5963 (computer grade dual triode) production dates?

For the most part.

Sometimes yes, sometimes no. Somethings are in detail, somethings are just tiny snippets of info.

I asked on the Tube Collector list, and there is currently a little debate about who made the design - GE or RCA.

-- Will

We'd toss them around like toys...tubes were tubes, nothing more. I felt the same way when I unloaded a box full of NOS 350Bs a few years back...it's just a well done 6L6. The buyers didn't think so, enriching me with thousands of dollars. Ditto for the 101A audio amps.

It does to someone like me who's studied Latin and Greek, yes.

I understand it perfectly.

I do miss those old days, not just for the old, 1930s through '50s vintage tube toll equipment, either. Everything was designed to maintained, everyone had a job, and the economy was good....not "good" like the Bushies and Wall St. lie about today, but REALLY good. Today, things are in bad shape, whether those living in it want to wake up and smell the coffee or not. The complex I worked in for years for both PT&T and AT&T had 1800 people working in it on any given day shift, with a couple of hundred on swing and graveyard. Now, that whole complex MAY house about 25 people max at any given time. Where did all those jobs go? Wally-Fart, Burger King, Starbuck's...the list is endless.

dB

It was *nothing* to do radio equipment routines (on non-WECo designed radios) and replace 2-300 tubes in one day. Of course

As you noted quite accurately, WECo designed equipment was often turned down 30-40 years later with some of the original tubes still working! (Except of course in 43A1 terminal units and X type SF units! What a pain those were...)

...

I sympathize with your feelings, but can't agree with what you say.

I worked at small places. But the ratio of circuits to people was about the same. There was a time when each technician on shift might cover for 200-300 circuits or so. Then it became 2000, and kept going up.

Today AT&T covers whole regions with a handful of people, and I have no way to calculate a ratio of circuits to techs that relates to what we had 40 years ago.

But I *can* tell you what the difference is! Long distance today costs less than a penny per minute. The actual bill that we pay for a long distance call still has more to do with the cost of local telephone service than any other single factor. And we have the Internet (which is transported on the same long distance facilities that voice calls are).

To put it mildly, the transformation that modern telecommunications technology has brought to the world in the last 40 years has had more to do with *improving* the quality (and length) of life for the people who have access to it (the industrialized world) than any other single cause.

Which probably means that without it, you and I would by now be dead an buried rather than alive, healthy, and kicking on Usenet!

More. The 416C, the last version, went for a nice $460 apiece. The

We used to send out 3KW klystrons in for rebuilding, if they weren't too far pitted. That'd save a cool $4K a unit.

ACK! SF UNITS! BOOO....HISS!

More. The 416C, the last version, went for a nice $460 apiece. The

We used to send out 3KW klystrons in for rebuilding, if they weren't too far pitted. That'd save a cool $4K a unit.

ACK! SF UNITS! BOOO....HISS!

You pay a penny a minute (cell phoners generally pay zero per minute) and local POTS goes up and up to compensate. Now that the Bell System is about 75% back together, things will change.

How it relates to "length of life," I have not a clue.

Somehow, I'd lay that increase at better medical care I get from my non-profit HMO than communications availability. As for POTS and LD service, it's no better than it was circa 1990, and probably a little worse. Reorders in peak periods are common now; they were a mortal sin back in the '70s. Transmission impairments due to over squeezing bitstuffing schemes yields degraded transmission qauality as well on DDD access, but with so many people on those rotten cell phones, it's hard to pin down where the problem lies.

dB

The entire health care delivery system is *totally* dependent upon it.

We *don't* want to go back to 1950...

How do you think they are able to provide better care!

I don't agree. SS7 is one difference, and of course in 1990 digital switching was ubiquitous for LD, but most of the US was about 33% digital.

On the other hand, I'd agree that since about 1995 we haven't gone far. But the decade from 1985 to 1995 was the one where the *huge* change took place.

