Great to see the 1AESS / 2BESS / 3ESS commands again. This machine was not designed to be powered down and then back on. Back in the 1970s, when a 1ESS was put into service, the Engineering goal was for the switch not to be Out Of Service for more than 2 hours in 30 years.
We had a 2BESS at IBM in San Jose in the 80s. It’s most likely been changed out in 2022. I heard they also had a 2B machine that served Catalina Island in SoCal in the 80s. The good ole PAC*BELL days.
@@steve94044 I worked on the 3ESS system that went to Avalon on Catalina Island around 1978. We shipped it special in two pieces so It would fit the ferry that took it to the Island.
@@535tony That’s a cute building the co is in. My co worker was the island phone tech in the 1980’s. He lived in one of the pac bell rentals there. His first name is Rod.
@@martello44 I don’t know where the 3ESS was developed. The 3ESS was mostly built at the Oklahoma works. After I worked on the 3ESS for two years all production was moved to Oklahoma works for the 3ESS. I then started working on the 3B processor that was used in the 5ESS. These were developed at the Northern Illinois Works in lisle Illinois where I worked. I worked from 1979 to 1983 on the 3B processor from prototype to production. The Northern Illinois Works was shut down in 1983 when I left.
Sarah and Astrid, you both have an almost superhuman capacity for patience, good humour and calmness, even when faced with what are clearly frustrating and complicated technical issues with these beloved and amazing beasts of historical systems. I greatly appreciate both of you for achieving something infinitely rare in combining your tenacity with being genuinely awesome human beings. Many others in network engineering would do well to follow your example.
I built the 3ESS systems in the late 70’s from the Northern Illinois Works in Lisle Illinois. Great memories. I also built the 3B processor that was used in the 5ESS till I left Western Electric in 1983.
This is so cool. I work on old heating controls and I love seeing operators who know the "quirks" and "personalities" of their equipment lol. Yall are legends!
Ahh yess..."intermittent conditions". When I started my telecommunications career on a main distributing frame, miles and miles of copper wiring gave me a "few" of those...daily, lol. Great channel!
Hi, I'm currently underway with network engineering studies. While I'm engaged with enterprise level state of the art systems, I also know the value of understanding our networking heritage and where these modern systems came from. That being said, I appreciate these videos even moreso and intend to utilize aspects of these systems in my studies as well. It's just so different when you're passionate about the subject. If I try and talk about things with others, their eyes start to glaze over after a few minutes. It's so beyond refreshing listening to my people discuss these systems. Just top down, I love troubleshooting. I love having to drill down into issues where you don't have proper schematics available and not just memorizing button press sequences but knowing WHY those buttons are pressed and what they do.
I remember interviewing with Bell Labs in Napierville, Illinois in 1981 (no, they didn't hire me). While I was there, they showed me a 3ESS and I heard the horrible noise it made when everything was being reset - all the relays were switching. The person with me commented that if you ever hear that sound in a central office that it meant they were having serious trouble because they rarely reset an ESS.
I was a Test Engineer for GTE back in the 70's and worked on the 2EAX, installing and configuring it at various locations around the country. Great times!
Was going to say, the ones I've worked on are literally never fully powered down until they're decomissioned. Different function, but on an HLR (Home Location Register) I worked on, the one time we had to power down a whole frame (because we'd lost commercial power, diesel generator had failed to start and the battery plant was running at 36VDC), when it came back online it had lost 4 of 8 SCSI drives and needed me to fly out and nurse it back to health. Fun times...
I recall the conversation from Alcatel when we were trying to sell our 600E in Manhattan... "If you turn it off, we'll never recertify it!" I think we sold it _and_ the building. :-) That was in the early 2000's; it had been running continuously since the 70's. (No one was sure it _could_ be powered back up.)
@@jfbeam early in my career, I worked for an oil refinery in the Midwest. The place opened in 1970 and by the time I came along, almost all the people who opened it had retired, and in that time they had never needed to stop the whole facility for safety or anything .. put another way, nobody knew how to turn the machines back on from a cold start. To my knowledge they still haven’t had to cold start the place (but there was a lot of anxiety around writing the documentation when I was there)
@@satsuke It's growing problem in technology fields. Few ever consider the issue of bootstrapping their operations. My former employer used to have a yearly "shutdown" process where everything is shutdown and restarted once a year, around the time most people take off on summer vacations. Of course, that wasn't an ISP, telco, cable company, power company, etc. If we turn off a building, it doesn't really effect anyone else. At my last telco job, rebooting _anything_ was a paperwork nightmare; and there are things you simply _DO NOT_ reboot, much less even look at the power cables.
@@jfbeam I was recently RIFF’d from my telco job, and it wasn’t so much an issue for us as they went with an aggressive trinary redundancy platform, so the Nokia could take an entire site offline for a month to perform upgrades and bring them up in order. It was still a nightmare of planning because of the dependencies when it came to data model changes and trying to support GSM/UMTS/HSDP/LTE/NR all at the same time
@@jfbeam bootstrapping itself wasn’t that big a deal though as it was a docker/kuberneties on VM arrangement .. So we could move a live node to a different piece of hardware with it still running without traffic loss. It honestly wasn’t as satisfying as the old ways when we’d troubleshoot at small scale .. it is easier to throw away the VM and spin up a new one if something isn’t acting correctly then it is to try and troubleshoot
I had the same thought train! I was disappointed that the tape emulator didn't have a speaker to emulate the sounds of whirring drive mechanics though :-(
PUC data links were my bain back in the day! 45 years in Telecom and I am amazed that these curators took this tech ( and read the BSP's) and have it working again... Congrats and Thank you!
