>I recently purchased a US Robotics Sportster Voice Internal 33.6
>Faxmodem with Voice Personal Mail. (model 1171). I bought this
>product with the specific intention of setting up a voice-menuing
>system to be used with 3 distinctive ring patterns on a single
>This is a long post, so I'll give the bottom line first, and then
>follow with all the gory details. The Sportster Voice distinctive
>ringing feature is not compatible with Northern Telecom's DMS100
>switch, so if your telephone company uses this type of switch
>equipment you are out of luck. NYNEX tells me that the Northern
>Telecom DMS100 switch is one of two types of switch equipment commonly
>used (on up to half of the circuits in their territory). In two
>areas that I have checked, Port Jefferson, LI, New York, and
>Brookline, Mass. Nynex uses the DMS100, although they use an AT&T
>switch in other areas of the same cities. I don't know if the
>Sportster Voice works well with the AT&T switch, I have been unable to
The Switch is not the problem. The problem is a matter of
adhering to Cadences. The modem answers the appropriate RING
when it see's the ring. It needs to receive a certain RING
pattern in order to answer the call. Below are the ring
patterns. If the Switch and Software does not follow these
pattern the modem will not answer.
Type A: 2.0s on, 4s off
Type B: .8s on, .4s off, .8s on
| | | |
| | | |
------+ +----+ +----------------------
Type C: .4s on, .2s off, .4s on, .2s off, .8s on
+--+ +--+ +------+
| | | | | |
| | | | | |
------+ +--+ +--+ +----------------------
Type D: .3s on, .2s off, 1.0s on, .2s off, .3s on
+-+ +--------+ +-+
| | | | | |
| | | | | |
------+ +--+ +--+ +----------------------
>After waiting about 20 minutes on toll-hold, a technician guided me
>through the setup process. It turns out that the PNP features of the
>1171 will set the modem up to COM3, IRQ7, if it detects COM1 and COM2
>as being used, and if IRQ5 is in use, as it is on my system. Does
>your LPT port use IRQ7 also? Too bad, the PNP features of the 1171
>will still use IRQ7, and you will have a problem that will not be
>reported. Win95 Control Panel, System, Device Manager, did not report
>the problem, and reported "no conflict" under each device inlcuding
>com1, com2, lpt1, and the modem on com3. Solution: Give up the
>com2 port, since I need the LPT port.
>HINT for USR: PNP is very nice, but more choice in IRQ's is even
>nicer. Why do you limit the 1171 to IRQ's 2,3,4,5 & 7? Why don't
>you offer 9,10,11, & 12 as additional options? There are lots of us
>that need 2 com ports plus the modem and have sound cards and lpt
The modem is a 8 bit card, so it will only use IRQ 0-7. It
would have to be a 16 bit card to accomplish this. If your
Sound Card is a 16 bit card, then assign it a higher IRQ.
>I decided to plunge into setting up the distinctive ringing features
>without waiting for the new RCV or User's Guide. It's just as well,
>since there is not one word in the User's Guide about DR. There is
>also nothing in the USR hardware book "Installation and
>Troubleshooting" about DR. Time to call USR Tech support again.
>Toll-hold once more for about 20 minutes, to be told that I must edit
>the hardware.ini file for RCV, to set DistRingSupported=TRUE, changing
>from the default FALSE setting. You may then enter RCV modem setup
>and verify the settings of the S41 register on/off and the RING A, B,
>C,D reporting strings. (Use "default" settings to autodetect.)
>Sure, this is easy to do, once you are told how, but it is completely
>undocumented. Don't be misled by information at
>There is discussion there of S41 register, but it has nothing to do
>with DR (???).
>HINT for USR: If you have time enough to design a new feature, and
>print new boxes that proudly proclaim "Caller ID and Distinctive
>Ring" then make the time to print an addendum to your manual showing
>the S registers and necessary editing of setup files.
>Now comes the fun part. I wanted to set up DR for a typical home
>office situation. DR uses different ring patterns or cadences, which
>I will call A, B, C, D, following the USR terminology.