I don't think it is so much that people are on cell phones, as it is that there is no longer *anyone paying attention* to what any given switch is actually *doing*.

The network is monitored from a centralized NOC, where nobody knows what any particular switch is actually doing. The don't see specifics, only trends. If there is a trend for some particular problem, then someone will analyze it to death, and maybe even find a cause. But if some particular switch as a single problem, it just may *never* get notice if it doesn't actually cause an alarm.

SS7 was already in deployment in major toll centers by '90. I was involved in that regarding international trunking of Pacific Rim countries from an INTELSAT Class A earth station to various 4ESS switches in California.

Not really. The POTS user wasn't really affected for the better by SS7 deployment. The big change for them was the move of toll switching from 4A crossbar to 4ESS machines and the move of trunks off of FM radio and coaxial cable onto lightwave. That difference was huge in the '80s.

Quality doesn't matter to dumb consumers. If it did, there would be very few cellular customers who were happy with their service. Now, hosed up transmission, dropped called, and all the usual cell phone garbage is somehow "normal." If it were happening in the regulated era, PUCs and the FCC would have been overflowing with complaints.

You're talking to someone who's actually been in the AT&T NOC in Bedminster. They know a LOT about what each switch is doing, trust me on this. Of course, the competitors, MCI (now VZ) and those other losers haven't got a clue until there's a major failure. The Loma Prieta earthquake in 1987 made it abundently clear that MCI was somewhat of a fraud. Not only did all their switches fail, but their "NOC" had no clue as to how to reroute traffic. Even after the Oakland toll center was condemned, the 4ESS kept right on putting traffic through, although in simplex mode.

dB

Sure, but "in major toll centers" is *very* limiting, because the advantages of SS7 were only realized when it was extended to the Local Exchange Carriers. Given that the first full implementation (by Sprint) for a toll network was in 1988 (and AT&T very shortly after that), there had been no incentive for LECs to implement SS7 *until* 1990. In fact, many of them were still using non-digital switching systems that could not make use of SS7.

The real import of SS7 was not available until *after* 1990, and it was rather rapidly adopted between then and 1995.

Exactly. And it wasn't until *that* was complete before the LECs could actually make use of it.

Not by the SS7 deployment of the IXC's! But they were certainly affected between 1990 and 1995 by the massive move of LECs to SS7.

Yes... the *late* '80s. But note again that you are considering only Long Distance (4ESS etc.), not line switching by LECs.

True for the quality of transmission characteristics. But of course there are other measures of "quality", and some of them

I've spent a great deal of time studying how and why a NOC functions the way it does (or doesn't), in efforts to design better trouble ticketing systems.

Trust me, there is a *huge* amount of what is going on that we don't know. Remember back when every switch was maintained by a crew that had at least a few old switchmen... and they used to be so tuned to what the switch *sounded* like, that when they heard an unusual noise they set about to figure out what was wrong? We don't have that level of familiarity with the switching systems anymore.

I wrote software to analyze log reports at the Fairbanks Toll Center back in the early 1990's. This is a topic that I have more than a casual introduction to. The biggest problem is spotting unique troubles that are *not* trends. For example, a digit receiver that drops the number every 5th time it is used... will most likely *never* be discovered. From a signal device, the numbers are not high enough to trigger anyone's attention. If the device itself does not go into alarm, it will continue to dump call forever.

Ha ha, you think AT&T is any better????

AT&T is so hung up on how to administer the operations of a large company that they can't *operate* any given small part of it effectively. The result is an "efficient" top level, managing a large number of ineffective (hence not profitable) operations.

Basically they can't implement anything in any part of the company unless it is something that can be done company wide in an efficient manner.

We can find just as many isolated incidents of AT&T being just as confused. Ask someone why they have both a 4ESS *and* a DMS-200 in the Anchorage Toll Center...

