larry@uunet.uu.net (Larry Lippman) (04/08/90)
In article <5991@accuvax.nwu.edu> MAP@lcs.mit.edu (Michael A. Patton) writes: > Loop Start provides no way to interlock against glare, > Ground Start interlocks intrinsically in the protocol for picking up > the trunk. However, glare cannot be totally eliminated, even with ground start trunks. A trunk for an outgoing call is selected by an idle condition of open on the tip side of the line. Depending upon the type of PABX, there could well be a one second or more "window of opportunity" for an incoming call to arrive on that very same trunk before the PABX actually *seizes* the selected trunk by placing a ground on its ring side. To quote Murphy's Law of Telephone Anomalies, "If it can happen, it will happen, given enough traffic." :-) > On the other hand Ground Start > trunks can be configured (I don't remember if it's the default) to > return positive supervision so accurate billing can be done. The only way that a ground start trunk can signal any type of supervision is by means of T/R battery reversal. Traditionally, T/R battery reversal was provided on ground start PABX trunks as a tariffed option *solely* for toll diversion purposes. I have personally never seen a PABX which considers a ground start trunk T/R battery reversal as indicative of any type of answer supervision, nor have I seen a tariff for this type of offering - by New York Telephone, at least. On the other hand, I don't exactly claim intimate familiarity with all the new-fangled PABX's on the market these days, so I suppose anything is possible. > The lack of supervision also means that the PBX is solely at the mercy > of the local user to know when a call is over. If an outside caller > calls you (or you call out), the switch will keep the circuit up until > you hang up your phone, it can't tell when the outside party hangs up. > This means that among other things --- if you leave your phone off > hook rather than hanging up on an inbound call --- you are tying up a > CO trunk. In some cases the CO will detect this condition and drop > the call on its side, opening the line up to glare or other problems. I have never seen a PABX without some type of CPC relay or equivalent detector circuit which operates on CO trunk loop current. Don't forget, loop current is loop current, whether the trunk is loop start or ground start. A momentary loop open on an outgoing call, as created when a callED party disconnects, is detected by the PABX and forces release of the station from the affected trunk. The same thing holds true for an incoming call, in which case the momentary loop open is created when the callING party disconnects first. > Now finally, in case you haven't figured it out, the above is the > reason you can't forward incoming calls back out without Ground Start > trunks. The PBX has no one to watch for final disconnection. There is no reason why a PABX cannot consider a momentary loop open on either the incoming trunk or outgoing forwarded trunk as the basis to disconnect the forwarded call. Even on a loop start line, a momentary loop open WILL be provided when the incoming call disconnects. I won't disagree that the tip supervision on a ground start trunk is better because it is a *sustained* rather than momentary open, but there is no fundamental reason why loop start trunks cannot function in this application. Don't forget that station-installed call diverters, such as those made by Ford Industries, have been available for almost 20 years. Such call diverters *always* utilized two loop start lines. > The reason I know all this is that I used to work as telecom manager > (among other things) at a company that had a Loop Start only PBX which > did allow forwarding to outside lines and I had to regularly go > through all the trunks into the system to find the ones that were hung > and fix them up (disconnection for about 5 mins seemed to do it). I don't doubt your experience, but I suspect it is the result of a PABX which was poorly designed with inadequate CPC control, or the result of an incorrect installation (i.e., perhaps wrong wiring options). <> Larry Lippman @ Recognition Research Corp. - Uniquex Corp. - Viatran Corp. <> UUCP {boulder|decvax|rutgers|watmath}!acsu.buffalo.edu!kitty!larry <> TEL 716/688-1231 || 716/773-1700 {utzoo|uunet}!/ \uniquex!larry <> FAX 716/741-9635 || 716/773-2488
Brian Kantor <brian@ucsd.edu> (04/09/90)
Our Ericsson MD-110 switch seems to have at least two types of analog line interfaces: one does not provide loop supervision to the called extension when a callER disconnects. Instead, the audio drops immediately to a reorder tone and the line is marked out of service if it does not go on hook within a few seconds. Going on-hook for a while (more than a few seconds) will return the line to in-service status without manual intervention - a good thing, since our switchroom is unattended from early evening on. As you might guess, this lack of supervision is a major pain for answering machines, modems, and such, requiring that they sense the reorder tone and disconnect. Luckily, the reorder tone (a high/low warble) is not sensed as carrier for very long by most modems, although many spew incredible amounts of garbage on the screen when this happens. It's even more annoying in that calls to out-of-service lines are forwarded to the attendant's position, which if unmanned after hours, either gives an inappropriate recording or just rings no answer. I am told that the number of analog lines with loop supervision is also quite limited - apparently the interface cards with loop supervision can only support a few lines each, whereas they get several lines per card for the non-supervised lines. - Brian
Vance Shipley <vances@xenitec.on.ca> (04/11/90)
In article <6170@accuvax.nwu.edu>kitty!larry@uunet.uu.net(Larry Lippman) writes: >I have personally never seen a PABX which considers a ground >start trunk T/R battery reversal as indicative of any type of answer >supervision, nor have I seen a tariff for this type of offering - by >New York Telephone, at least. On the other hand, I don't exactly >claim intimate familiarity with all the new-fangled PABX's on the >market these days, so I suppose anything is possible. The SX series of PBX's from Mitel do recognize answer supervision on analog trunks. I believe the SL-1 by Northern Telecom can also. The reason they do is that answer supervision is tarriffed by Bell Canada (and others I asume) in one case. That case is metered local service in a hotel environment. You have the option of receiving TM3 trunks from Bell at no charge to you (they determine how many). These trunks are for local calling from guest rooms, the hotel is charged 3 cents per COMPLETED call. The hotel passes the charges (inflated of course :'>) back to the guest. Vance Shipley SwitchView - Linton Technology (519)746-4460 ... uunet!watmath!xenitec!ltg!vances
john@bovine.ati.com (John Higdon) (04/12/90)
Vance Shipley <vances@xenitec.on.ca> writes: > The SX series of PBX's from Mitel do recognize answer supervision on > analog trunks. I believe the SL-1 by Northern Telecom can also. Add to that list the ITT System 3100, as well as the Fujitsu Focus Elite. The 3100 was designed in Canada (were reversal is available) and the Focus is popular for Hotel/Motel service. John Higdon | P. O. Box 7648 | +1 408 723 1395 john@bovine.ati.com | San Jose, CA 95150 | M o o !