Very cool, I worked at Bell Labs in the 80's and 90's in NJ, where I wrote a terminal interface (no API's back then) for the 1ESS, 2ESS, 3ESS, 5ESS, and DMS100. My interface created a single user interface to repair centers so they did not need to learn all the different switch commands of each switch type. This was very popular during the time when they had to verify the PIC code, during the growth of new long distance carriers, like MCI and World Comm. The thing about the 3ESS was it was a very small and remote located switch and the only access was via a 300 baud modem. How times have changed.
When I went to network switching systems operation school in 2002 I was introduced into the world that you are showcasing. Thank you so much! I've been a phreakophile my whole life.
How was I born in Puyallup and yet had NO CLUE the connections museum was a thing?!?! I could nerd out in there ALL DAY!!!! I'd be able to listen to you explain this stuff all day.
I'm so happy to see another video from the Connections Museum! Feels like we were waiting forever; it was worth it. :) I've never touched a 3ESS before, but I managed an Avaya Communication Manager PBX for years. The shared "heritage" between CM and 3ESS amuses the hell out of me. :D When I started watching videos on 5ESS and 3ESS, I was astounded by how much seemed familiar.
When I worked for Nortel installing and testing the DMS I would sometimes see these in the office. I miss working in central offices. It was a good company to work for.
This stuff is fascinating. It's older, but not that different to powering up a late 90s aircraft. The act of powering it up can generate faults that need to be reset once everything is stable. It's quicker than this, though haha
What's the relay cadence I'm hearing that starts at 6:24 when booted? I recall hearing that exact same cadence when placing a call between the rings - mid 80s - Detroit area.
That is the office interrupter (OI). It provides timing pulses for the various circuits that need them. The CPUs operate the relays at 60 and 120 pulses per minute.
Although I haven't seen a cartoon page or an RCV menu since about 2006, now I know where all those funky RCV commands I used on the HLR came from (at least I didn't have to wait for everything at 110 baud though!). Surprisingly I remembered all the status returns ("NG"=="No good"; "IP"=="In progress", etc.)
Yeah. Sometimes those RCV's were 2400. I was "lucky" to be working with CALEA, so I always had access to dedicated 9600 ports. (until the minion from Lucent pointed out full menu access wasn't supposed to be enabled on those ports. 'tho there were only 3 people in the entire company with access, I was the only one who knew how to use any of it. It's a good think we never received any tap orders.)
Finally! I have been waiting for this for years! I can't wait for the 3ESS to be back into full museum service! Thank you all for all of the hard work!
Thanks so much for doing this! It is interesting that these were designed when the control console was a teletype terminal and that the machine can't talk any faster than 110 baud.
Oh, this is absolutely beautiful to see!! Funny to think, this process was intended to be done only a few times ever at most, if even that, and it has to be done each time the switch is to be brought into action. Tedious in a way, but certainly magical to do -- to boot such a computer in such a hands on way, all for it to handle phone calls. I have to be that guy -- when might we maybe be able to make a call into or through this beauty, or will that likely forever remain a :"need to visit" kind of thing?
I spent a lot of years working in Telephone central offices. The sound of those power converters and the smell of the RH power cables are very familiar to me. That and the incessant relay clicks.
Whoooooooa, I've been missing you for so long :) Love the 3ESS tape emulators - modern electronics coupled with antiquated systems are really my thing... this said by someone who built a few Raspberry Pi based controllers for a Monotype composition caster.
I've seen that done with an old PDP 11/34 since they no longer could get disk packs for it. The room seemed so empty without the fridge sized boxes of disk platters. I felt sad when I saw outlines on the floor where they used to sit and knew what was missing.
@@PWingert1966 the mechanical aspect of old time computers definetly add a certain charm to them. The transition to new CPUs is like replacing Prague astronomical clock with just a digital display. Yes, it's more precise, smaller, but it has lost the charm, the feeling that if you try enough you might understand what's doing what.
Remembering all the commands and how everything works is a real skill set very few people have I wish my brain was better I can't remember things like this I try to learn new things but I just can't get it to stick in my brain Same thing with another language. I couldn't get it to stick either It is good there are people like you in the world
I think the university here still uses the 5ess.. as an analog backbone.. still used for some feeds and the national weather service and emergency still uses it for clock and broadcasting relay... crazy huge mainframe with 5 levels of entrance cables feeding the campus.. most are cut or repurposed but still have the glass batteries and switchgear running 24/7.. gotta use that old software to set up..
Ammazing. I really like this kind of installation, driven by simple mini-computers. Lots of electromechanics driven by a small, limited, slow brain. It is very interesting to understand how things were designed with such (very) limited computer ressource. Thank you for sharing all that.
I hope this process they're describing wouldn't involve fooling with live electricity. As I am sight challenged, I'm really fascinated by machines that have to do with telephones. I hope that one day I can come to this museum.