Well, considering that GTE was still running SXS in the majority of their offices in California, that would be true...for them. By the '80s, the RBOCs were already phasing out 1A-ESS to be replaced by

That was true in the SXS and XBAR days, but died with ESS. Most 4A men could tell when the switch was going to dump a card due to a bad link frame or other problem just by the sound of the relay sequence while putting a 101 call through the switch.

Plus, you were probably stuck with ALASCOM, too!

Not anymore, but in the analog days after the Wendover disaster in '61, AT&T Long LInes was VERY good at isolating and rerouting around trouble. When Bob Allen took over as CEO of AT&T, all that went to crap.

Nice try, but simply not true. The failuire of AT&T in later years was due to one reason and one reason only: incompetence at the top.

NorTel = NO tel

"Phasing out" is not the same as "phased out". It wasn't anywhere near completed by 1990.

Granted that all of that is true, but it doesn't make the RBOC's all that much better... :-)

Which at the time was significantly *better* than AT&T. Of course since 1995 it *is* AT&T.

(I should probably qualify that statement. When Pacific Telecom Inc owned Alascom they had *exactly* the same upper management problems that AT&T did: a total lack of vision to encompass the future of telecommunications using modern technology. The biggest difference was size, which translates to resourses. The PTI Board of Directors and the Alascom Board of Directors, assisted by their management team, came to the conclusion in the late 1980's that they had no idea what direction the company should go, and decided to sell it to AT&T. It took them until

Which of course demonstrates *exactly* my point.

Except that I wouldn't say Bob Allen was at fault in quite the manner you describe. His failing was not being able to root out the entrenched Long Lines hold on the upper levels of AT&T management. He was replaced because he failed at that. His replacement was also replaced for the same reason, as was the next replacement...

And of course the reason AT&T failed, and has been sold, is because nobody was actually ever successful at restructuring the upper management to operate in a way that made sense with modern technology.

That is exactly what I said.

Innovation at the lower levels was absolutely ignored by upper level management when making decisions. They would not allow a regional or departmental (or any other from of compartmentalism) managment team to implement anything unique to themselves. If it could not be micromanaged from the upper levels, it was not allowed. It had to be "one size fits all". That is perhaps efficient somewhere, but not for the dominant long distance carrier in the US.

Yeah, sure. And that is the sort of brainlessness that cause AT&T to fail.

That is, for example, why Northern Telecom in a ten year period from 1975 to 1985 managed to grab 40% of the US market? (Not the least of reasons being they were first with an all digital switch that included a digital network fabric.)

But more to the point, the reason there is a DMS-200 in the Anchorage Toll Center is because the 4ESS cannot do what the DMS-200 does. Therefore the DMS-200 stays. The 4ESS was installed simply because when AT&T bought the business in Alaska they could not imagine a Toll Center without a 4ESS. Without determining whether it was required or even useful, they installed one. It was neither.

On the other hand, I've never heard of anything a 4E can do that a DMS-200 won't also do. I don't really see either of them as significantly better than the other. The same is true of the 5E compared to a DMS-100. The one catch might be that a DMS-100/200 is more flexible than a 5E.

You sound like an Air Force Tech Controller. How about those lenkherts (sp?) eh? 2600hz and all for you! Wink!

-Bart

If you've never spent all day, weeks on end, adjusting 255A relays or X-Type SF Units...

Keep in mind that the Bell System was designed by geniuses to be operated by idiots.

(On the other hand, in Alaska we had a PSTN that was designed by idiots, and because of that required geniuses to operate it.)

I thought the first installation of SS7 in the Bell system was connected to the last of the crossbar toll switches that Bell installed, at least that is what Bell claimed in an old issue of the "Bell Labs Record" IIRC?

Regards,

John Byrns

Surf my web pages at,

The TELSAM scores told the tale...the "indies" sucked. Some of the worst came from various GTE properties, with United a close second. Even some of the few "mom 'n pop" telcos out there scored better.