Following terminology most of the industry uses. <g>
>RingA is one
>very long ring, Ring B is two long rings, Ring C is 3 rings, short,
>short, long. Ring D is 3 rings too, short, long, short. I wanted
>RCV to ignore RingA, the line for personal use and/or answering
>machine independent of the computer. I wanted RingB to activate the
>voice mailbox features of RCV, and I wanted RingD to activate the
>fax/modem features only of RCV. (Note: In many areas of the country,
>you may subscribe to 3 DR patterns, and you will be assigned RingA,
>RingB and one of RingC or D. I am not aware of any telcos that offer
>four DR patterns, which could use A, B, C, D, but they may be out
Ring D is very difficult to implement. Not many applications
or modems support this correctly (With the exception of our
Courier). The Sportster does not implement this RING type
and see's it as a RING B because it uses the same amount of
pulses and in a way close to it.
>In any event, you normally do not have any say in what
>patterns your telco gives you, and A,B, D seems to becoming more
I do not know if your TELCO WOULD implement a RING C instead
of a RING D. Considering it is probably controlled by
software, it should be easy to change. I also want to know
how you have found out RING A, B, and D are most often used?
Is this only NYNEX or is this with Ameritech, Bell South,
>After much experimentation, I determined that when I called the line
>connected to RingD, the modem was answering with the setting I had
>chose for RingB. (Ring B also activated the RingB setting.)
>Another call to USR tech support was made. After another long
>toll-hold, I was kicked upstairs to "Level 2 Support Operations" and
>given Call Reference number 1411612. Troubleshooting involved using
>the Terminal mode of RCV, and watching the modem readout while the USR
>technician called back on each of my 3 DR numbers. Sure enough, the
>first 2 worked just fine, and the third pattern (D pattern) failed.
>Instead of reading out RingD on the RCV Terminal mode screen, it
>alternated RingB, RingA, RingB, RingA, on different lines. Thus, the
>fault is within the modem hardware and the RCV software is just doing
>its job and answering in B mode, when the modem reports "Ring B" to
Again, RING D is not supported in the Sportster product. The
Courier does support RING D.
>Followup troubleshooting with Nynex Repair service identified the
>specific problem. Briefly, USR defines a "very long ring" as a 2
>second burst of ringing voltage, a "long ring" as 800 msec, and a
>"short ring" as 200 msec. The intergroup interval is 4 seconds.
>Details are at http://ae.pcd.usr.com/techref/dring.htm
See above patterns.
>USR tells me that the pattern at this www page follows the BellCore
>standard. NYNEX tells me that in both Long Island and Boston areas,
>the lines that I use are served by a Northern Telecom DMS100 switch.
>Unfortunately, the DR patterns used by the DMS100 are different from
>the ones USR uses. A short ring on the DMS100 is 500 msec of ringing
>voltage followed by 500 msec silence. A long ring on the DMS100 is 1
>second on, followed by 500 msec off. The intergroup interval is 3 or
>3.5 seconds. From these specs, and the behavior of the modem, it is
>clear that the 500 msec short burst of the DMS100 is detected by the
>USR 1171 as a long ring. Presumably, these patterns are coded into
>firmware, but in any event they are not adjustable by the user.
AH HAH.... Again the problem is based on Cadences which are
programmed into the modem. These Cadences are based on the
top 3 used. RING A, RING B and RING C. Also there can be a
little varation in the RING patterns. To my knowledge, the
modem isn't only looking for a timing sequence, it is also
looking for a specific amount of pulses in a specified
>I called USR and reported this information in great detail to Level 2
>Support Operations, as a followup to report 1411612. The technician
>indicated that I had no option other than to return the modem due to
>incompatability. He did not seem to take my point that many other USR
>customers will face the same problem.
>NYNEX tells me that the Northern Telecom DMS100 is a very commonly
>used switch, one of only 2 types used in almost the entirety of the
>HINT for USR: You really should have another look at the distinctive
>ringing specs. If you had done your homework on this one, you could
>easily have made the 1171 work with both types of switches.
Nothing to do with the switch, it is how it is programmed.
Also keep in mind the Sportster implements the 3 most common
Online Support Analyst