My dad worked for AT&T from the 60's through the 90's. We had a local mom and pop telephone provider in Califon, NJ who had advertised to come see our new switch gear one time . Once he saw it he knew it was an old 3ESS system they had dumped.
I used to commission Philips EBX8000 PBX’s all over the world. Although these were a little less complex, not much so: they could have up to 8,000 extensions. Dual processor reed relay switch under stored program control, loaded with paper tape. Operated in hex with 16+ switches. Remarkable how common many of these concepts and procedures are! And it used to take me an hour to boot up an EBX8000: these too were designed to stay on.
I guess this is how non tech people see IT people. I don't even know what I don't even know. But all of this is seriously impressive, how everything is analog as hell but still digital. And it looks like something out of a cold war bunker.
Does this switch have coin control for pay phones? That would be cool to see pay phones connected here and have coin return and coin collect work properly.
That's really cool, I imagine this is similar era as DMS10? We still have lot of DMS10's in operation, this kinda reminds me of them. I've never had to INI one myself, I'm just a NOC tech so we don't dive too deep.
Similar era, yep. The DMS10 was miles ahead of this switch though, and this one never really took off. The DMS was cheaper, more advanced, and (as we now know) lasted well into the 21st century. The 1A ESS and 2B ESS, which served larger towns were both very successful. I guess there was just a better cost/performance ratio there.
I love that this museum exists. Part of it is that for part of my brain it is fascinating. Another part sees it , excuse me, as wonderfully absurd. I wonder if it is staffed with Logopolitans who at night as a hobby do block transfer computations. I am looking forward to visiting in the fall. Thanks for being there and posting these videos.
Here in Toronto near the Eaton center bell had a building right beside it and tucked away in the basement behind a glass wall was a fully functioning switch that was for the building and likely the downtown core. I managed to grab a few pictures for a project I was doing at the time. A couple of years later they frosted the bottom five feet of the glass wall. It made me sad to know what was in there and that it would never be visible again. A few years later they dismantled it. and the room's lights were turned off and then the whole section of wall was replaced with Gyprock. I would love to learn more about that if anyone has information.
Back in the late 80s I replaced one these with a DMS100 remote in a small community outside of Witchita KS. I dont remember it looking like that. I had to hook up the Nortel Board to Board tester no a no test trunk on the #3. Took me forever as none of the local techs seemed to know much about them. Finally got it to work and that was the last I ever saw of those machines
Is there a resource or manual describing the terminal interface for the #1 and #3 ESS digitized somewhere? I'm really curious and would love to learn it.
I love when massive ancient computers are hooked up to small modern computers running minicom or something because genuine teletypes are a pain to obtain lol
I'd like to see the boards in the 3A CCs themselves. I understand the 3ESS is a very rare machine, built for small offices and not produced in particularly large numbers.
It's called TL1. I'm sure you can find some generic documentation that describes the format and what everything means. A lot of words are abbreviated so that they can be transmitted quicker over slow modems.
I was a test engineer at OKC Works where I ttouble shot manufacturing defects on #2 ESS processor and AIS system. Used to design machine code programs and hand mag twister memory cards as trouble shooting aids. Lotsa fun. Would like to see a 2ESS processor video because I would have likely tested it back in early 70s.
How the HELL does one even begin to learn all this?? I bet the two people who work on this are the only 2 out of approximately 7 BILLION people on Earth who could get this up and running...that's pretty amazing if you ask me.
Wow! Reminds me of the days back when I worked for MobileComm, a Bell South Company, rebuilding voicemail machines. The VRS in particular was great when it worked, and pure hell when it didn't.
Right. They're known as the "3A CC" or "3A Central Control". They were designed specifically to control a telephone switch, and are not very good at doing math, or more general purpose stuff.
I have a working ASR33 that I am looking for a home for. I would gladly send it to you. It has an RS232 interface and worked when I ran it about 2 years ago.
This is the system that replaced the step by step 355a office that I took care of in 1978 for southwestern bell. This was quite a change. Thank you so much for this interesting video, brings back alot of memories. If I ever get to Seattle, can I see this system?
Yes, you can see it, this museum shares space in the Century link CO. @ 7000 East Marginal Way. I think it is called The Connections Museum and they are open to the public.
@t13 fox ... I had to laugh when I read your Comment. I remember SXS from back in the day with GTE...I started as a Courier and remember taking a break in a SXS switch while the Bears and Packers played... Everything was dead silent. ... until Halftime and that Rural Central Office erupted with an imitation of the D-Day Invasion and I thought I would go deaf!
Here’s the description and theory of operation documents. telecomarchive.com/docs/bsp-archive/254/254-300-110_I1I.pdf telecomarchive.com/docs/bsp-archive/254/254-300-120_I1.pdf
Did you mention that the two boxes that take 10 minutes to boot (with the marching lights) are trying to 'come into phase' with one another or another system.?
So awesome to see this stuff running, it would be neat to have an actual teletype connected to that serial port too. Did the 3ESS have support for remote management over a modem?
Sarah here -- I think Astrid and Colin eventually want to hook up a real Teletype to the local serial port, at 110 baud, and then have a remote session via a modem to one of the other serial ports. That should run at 1200 baud, I think.
@@KJ7BZC I believe this is discussed in this or a similar video, and you can clearly see several sleek looking black rectangular Dataphone modems buiilt into the backside of the rack where the terminal resides.
