Ascend Archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
(ASCEND) problems with connection to 962-1621
>Date: Mon, 23 Mar 1998 09:38:29 -0500
>To:Jamie@clark.net
>From:rweeks@pop3.clark.net (Robert Weeks)
>Subject:problems with connection to 962-1621
>
>I have just read Jack Rickard's paper on the 56k modem technology.
>http://www.boardwatch.com/mag/98/mar/bwm24.html
>
>In the tests performed by Rickard, he documented the improvement of using
>the long distance network vs local calling. I have noticed that I can
>make connections to Clark's Belair modems which require using the toll
>facilities, but any attempts to make a connection to the 962 number
>generally cannot be completed-either with the 56k Flex modem or with a
>14.4 modem. The 14.4 modem uses ATT (now Lucent) chips and the 56k unit
>is presumably Rockwell( Made by Global Village) Additionally connections
>from the former 919 area code (now 252) to the 962 numbers have been very
>relaliable, but only tested at the 14.4 rate.
>John Powell of USR suggested that the local telco circuits used a 3dB pad
>where the long distance calls used a 6dB pad. Obviously the problem with
>connecting is the a result of signal levels! Although my hearing is far
>from perfect, I was able to easily hear the signal level difference in the
>modem speaker when communicating with Baltimore and Belair. Powell's
>statement does confirm this difference.
>
>Paths through the phone co go through Ulaw and Alaw codecs and there in
>lies the problem that I am experiencing when calling form the 997 exchange
>to the 962 switch. It is unfortunate that Bell Atlantic cannot give a
>customer a concise technical answer and I had to work this out for my
>self, but I am satisfied that what I am seeing is just inherent with the
>workings of the phone company such as their use of -48 volts and 20 cycle
>ringing.
>
>However- The modems are supposed to be able to set their internal workings
>to accomodate the vageries of different line configurations and levels.
>At least that is what I thought. (I still do not know why the 997 to 995
>connection uses a path that is compatible with the modem pool on the
>995-0271 lines. My guess is that the 28.8 modems are more capable of
>dealing with the signal level differences than the Ascend Max 4000 modems
>with their current rev of S/W.)
>
>I have read that this is a problem with many ISP's and that I am not the
>only customer of Clark that is having this problem. There are work
>arounds for the ascend Max 4000 ( at least for dialing out) that force a
>resolution for the Alaw, Ulaw compression incompatibilities. Are there
>any that need to be applied to the local modem or is it only valid to
>correct the s/w at your end?
>
>I also do not have any assurances that the move to V.90 will fix any of
>these anomalies.
>
>Getting back to Rickard's paper, he reported that the K56flex modems were
>difficult to connect and he had to set his connect timeout to 60 seconds.
>Their connect rate was a miserable 80% connection rate. (At least thils
>was better than my 3 or so connects out of 25!. Rickard's conclusion was
>a problem with the Rockwell chip set!
>
>If there is any thilng that I can do to help resolve this problem pleaes
>contact me.
>
>Bob Weeks
>
>
>
>
>
>
>
>
>
>
++ Ascend Users Mailing List ++
To unsubscribe: send unsubscribe to ascend-users-request@bungi.com
To get FAQ'd: <http://www.nealis.net/ascend/faq>