When I went to the Satellite Transmission Systems/Varian factory in Melbourne, FL to check on an order of IBS modems, there, in the back, was an old, beat to crap Alascom "mobile unit" that they'd somehow shoehorned into an old GMC motorhome...you know, the fiberglas ones from the '70s? It was actually worse than some of the engineering boondoggles I'd seen out of New Jersey...but not by much!

All wrong. Bob Allen was "replaced" because Leslie Stahl on CBS outed him on "60 Minutes" for being indifferent about the company's (and the employees') situation at the time. Prior to the "60 Minutes" surprise, Wall St. loved Allen, because every time T's common stock would take a dip, he'd lay off more people, and thus buoy the stock. When Allen got thrown out, there weren't enough "indians" to do the work in the field, while management was as fat as ever. Armstrong did one good thing, in that the next round of downsizing was mostly lower and middle management, who'd been building empires and licking Allen and his underlings' boots for years. When Allen left, there were 1.15 managers for every working employee. Chew on that for a moment.

Technology wasn't the problem...it was failed upper management, period. AT&T had all the resources, and, instead of capitalizing on that, they put the guy in there as CEO that 1.) approved the AT&T PC

Not true. During Allen's "reign of stupidity," district and above level managers kept complaining about having to make decisions they felt were "too small" to be their concern, and so, decision making was driven to the absolute lowest tier, sometimes to the techs themselves, with no clue as to strategy, funding, logistics...nothing. What emerged was even more "empire building" as first levels and even techs, left to their won devices, would try to carve out a layoff-proof niche in the crumbling AT&T structure.

Ask Pac Bell (later SBC) about NorTel switches. PB bought a shitload of them right after Divestiture and wound up wrecking them out in as little as three or four years due to reliability and support problems, opting to go with the 5E almost exclusively.

The DMS-200 could do end office functions. The 4E was a huge toll switch, period. Next? The 5E could do both (well, not being a HUGE toll switch, but a medium toll switch), and more, although it, too, had design problems which took some years to fix.

I have...it'll last 50-100 years. DMS-200s were good for...five? Ten at the outside?

Why were 5Es outselling DMS-100s almost 5 to 1 after it became common knowledge that NorTel couldn't support their own product? If it weren't for the 5E, Lucent would be gone already. It was their only profitable and in-demand big product line for a long, long time. No one buys 4Es anymore, either...the ones in service just get added on to and keep on going.

I was USAF, but not for telcom.

Lenkurt actually built some good, although quirky, stuff back in the day before GTE gutted them. Lenkert 75A and 775A MW radios were as reliable as WECO counterparts, although their protection switching arrangement was a little hosed up. Most of the quality of those radio was, basically, due to the Varian TWTs, I'm sure. Lenkurt could also give you 12 channels of P-T-P MW for about half of what WECO could, and without all that unneeded WLEL gear that HAD to be sold with each WECO radio installation. I worked on almost all WECO radios from the TA (2 GHz) to the AR6A (6 GHz, single side band AM...an unmitigated disaster), and I wouldn't sell Lenkurt stuff completely short. Some of their stuff, though, was just plain...weird.

Worst MW radio I ever saw: Collins.

Most interesting piece of Lenkurt gear I worked with: 51A remote telemetry system...same system that, as of 1990, was controlling everything inside Grand Coulee Dam for the DoI.

Once word came down that GTE was axing Lenkurt and Automatic Electric and the layoffs started, you couldn't get too much good out of Lenkurt up there in Sunnyvale. Another American company bit the dust.

Two other small indies that built good stuff back in "the day"...Farinon and Lynch.

dB

Oops, sorry, please cancel that, I was thinking of the first Bell system installation of SS6 which connected the last crossbar Toll Switch installed in the Bell System with the first ESS Toll Switch installed in the Bell System.