Why does this give me like DEC vibes? I know Bell built all their own stuff, but that looks awfully similar to regular mini computers of the day, specifically a PDP or the IMSAI 8080 🤔
The later 4ESS and 5ESS switches had 32 bit processors, which - someone correct me - ran an RTOS called DMERT. Bell labs also produced a series of commercial computers which had the designation 3Bxxxx which used similar technology. The 3B series ranged in capability from a desktop unit which could handle a handful of terminals in a lab to large systems which could run a whole business office. Early machines such as the 3B20 were 32- it word size built from 4-bit wide AMD 2901 bit slices in the 1970s. Later systems incorporated the Western Electric WE32000 processor chip. Many of the 3B systems were fairly powerful and equated more to a DEC-10 system than a PDP-11. Side note: After about 1977, many PDPs got rid of the toggle switches in favor of a keypad and seven segment LED digits displaying in octal. Only a very old PDP-11/34 I worked with early on had the front panel toggle switches. The old machine required manually keying in a start address and executing a jump instruction on the switches to execute the bootstrap to start loading RT-11 O/S (in our case from an RK-05 disc). I later worked for a university where we ran System V UNIX on a DEC Vax. I got a chance to meet a few of the AT&T UNIX old timers: Mark Horton who contributed to the vi editor and ditroff and Perter Honeyman of Honey DAN BER UUCP. We were the first site outside of Bell Labs to run the revamped UUCP. We kept a couple of Telebit Trailblazer 23K modems busy pretty much 24/7 transferring our USENET feed. I personally had an AT&T 3B1 aka Safari PC7300. The 3B1 was actually made for AT&T by Convergent Technologies, was based on the Motorola 68000 CPU and ran System V r2 UNIX. The 3B1 had no commonality with the 3B2 series other than running a version of UNIX. The 3B1 could have up to 4 megs of RAM and an 80 meg MFM drive. I also had a cartridge tape and Voicepower board which I got at swap meets. The Voicepower was really cool because it had a high quality CODEC and an entire IVR system could be written with Bourne Shell commands. Graphics were green screen Hercules compatible 720 x 384. The 3B1 was a pretty powerful desktop for 1984, but it didn’t sell well and AT&T started liquidating them in a so-called fire sale in 1986. The 3B1 is sometimes blamed for Convergent Technologies demise.
When I build 3ESS in the late 70’s we used a PDP 8 computer to test the system by making simulated telephone calls. We would run the systems for days at full capacity before they would pass.
I hope all the information about starting up the system is documented somewhere. It is an incredibly complicated procedure that you won't be able to do by playing around.
Great to see the 1AESS / 2BESS / 3ESS commands again. This machine was not designed to be powered down and then back on. Back in the 1970s, when a 1ESS was put into service, the Engineering goal was for the switch not to be Out Of Service for more than 2 hours in 30 years.
We had a 2BESS at IBM in San Jose in the 80s. It’s most likely been changed out in 2022. I heard they also had a 2B machine that served Catalina Island in SoCal in the 80s. The good ole PAC*BELL days.
@@steve94044 I worked on the 3ESS system that went to Avalon on Catalina Island around 1978. We shipped it special in two pieces so It would fit the ferry that took it to the Island.
@@535tony That’s a cute building the co is in. My co worker was the island phone tech in the 1980’s. He lived in one of the pac bell rentals there. His first name is Rod.
@@535tony What part of the bell system designed the ESS and at what location?
@@martello44 I don’t know where the 3ESS was developed. The 3ESS was mostly built at the Oklahoma works. After I worked on the 3ESS for two years all production was moved to Oklahoma works for the 3ESS. I then started working on the 3B processor that was used in the 5ESS. These were developed at the Northern Illinois Works in lisle Illinois where I worked. I worked from 1979 to 1983 on the 3B processor from prototype to production. The Northern Illinois Works was shut down in 1983 when I left.
Sarah and Astrid, you both have an almost superhuman capacity for patience, good humour and calmness, even when faced with what are clearly frustrating and complicated technical issues with these beloved and amazing beasts of historical systems. I greatly appreciate both of you for achieving something infinitely rare in combining your tenacity with being genuinely awesome human beings. Many others in network engineering would do well to follow your example.
I built the 3ESS systems in the late 70’s from the Northern Illinois Works in Lisle Illinois. Great memories. I also built the 3B processor that was used in the 5ESS till I left Western Electric in 1983.
Hey Scott! Good to see you out there. We had some fun times building 3ESS on our team.
Did you know Terry Webster?
@@jameschristiansson3137 I don’t recall that name.
Did anyone else use 3B15/3B20 mainframes besides AT&T?
@@douro20 we made 3B15’s and they were all for AT&T. We mostly made the 3B20’s and they all went into 5ESS.
This is so cool. I work on old heating controls and I love seeing operators who know the "quirks" and "personalities" of their equipment lol. Yall are legends!
Its amazing just how reliable these old machines were compared to the telephone equipment of today.
Holy sh!+ that is nuts. I'm a 48 year old vintage (80's) video game and computer collector. I have never seen any thing like this! Thank you!
Coolest thing I’ve seen all week. Wonderful to learn from an engineer who is not just an operator but understands the system all the way down.
