Conference 7.286::fddi

1608.0. "SMF db loss issues ?" by 45279::BARKER_E (test dummy) Fri Mar 03 1995 07:41

	Hi,
    
    	I'm x-posting this from the hub notes file as it seems to be
    widening out into a general FDDI issue. If anyone can advise I'd be
    very grateful !
    
    	Regards,
    
    		Euan Barker
    
    
            <<< NETCAD::KALI$USER3:[NOTES$LIBRARY]HUB_MGNT.NOTE;1 >>>
                   -< DEChub/HUBwatch/PROBEwatch CONFERENCE >-
================================================================================
Note 2066.0               SMF MOD PMD (DEFXS) problems                 3 replies
BRIEIS::BARKER_E "test dummy"                        21 lines   2-MAR-1995 15:36
--------------------------------------------------------------------------------
    Hi,
    
    I'm looking for some advice. A customer has ordered a DEF6X-MA
    (DECconcentrator 900MX with no PMD's) and then ordered a number of MMF
    PMD's and one SMF PMD to be fitted. I turned up on site to configure an
    FDDI ring , got presented with this so had to go ahead and fitted
    them...
    and the SMF module didn't operate correctly. I was incredibly careful
    with the fibre loops etc, honest ! We tried another DEFXS-AA today and
    that gives the same problems. Is there any testing that can be done on
    the concentrator and DEFXS-AA or is the next stage to a/ try DEFXS in
    another concentrator and b/ swap the concentrator ?  Seriously, I
    didn't exceed any bend radius' not already in the packaging and tried
    to ease out the cable to as large a radius as possible, one problem was
    that the way they are packed the fibre seems to get a memory for being
    bent one way in the packaging and although we had two free slots next
    to the SMF PMD this still made life awkward.
    
    	Thanks for any pointers and advice,
    
    		Euan
================================================================================
Note 2066.1               SMF MOD PMD (DEFXS) problems                    1 of 3
NETCAD::B_CRONIN                                      7 lines   2-MAR-1995 16:49
--------------------------------------------------------------------------------
    
    Can you connect a single mode fiber as a loopback connection on 
    that one SMF port? If the module is good, the loopback connection should 
    result in a blinking yellow light, indicating that the connection is 
    being rejected (because of a topology reject). That will tell you if
    the module is dead. 
    
================================================================================
Note 2066.2               SMF MOD PMD (DEFXS) problems                    2 of 3
BRIEIS::BARKER_E "test dummy"                        17 lines   3-MAR-1995 06:32
--------------------------------------------------------------------------------
    Hi,
    
    	Thanks for the reply, I can try that but I will have an issue with
    the minimum length for the loopback due to the inline loss on a short
    cable, the customer has no inline loss elements as far as I'm aware.
    I'll patch around the full length of the ring on one core which will
    total 22km.
    
    	I have a suspicion the customer may have tried the above with 10m
    SMF patch cables directly between two concentrators. What would this
    do, destroy the DEFXS MOD PMD ?
    	
    	Thanks,
    
    		Euan
    
    
================================================================================
Note 2066.3               SMF MOD PMD (DEFXS) problems                    3 of 3
BRIEIS::BARKER_E "test dummy"                        21 lines   3-MAR-1995 07:39
                    -< Could be a general db loss issue ? >-
--------------------------------------------------------------------------------
    Hi,
    
    	MMmmm, this is leading somewhere, the loss adjuster issue in .2 got
    me thinking, phoned the customer and guess what, the db loss on the
    circuits is 4.2 on one (plus whatever 2 patch cables introduce) and 6.8
    on the other (plus whatever 5 patches inroduce).  The 6.8 plus 5
    patches route is the one that is working, the 4.2 one isn't.  What is
    likely to happen to the DEFXS-AA in this situation ?
    
    	This could get sticky, we (as a project team) specified the ring
    must be available and operative before we got on site, and once there
    assumed it matched our required spec. It is running in
    parallel with a Cisco network on the same fibres and having talked to
    the Network Manager they didn't check the db loss we require. It is
    about to be tested. Anyone know of the Cisco kit would specify a db
    loss range different to ours of 12-22 db or is 12-22 a standard spec. ?
    
    	Thanks again,
    
    
    	Euan