History -- why not 20 pulse/sec dials? [telecom]

Toll operators did use 20 pps dials. Of course they were dialing on a short loop while many customers were dialing on longer, or even very long loops. The transmission properties of long loops distort dial pulses, so the "safe" thing to do was to design customer equipment to avoid problems on any circuit.

E. Tappert

I remember doing that in the 1950's - at least 20 pps and up to where the CO wouldn't keep up (Boston Metro).

Bell System Inertia (highly technical):-(

If I recall correctly, at one point I owned a modem with hook control but not touch-tone... dialling was accomplished by setting and clearing a bit on an I/O port under program control. I don't know how much a DTMF generator would have added to the cost of the modem, but we really didn't need it either. And, yes, many terminal programs offered to dial faster than 10 PPS.

I remember an old timer telling me one time that for 20pps to work you had to be fairly close to the CO. I wonder if it had to do with what the pulses looked like through a 20 mile loop out to some farm?

- Ron

> If I recall correctly, at one point I owned a modem with hook control but not
> touch-tone... dialling was accomplished by setting and clearing a bit on an
> I/O port under program control. I don't know how much a DTMF generator would
> have added to the cost of the modem, but we really didn't need it either.
> And, yes, many terminal programs offered to dial faster than 10 PPS. >

Makes me think of all the money I wasted on the ever-better Hayes modems before the Internet became ubiquitous.

I don't recall the specifics of the register setting, but you could make the later Hayes spit out the DTMF string really fast. The DTMF orgination hoppers (or whatever they're called) on my local DMS-100 could correctly process the rapid fire stuff until it was set so fast it sounded like a blur. I doubt those DMTF readers on a SxS office would have done quite as well.

Another poster also pointed out the distance issues, which is a good point.

I was thinking that the high speed dials would be on city lines served by Panel or #1 crossbar, not so much suburban or rural offices, which might have longer loops. Back in those days (1950s) No. 5 crossbar wasn't too common in city service, and I suspect panel and #1xbar, being a high volume city exchange, didn't have 20 mile farm loops.

I'd have to check on a map to see the distance from us to our exchange, but roughly I'd estimtate about 3-4 miles (that is, it worked at that distance.)

I realized another issue: concentrators. Did big city subscribers say 4 miles from the CO have their own direct copper line or was it concentrated?

FWIW, our exchange building handled a high volume of lines, which was primarily residential:

LI 8 LI 9 WA 4 WA 7 HA 4 CA 4

By the time modems were cheap enough to be used at home (late 1970s), the cost of electronics had dropped enough so that frequency converters were not as expensive. By the late 1970s Touch Tone was available to the majority of subscribers and gaining in popularity. Further, Panel was pretty much gone by then and ESS was quickly coming on-line.

I was thinking more for the 1950s and 1960s.

On Thu, 07 Aug 2008 21:11:58 -0400, Sam Spade wrote: ......

The major problem with rapid DTMF is the gap between identical digits - if the receiving end setup needed a decent time to discriminate between two separate identical tone pairs (rather than guess it was one digit with an interruption) then you could get into all sorts of trouble.

I ran into this with voice mail systems with in-band signalling for Message Waiting indicators, you could tweak the DTMF send rate and think that you had a system functioning "better" than the default settings - but then be hammered by support calls by people saying that they don't get reliable MW while other users have no problem at all, and then eventually figure out that *all* the problem users have a pair of identical digits in their extension numbers...... :-(

FWIW, when I got my first home computer around 1992 I had rotary service. The modem was easily set to 20 pps by a AT command string. (It could also be set to different make/break ratios for use in other countries). The ESS had no trouble.

I bought an phone with an auto dialer at that time. That dialed pulse very slowly, I could dial faster on a plain rotary phone, which was frustrating. It spit out Touch Tone fast.

I don't think line concentrators came into play in the Bell System before the late '50s. Rural carrier may have been earlier; can't speak to the dates on that.

I don't remembe the register number either, but you used it to specify, in milliseconds, the duation of the tone burst _and_ the inter-digit delay. Default setting ws '70', which dialed a local (7-digit) number in almost exactly 1 second.