Ahh yess..."intermittent conditions". When I started my telecommunications career on a main distributing frame, miles and miles of copper wiring gave me a "few" of those...daily, lol. Great channel!
Hi, I'm currently underway with network engineering studies. While I'm engaged with enterprise level state of the art systems, I also know the value of understanding our networking heritage and where these modern systems came from. That being said, I appreciate these videos even moreso and intend to utilize aspects of these systems in my studies as well. It's just so different when you're passionate about the subject. If I try and talk about things with others, their eyes start to glaze over after a few minutes. It's so beyond refreshing listening to my people discuss these systems. Just top down, I love troubleshooting. I love having to drill down into issues where you don't have proper schematics available and not just memorizing button press sequences but knowing WHY those buttons are pressed and what they do.
I remember interviewing with Bell Labs in Napierville, Illinois in 1981 (no, they didn't hire me). While I was there, they showed me a 3ESS and I heard the horrible noise it made when everything was being reset - all the relays were switching. The person with me commented that if you ever hear that sound in a central office that it meant they were having serious trouble because they rarely reset an ESS.
I was a Test Engineer for GTE back in the 70's and worked on the 2EAX, installing and configuring it at various locations around the country. Great times!
Was going to say, the ones I've worked on are literally never fully powered down until they're decomissioned.
Different function, but on an HLR (Home Location Register) I worked on, the one time we had to power down a whole frame (because we'd lost commercial power, diesel generator had failed to start and the battery plant was running at 36VDC), when it came back online it had lost 4 of 8 SCSI drives and needed me to fly out and nurse it back to health.
Fun times...
I recall the conversation from Alcatel when we were trying to sell our 600E in Manhattan... "If you turn it off, we'll never recertify it!" I think we sold it _and_ the building. :-) That was in the early 2000's; it had been running continuously since the 70's. (No one was sure it _could_ be powered back up.)
@@jfbeam early in my career, I worked for an oil refinery in the Midwest.
The place opened in 1970 and by the time I came along, almost all the people who opened it had retired, and in that time they had never needed to stop the whole facility for safety or anything .. put another way, nobody knew how to turn the machines back on from a cold start.
To my knowledge they still haven’t had to cold start the place (but there was a lot of anxiety around writing the documentation when I was there)
@@satsuke It's growing problem in technology fields. Few ever consider the issue of bootstrapping their operations. My former employer used to have a yearly "shutdown" process where everything is shutdown and restarted once a year, around the time most people take off on summer vacations. Of course, that wasn't an ISP, telco, cable company, power company, etc. If we turn off a building, it doesn't really effect anyone else. At my last telco job, rebooting _anything_ was a paperwork nightmare; and there are things you simply _DO NOT_ reboot, much less even look at the power cables.
@@jfbeam I was recently RIFF’d from my telco job, and it wasn’t so much an issue for us as they went with an aggressive trinary redundancy platform, so the Nokia could take an entire site offline for a month to perform upgrades and bring them up in order. It was still a nightmare of planning because of the dependencies when it came to data model changes and trying to support GSM/UMTS/HSDP/LTE/NR all at the same time
@@jfbeam bootstrapping itself wasn’t that big a deal though as it was a docker/kuberneties on VM arrangement .. So we could move a live node to a different piece of hardware with it still running without traffic loss.
It honestly wasn’t as satisfying as the old ways when we’d troubleshoot at small scale .. it is easier to throw away the VM and spin up a new one if something isn’t acting correctly then it is to try and troubleshoot
I hear "tapes" and think: someone should make an emulator for that. Three minutes later, you show the emulator. That's fast hardware development! :D
I had the same thought train!
I was disappointed that the tape emulator didn't have a speaker to emulate the sounds of whirring drive mechanics though :-(
PUC data links were my bain back in the day! 45 years in Telecom and I am amazed that these curators took this tech ( and read the BSP's) and have it working again... Congrats and Thank you!
Very cool, I worked at Bell Labs in the 80's and 90's in NJ, where I wrote a terminal interface (no API's back then) for the 1ESS, 2ESS, 3ESS, 5ESS, and DMS100. My interface created a single user interface to repair centers so they did not need to learn all the different switch commands of each switch type. This was very popular during the time when they had to verify the PIC code, during the growth of new long distance carriers, like MCI and World Comm. The thing about the 3ESS was it was a very small and remote located switch and the only access was via a 300 baud modem. How times have changed.
When I went to network switching systems operation school in 2002 I was introduced into the world that you are showcasing. Thank you so much! I've been a phreakophile my whole life.
How was I born in Puyallup and yet had NO CLUE the connections museum was a thing?!?! I could nerd out in there ALL DAY!!!! I'd be able to listen to you explain this stuff all day.
I'm so happy to see another video from the Connections Museum! Feels like we were waiting forever; it was worth it. :)
I've never touched a 3ESS before, but I managed an Avaya Communication Manager PBX for years. The shared "heritage" between CM and 3ESS amuses the hell out of me. :D When I started watching videos on 5ESS and 3ESS, I was astounded by how much seemed familiar.
When I worked for Nortel installing and testing the DMS I would sometimes see these in the office. I miss working in central offices. It was a good company to work for.
Love this channel and this comment section! It all fills me w hope tbh
Oh my! The patience involved. Thumbs up!