Regards,

John Byrns

Surf my web pages at,

GTE was built by idiots, putting together broken down mom 'n pop locals, and was run by idiots. I know...my father-in-law was there and saw it all. It's good that Bell Atlantic was the surviving company of the VZ merger; else, the whole works would've been eaten up by Whitacre at SBC...and still may be.

Never sell Whitacre short. He's a Bell System old timer in the best tradition, and knows what he's doing. Eating AT&T for him was most likely a "See? Told ya so" to the Noo Joisey morons that ruined T and kept the associated companies under their thumbs for decades . Whitacre dreams of reuniting the Bell System, but doesn't seem to face the fact that when and if that happens, reregulation will naturally come with it....his only flaw of strategic vision I've seen so far.

The Labs did indeed cook up a harebrained scheme to retrofit #5-XBAR translators for SS7, which, of course, was ludicrous and was quietly dropped. You'll also note in the Record that XBAR engineering groups at the Labs also cooked up "custom calling" features, borrowed from the #2-XBAR AUTOVON switch, to retrofit #5 machines...about as good a case of "empire building" as I can show. ALL of the big 4As were gone by the mid-80s, with a smattering of #5 and 4M tandem switches still lingering in locations where a "drop in" of a 5E wasn't easily done, or where the office was simply slated for elimination from the switched network.

Last case of this I remember was the #5-XBAR at Death Valley Jct., CA. This switch was built in the '70s to do exactly what the #5 machine was originally designed to do...be a variable density end and tandem switch in one office. The 5E has all that capability, but due to some odd problems, such as analog 4 wire short haul carrier systems feeding "local loops," it took some time to get rid of, as the outside plant had to be also digitized (SLC96, etc) in order to interface directly with the switch and prevent a major and costly kludge.

Pacific Telephone, who owned the office at the time, got a "deal" on that XBAR machine, and couldn't justify installing a new, but quickly obsolete, 1A-ESS for about 500 subscribers, nor would a 1A do toll functions or digital facility inteface. Add to that fact that it was one of only five (?) end offices owned solely by AT&T after Divestiture, and the legal and political impediments became astounding. Also, the DVLY 5-XBAR was a very new machine, one of the last #5s built by the Oklahoma City Works before it shut down production, and wasn't fully amortized at the time. Once the bean counters saw that it was fully amortized, it was scrapped. The #5 machine in that place replaced a 100 line SXS CDO that had existed since the mid '40s. The major indie telco in that whole desert area at the time was formerly Califoria Interstate Telephone, which was gobbled up by ConTel in the '60s. They had all manual board toll service and open wire C carrier way into the '70s, while Death Valley had XBAR DDD.

And no, the DVLY #5-XBAR machine was never retrofitted for SS7.

The first 4ESS was installed in Baltimore, MD in 1964. At the same time, Long Lines was installing underground "hardened" 4Es across the country as part of the AUTOVON military network. That predates SS6 by quite a long time, but surely the switch was retrofitted later. There was no XBAR at all in Baltimore, as it was the Labs' "ESS City" that had been first envisioned in the mid '50s when the ESS project first got off the ground. That machine, although completely retrofitted, is still in service.

You may recall that Baltimore was the first "TouchTone-only" city as well, as ten button 2500 tel sets were the only option. Rotary sets were not deployed there after the last of the 1A-ESS end offices was finished in the early '70s. Also, all "N" and other short haul analog carriers serving the area were wholesale replaced with early DS1 systems, another first, in order to feed the 4E directly without T-CXR-to-analog channel banks at the toll office. The Labs used the Baltimore experience to pattern its nationwide elimination of electromechanical switching and analog short haul carrier that came later. Much of the documentation and procedures of the big LA cutover of the 70T and 92T 4E switches in 1979 and 1982 came from the Baltimore project, as they all did. Both of those massive cutovers happened without a single minute of call blockage in the metro area, a feat that could not be done today. The only impact on callers the next morning was that their toll calls zipped through a lot faster and there was no noise on their toll trunks in the city anymore.