Minimum value the genuine Hayes would accept was 20, which was shorter than most COs would recognize _accurately_. A setting of 35 generally worked on in-town lines.

For instance, some pay phones were pre-pay, some were post-pay.
> Also, they maintained stocks of non-dial phones for areas with
> manual service and certain PBX extensions (such as in a plain
> motel.)

In the late 1940s or early 1950s I had an occasion to use a pay phone in Muskogee, Oklahoma, a fair-sized manual exchange.

It had manual pre-pay service...similar to a dial office. You put your money in the slot, the operator said "Number, please," took your number and dropped off. From that point on everything proceeded just like you had dialed the number on a pre-pay phone.

Wes Leatherock snipped-for-privacy@yahoo.com snipped-for-privacy@aol.com

To clarify for our younger viewers:

Pay telephone had two types: In cities there was the pre-pay, in which you deposited your coin to get a dial tone. If your call wasn't answered your coin was returned. This required the phone have to a coin-holding area and a switchable opening to either collect the money or return it to the customer. A special signal from the CO operated that opening.

In rural areas there was a simpler phone known as post pay. You got a dial tone, dialed the number, and waited for answer. If someone answered, you then promptly inserted your coin. There was no holding area, your coin went directly into the collection box. The was a polar relay inside the phone so that when the coin was dropped the relay would connect your transmitter circuit so you could talk, at the end of the call, polarity was reversed to reset the relay.

The post pay was a simpler phone and had simpler CO interfaces.

My point of all this was that the Bell System had to maintain two distinct types of pay phones--that it was not a universal inventory.

Further, phones had to be maintained in inventory or modified to have a dial or a dial blank on them; another separate task.

Thus, maintaining or modifying dials to be 20 pps would not be a big deal.

The "specal signal" from the CO was either -130 volts from tip to ground (Coin Return Battery), or +130 volts from tip to ground (Coin Collect Battery). Students at M.I.T. implemented a workaround, called a "T", witch assured that Coin-Collect Battery would be changed so that the phone received Coin-Return Battery instead.

This device was call a "T" because it was one of a triumvirate of weapons that the kids employed in their never-ending battle against boredom and Ma Bell: the "Black Box", code named "Agnew", and the "Blue Box", code named "Spiro". The three weapons were employed so effectively that Western Electric developed the "Fortress" model of pay phone, which would lock up and become innoperative if the CO talk battery (-48 & Ground) was reversed, thus preventing the use of "T" rectifiers. Local Testmen had a lot of trouble with them: they had to put talk battery on the pair before attempting to collect a stuck coin, or the station totalizer wouldn't "home", which locked the phone. Unfortunately, not all the cable test tags were consistent: some had "tipover" errors in their wiring, which caused the Fortress phones to lock when the testman applied talk battery.

The ESS killed the Agnew, and SS7 caused the (long mourned) demise of the Spiro. Ma Bell wouldn't let us play with her toys anymore, except, of course, for those who still use Red Boxes. Red Boxers have a standing joke: when asked what a payphone call costs, they answer "It's always Five Cents".

Bill "When you get that first call from Homeland Security, you know you've arrived" Horne Temporary Moderator

Please put [Telecom] at the end of your subject line, or I may never see your post! Thanks!

We have a new address for email submissions: telecomdigestmoderator atsign telecom-digest.org. This is only for those who submit posts via email: if you use a newsreader or a web interface to contribute to the digest, you don't need to change anything.

The Anchor Signalman did this. They advertised that they had autodial, but in fact they had the ability to control the hookswitch under software control, which was not useful when connected to a dumb terminal.

--scott

I never thought about that... the ones I was thinking of were VICmodems (1650 and up - the 1600 didn't have hook control - and workalikes from various manufacturers) which plugged directly into the user port of a VIC-20, Commodore 64, Commodore 128, etc.; since they couldn't be plugged into a VT-100 style terminal, I guess the limitation is somewhat more forgivable.