This stuff is fascinating. It's older, but not that different to powering up a late 90s aircraft. The act of powering it up can generate faults that need to be reset once everything is stable. It's quicker than this, though haha
Thank you for your effort in documenting this amazing technology, your clearly very talented
What's the relay cadence I'm hearing that starts at 6:24 when booted? I recall hearing that exact same cadence when placing a call between the rings - mid 80s - Detroit area.
That is the office interrupter (OI). It provides timing pulses for the various circuits that need them. The CPUs operate the relays at 60 and 120 pulses per minute.
Although I haven't seen a cartoon page or an RCV menu since about 2006, now I know where all those funky RCV commands I used on the HLR came from (at least I didn't have to wait for everything at 110 baud though!). Surprisingly I remembered all the status returns ("NG"=="No good"; "IP"=="In progress", etc.)
Yeah. Sometimes those RCV's were 2400. I was "lucky" to be working with CALEA, so I always had access to dedicated 9600 ports. (until the minion from Lucent pointed out full menu access wasn't supposed to be enabled on those ports. 'tho there were only 3 people in the entire company with access, I was the only one who knew how to use any of it. It's a good think we never received any tap orders.)
I’m sick in bed today and this was a great distraction! Thanks!
Finally! I have been waiting for this for years! I can't wait for the 3ESS to be back into full museum service! Thank you all for all of the hard work!
Need to put together a check list like an aircraft pilot uses then publish it here for the rest of us to follow. Thanks for the video.
Thanks so much for doing this! It is interesting that these were designed when the control console was a teletype terminal and that the machine can't talk any faster than 110 baud.
Oh, this is absolutely beautiful to see!!
Funny to think, this process was intended to be done only a few times ever at most, if even that, and it has to be done each time the switch is to be brought into action. Tedious in a way, but certainly magical to do -- to boot such a computer in such a hands on way, all for it to handle phone calls.
I have to be that guy -- when might we maybe be able to make a call into or through this beauty, or will that likely forever remain a :"need to visit" kind of thing?
that would be amazing! i'd like to know the answer to this as well
Even if we could just call and listen to a recording from there, that would be so awesome!!!
I spent a lot of years working in Telephone central offices. The sound of those power converters and the smell of the RH power cables are very familiar to me. That and the incessant relay clicks.
Wow, cool to see this again, I used to install mainly the 5ESS and did a lot of tearouts of 3ESS and others along with crossbar and power.
Sensacional, sou fascinado por centrais telefônicas antigas!!!! Brasil!
I work on cutovers from 1XB & 5XB to 1AESS in 1982 and later on with 1AESS to NorTel DMS-100
Whoooooooa, I've been missing you for so long :)
Love the 3ESS tape emulators - modern electronics coupled with antiquated systems are really my thing... this said by someone who built a few Raspberry Pi based controllers for a Monotype composition caster.
I've seen that done with an old PDP 11/34 since they no longer could get disk packs for it. The room seemed so empty without the fridge sized boxes of disk platters. I felt sad when I saw outlines on the floor where they used to sit and knew what was missing.
@@PWingert1966 the mechanical aspect of old time computers definetly add a certain charm to them.
The transition to new CPUs is like replacing Prague astronomical clock with just a digital display. Yes, it's more precise, smaller, but it has lost the charm, the feeling that if you try enough you might understand what's doing what.
This is so insanely nerdy in the best way possible. Love it! :D
Remembering all the commands and how everything works is a real skill set very few people have
I wish my brain was better
I can't remember things like this
I try to learn new things but I just can't get it to stick in my brain
Same thing with another language. I couldn't get it to stick either
It is good there are people like you in the world
I think the university here still uses the 5ess.. as an analog backbone.. still used for some feeds and the national weather service and emergency still uses it for clock and broadcasting relay... crazy huge mainframe with 5 levels of entrance cables feeding the campus.. most are cut or repurposed but still have the glass batteries and switchgear running 24/7.. gotta use that old software to set up..
Its amazing you have all those steps memorized. Wow.
Ammazing. I really like this kind of installation, driven by simple mini-computers. Lots of electromechanics driven by a small, limited, slow brain. It is very interesting to understand how things were designed with such (very) limited computer ressource. Thank you for sharing all that.
I hope this process they're describing wouldn't involve fooling with live electricity. As I am sight challenged, I'm really fascinated by machines that have to do with telephones. I hope that one day I can come to this museum.
My dad worked for AT&T from the 60's through the 90's. We had a local mom and pop telephone provider in Califon, NJ who had advertised to come see our new switch gear one time . Once he saw it he knew it was an old 3ESS system they had dumped.
This is incredible. Huge props to y’all!
As someone who works in modern telecoms and sees this equipment retired in place all the time, its really cool to see it in action.
I used to commission Philips EBX8000 PBX’s all over the world. Although these were a little less complex, not much so: they could have up to 8,000 extensions. Dual processor reed relay switch under stored program control, loaded with paper tape. Operated in hex with 16+ switches. Remarkable how common many of these concepts and procedures are!
And it used to take me an hour to boot up an EBX8000: these too were designed to stay on.
