this happen. My local CO is usually happy to help me in any way to perform tests but if they have to get any other Verizon offices in on the tests likehtier office in Florida that does translations, then they usaully have to open a ticket and if nothing is found to be a probelm on their end they need to bill that time off tho the customer and without a work order for the proper agencies, they cant get the billing done and the Co techs that help me out get in trouble for it.

anyway, thanks for the ideas. I will put them to use ASAP

 

[srmega41]

Been out of the loop all day, so I am somewhat late responding. If the signalling coming into the building is B8ZS, the light would be lit, if it was AMI it would not. If I remember correctly, the optioning can be set at multiple points (CO, fiber hut, external MUX), and that is why this can be such a hard issue to diagnose. It sounds like this is a rather odd one and it will be interesting to hear the results.

Another thought I had is it might be a cable issue at the physical layer. Have you considered asking for the LEC to re-terminate the span on new pairs? I know you said the CO is close, it is possible a bad copper connection could be an issue.

Just a thought,

Scott M.

 

[johnpoole]

if your less then 5000 cable feet you could be on pure copper. a known problem is a demark that is 4600 to 5400 ft, your either to close to balance with treatment or to close to the treatment. the voltage is a long shot, but without moving the switch 2000 feet closer your looking for a unique problem. i use standard cat 5 to the db 15, i don't care for nortel provided cable for that, they work but the double shield can cause a reflected clock, according to a post on the nortel alerts right after the pri was supported by their switch. they may have addressed that problem but for me the best place to use them on a switch i don't maintain.. this one had a mix of those, and hand made patches when i started here, now they are all factory cat 5 patch and all match, and are marked on each end...

john poole
bellsouth business
columbia,sc

 

[cilcomm]

Johnpoole,
I measured the voltages like you said at the demarc on each side of the PRI legs to ground.
On the first PRI that I have no trouble with, I measured 185Vdc to Ground on the Tip side and 5Vdc to ground on the Ring side.

On the PRI that I have trouble with, I measured 8Vdc on the Tip side and 5Vdc to ground on the Ring side.

I double and triple checked these measurements to make sure I wasn't seeing things.

Both of these PRIs are coming in as 2wire PRI DS1/B8ZS
they come into the building on the same 100 pair cable but do not reisde on adjacent cable pairs.

Any other ideas?

I have not been able to set up a vendor meet yet due to the government paperwork between me and Verizon but I am still working on it!

 

[johnpoole]

? i've never worked with single pair pri. usually two pair reaching a common point, the second pri is so close to dead, how can the co see a loop back?

john poole
bellsouth business
columbia,sc

 

[srmega41]

Sounds like the span is coming in using DSL technology. We have a number of these at locations, but I would think you need some device to go from 2 wire to 4 wire. The voltage issue seems to be very obvious, so I would continue to push that direction.

Any luck swapping the two spans? I know the D-Channel needs to see the other end, but I am not sure if the Backup D-Channel will complain if it is connecting to the primary from the CO's end. I think the config is made to support everything is the primary goes away, so the backup is a duplicate of the primary, just on a different span.

More food for thought.

Scott M.

 

[cilcomm]

srmega41,
are you famialiar with the voltages that I should be seeing on this type of PRi delivery? if so, what should I be seeing?
I think most of the staff is out for a couple of weeks starting next week. I will see if I can do the channel swap during that time.

Another curious thing I have noticed is that if I go to the smart jack ( this one is a 4 channel housing containin my other main PRI and 2 other PRI all on B8ZS) and remove the PRI card from the housing and reseat it, the error problems go away for about a week or more, then it all starts to cascade in slips, errors and the Dch on the backup PRi dropping in and out.

I will try to get an expllanation from the carrier about the voltages and the other things you have all mentioned before concenring hte round robin tests, etc....

thanks

 

[srmega41]

I cannot say exactly what voltages you should be seeing, but considering the smartjack setup you have, they should be consistent across all four spans. A good test would be to see if you can take all of them out of service, one at a time, and test each one. I would not be surprised if you see consistent voltage readings on the three that are working.

Scott M.

 

[johnpoole]

scotts makes a good point, your voltage is below what i would expect locally, but things can be locally unique. but they should match

john poole
bellsouth business
columbia,sc

 

[johnpoole]

i also check each pair for any loading. if you have an analog meter

john poole
bellsouth business
columbia,sc

 

[cilcomm]

What type of loading? do you mean loading coils within 1 mile of the CO? With voltage on the line, I cant check resistance or impedance. Is there something else I should check?

 

[johnpoole]

if you take a voltage reading, then flip the leads you'll see a cap kick on a loaded line, used to use the polarity switch on a simpson meter for that on a line with voltage. with the voltage on normal the needle stays steady, flip the poliarty, meter starts to peg, but half way to 0 volts, it kicks from the discharge. load coils are stil used but for a pri ckt SHOULD have been cut around, i've seen ckts turned up with loads still in place. a good tested will see a major db drop between 1000 cycle and 300 cycle return. back in the dark ages sending and rec 1000 cycle at 0 db was the only test we used. by sending a set level, from several freq's a tester could tell if a pair would end up to hot or cold at certian levels.. more then likely, your cable plant is newer then that. unless your rural, or working out of a distand office. 10 years back in this area, the remotes, usually less then a mile out, worked great, but pri's had to work out of the main, with repeaters. up to 10 miles away. most of our newer remotes now have software to handle pri's... i have ordered a new pri, just so i could turn off the "one" they couldn't fix. because of my postion now i can make that happen without billing the customer... in house tech for the largest hosp in the state, plus working for the ld, dial tone, and pbx provider helps

john poole
bellsouth business
columbia,sc