Wow, that's quite a process and a quite a priviledge being able to do this!
sure is. When I think of a dms , the only thing you did on those was turn on the power and then it would boot itself and load all the peripherals
The 1AESS was eventually upgraded to handle SS7, I wonder if the same was ever done or proposed for the 3ESS
I guess this is how non tech people see IT people. I don't even know what I don't even know. But all of this is seriously impressive, how everything is analog as hell but still digital. And it looks like something out of a cold war bunker.
Does this switch have coin control for pay phones? That would be cool to see pay phones connected here and have coin return and coin collect work properly.
Yep! That's on the list of things that we're planning to diagnose and fix.
so. freaking. cool!! I doff my cap to Astrid and Colin for work getting this piece of history working again.
What a romp down Nostalgia Lane. Thank you.
That's really cool, I imagine this is similar era as DMS10? We still have lot of DMS10's in operation, this kinda reminds me of them. I've never had to INI one myself, I'm just a NOC tech so we don't dive too deep.
Similar era, yep. The DMS10 was miles ahead of this switch though, and this one never really took off. The DMS was cheaper, more advanced, and (as we now know) lasted well into the 21st century.
The 1A ESS and 2B ESS, which served larger towns were both very successful. I guess there was just a better cost/performance ratio there.
I love that this museum exists. Part of it is that for part of my brain it is fascinating. Another part sees it , excuse me, as wonderfully absurd. I wonder if it is staffed with Logopolitans who at night as a hobby do block transfer computations.
I am looking forward to visiting in the fall. Thanks for being there and posting these videos.
Do you know how much of the difficulty is normal? How much is due to flaky hardware you haven't figured out yet?
Boy does that bring back memories I worked for Verizon for 35 years and spent a lot of time on a number 3 ESS switching system
Here in Toronto near the Eaton center bell had a building right beside it and tucked away in the basement behind a glass wall was a fully functioning switch that was for the building and likely the downtown core. I managed to grab a few pictures for a project I was doing at the time. A couple of years later they frosted the bottom five feet of the glass wall. It made me sad to know what was in there and that it would never be visible again. A few years later they dismantled it. and the room's lights were turned off and then the whole section of wall was replaced with Gyprock. I would love to learn more about that if anyone has information.
Thank you for all the shares - I’d love to visit, omg!!!
Back in the late 80s I replaced one these with a DMS100 remote in a small community outside of Witchita KS. I dont remember it looking like that. I had to hook up the Nortel Board to Board tester no a no test trunk on the #3. Took me forever as none of the local techs seemed to know much about them. Finally got it to work and that was the last I ever saw of those machines
No clue what is going on here but seeing two wonderful geeky women like me working and nerding out is awesome
This is mind blowing attention to detail !
yaaaasssssss Ive been waiting so long to see this video! thanks for making it!
Is there a resource or manual describing the terminal interface for the #1 and #3 ESS digitized somewhere? I'm really curious and would love to learn it.
Very cool. Hope I can visit some day
I love when massive ancient computers are hooked up to small modern computers running minicom or something because genuine teletypes are a pain to obtain lol
12:54 - "...taking a circuit diagram off the edge of the page"
Here be dragons.
I'd like to see the boards in the 3A CCs themselves. I understand the 3ESS is a very rare machine, built for small offices and not produced in particularly large numbers.
Wow, I'd love to learn more about this system! The command format is intriguing.
It's called TL1. I'm sure you can find some generic documentation that describes the format and what everything means. A lot of words are abbreviated so that they can be transmitted quicker over slow modems.
This going back previous century. You guys are like Doctor Who(s) in the Tarsus.
I was a test engineer at OKC Works where I ttouble shot manufacturing defects on #2 ESS processor and AIS system. Used to design machine code programs and hand mag twister memory cards as trouble shooting aids. Lotsa fun.
Would like to see a 2ESS processor video because I would have likely tested it back in early 70s.
We have a 2B ESS in storage, but sadly no room in the museum to move it in and display it. Maybe someday!
How the HELL does one even begin to learn all this?? I bet the two people who work on this are the only 2 out of approximately 7 BILLION people on Earth who could get this up and running...that's pretty amazing if you ask me.
".... and then it will usually come up cleanly" is a simple phrase encompassing much potential pain.
I'd really enjoy more videos about the 3ESS :D
Interesting .
At the end of boot , will show Arnim Zola ? (I already have trisuit from Switzerland)😐
Wow! Reminds me of the days back when I worked for MobileComm, a Bell South Company, rebuilding voicemail machines. The VRS in particular was great when it worked, and pure hell when it didn't.
Because this is a *3*ESS, and not the later 4 or 5, those processors are custom logic, not the 3B20D that would drive the later switches, right?
Right. They're known as the "3A CC" or "3A Central Control". They were designed specifically to control a telephone switch, and are not very good at doing math, or more general purpose stuff.
It's ALIVE!!
I have a working ASR33 that I am looking for a home for. I would gladly send it to you. It has an RS232 interface and worked when I ran it about 2 years ago.
Thank you! Since this video was posted, we have a 35 ASR as the console and its working well :)
What is the relay that you hear clicking on the background the whole time?
This is the system that replaced the step by step 355a office that I took care of in 1978 for southwestern bell. This was quite a change. Thank you so much for this interesting video, brings back alot of memories. If I ever get to Seattle, can I see this system?
Yes, you can see it, this museum shares space in the Century link CO. @ 7000 East Marginal Way. I think it is called The Connections Museum and they are open to the public.
@@aCraig-s3s excellent, thank you so much.
@t13 fox ... I had to laugh when I read your Comment. I remember SXS from back in the day with GTE...I started as a Courier and remember taking a break in a SXS switch while the Bears and Packers played...
Everything was dead silent.
... until Halftime and that Rural Central Office erupted with an imitation of the D-Day Invasion and I thought I would go deaf!
@@brunobiava4833 I remember that.
Fascinating video. Looking at the console, 110 baud is painfully slow.
WOW AMAZING TECHNOLOGY FOR TELEPHONY
this was really interesting, is there any further reading on the 3ESS available online? particularly on its processor?
Here’s the description and theory of operation documents.
telecomarchive.com/docs/bsp-archive/254/254-300-110_I1I.pdf
telecomarchive.com/docs/bsp-archive/254/254-300-120_I1.pdf
@@ConnectionsMuseum thank you!
@@ConnectionsMuseum There are... BSPs. Why am I not surprised. :-)
Do either of the Sam's have those in their libraries?
@@baylinkdashyt Yeah, the link I posted above is to Sam Etler's library.
Is Astrid still working in the Museum? I havent seen her for a long time in a video. :-(
Did you mention that the two boxes that take 10 minutes to boot (with the marching lights) are trying to 'come into phase' with one another or another system.?
Fascinating!
I worked on 1A for years. I think a few 3ESS guys came over to 1A.
You might as well add a turbo encabulator or does this model already have one?
Will you be making a D4 Video?
Curious to know what parts are pure digital, pure analog, hybrid
What is the oldest computer you have? Also, do solar flares affect some of the older units?
Super cool !!
So awesome to see this stuff running, it would be neat to have an actual teletype connected to that serial port too. Did the 3ESS have support for remote management over a modem?
Sarah here -- I think Astrid and Colin eventually want to hook up a real Teletype to the local serial port, at 110 baud, and then have a remote session via a modem to one of the other serial ports. That should run at 1200 baud, I think.
@@ConnectionsMuseum A Bell Dataphone modem would be perfect for this, how many serial ports does the switch have?
@@KJ7BZC I believe this is discussed in this or a similar video, and you can clearly see several sleek looking black rectangular Dataphone modems buiilt into the backside of the rack where the terminal resides.
Why does this give me like DEC vibes? I know Bell built all their own stuff, but that looks awfully similar to regular mini computers of the day, specifically a PDP or the IMSAI 8080 🤔
or Data General Novas
The later 4ESS and 5ESS switches had 32 bit processors, which - someone correct me - ran an RTOS called DMERT. Bell labs also produced a series of commercial computers which had the designation 3Bxxxx which used similar technology. The 3B series ranged in capability from a desktop unit which could handle a handful of terminals in a lab to large systems which could run a whole business office. Early machines such as the 3B20 were 32- it word size built from 4-bit wide AMD 2901 bit slices in the 1970s. Later systems incorporated the Western Electric WE32000 processor chip. Many of the 3B systems were fairly powerful and equated more to a DEC-10 system than a PDP-11.
Side note: After about 1977, many PDPs got rid of the toggle switches in favor of a keypad and seven segment LED digits displaying in octal. Only a very old PDP-11/34 I worked with early on had the front panel toggle switches. The old machine required manually keying in a start address and executing a jump instruction on the switches to execute the bootstrap to start loading RT-11 O/S (in our case from an RK-05 disc). I later worked for a university where we ran System V UNIX on a DEC Vax. I got a chance to meet a few of the AT&T UNIX old timers: Mark Horton who contributed to the vi editor and ditroff and Perter Honeyman of Honey DAN BER UUCP. We were the first site outside of Bell Labs to run the revamped UUCP. We kept a couple of Telebit Trailblazer 23K modems busy pretty much 24/7 transferring our USENET feed.
I personally had an AT&T 3B1 aka Safari PC7300. The 3B1 was actually made for AT&T by Convergent Technologies, was based on the Motorola 68000 CPU and ran System V r2 UNIX. The 3B1 had no commonality with the 3B2 series other than running a version of UNIX. The 3B1 could have up to 4 megs of RAM and an 80 meg MFM drive. I also had a cartridge tape and Voicepower board which I got at swap meets. The Voicepower was really cool because it had a high quality CODEC and an entire IVR system could be written with Bourne Shell commands. Graphics were green screen Hercules compatible 720 x 384. The 3B1 was a pretty powerful desktop for 1984, but it didn’t sell well and AT&T started liquidating them in a so-called fire sale in 1986. The 3B1 is sometimes blamed for Convergent Technologies demise.
When I build 3ESS in the late 70’s we used a PDP 8 computer to test the system by making simulated telephone calls. We would run the systems for days at full capacity before they would pass.
this was amazing!!! congrats!!
"Good Morning"
"Good Morning"
"What are we doing?"
"Being awesome".....
Wow. She’s like a walking encyclopedia!
Neat stuff, thanks for your hard work.
What switching equipment did Ma Bell use on its teletype network?
I hope all the information about starting up the system is documented somewhere. It is an incredibly complicated procedure that you won't be able to do by playing around.
Can it run crysis?
That is just amazing.