If you haven't watched Part 1, go do that for more context. :) • How does call tracing ... A Bell Labs Record article on the trunking and switching network of the 5XB: archive.org/details/bellsyste...
Thank you so much, Sarah, for planning these videos, recording them in high quality with good sound and lighting, editing them well, and sharing them with all of us. I have seen every Connections Museum video and I love them all. I hope to visit some day but it’s a long trip from Asheville, NC to Seattle. Circuit switch networks are super interesting to me. Wish I could have my own switches. Y’all are amazing!
Great stuff, brings back memories. In the late 70's I was an apprentice Telecoms Tech and got sent to a crossbar exchange for work experience. Of course I got the jobs that no one else wanted. I got shown call tracing and it was totally confusing. But then it clicked and I enjoyed it. So my job ended up as either mainframe jumpering or call tracing. But I enjoyed it. Still miss the smell and sound of a loaded exchange.
The sheer complexity of the nature of telephone switching cannot be understated -- the logic is just so incredible, it's hard to think people actually came up with it -- let alone thinking of people having to back-trace a connection through the amazing tapestry that is telephony switching fabric in the pre-digital era.
As bonkers as a trace is, one must remember how many people the bell system family employed. Each tech was trained to do any job and then was dispatched first in, first out. On a call trace, for police activity, the system would have a different tech perform each step you're preforming. Each would yell out to the next tech where the call had been traced back to, then that tech would preform the task and yell out the results to the next tech, and culminate at the switch panel to hand the trace off back upstream, untill the origination number was known. It took them something like 1 1/2 minutes to trace a call locally; so, it's a lot on one person. But in a large switch, there were plenty of techs for a trace. Cool video, Sarah. Thanks.
Oh yeah, so do I! Sarah is a genius and she explains it all with Fran-tastic clarity and a down-to-earth approach, which I also try to do on my own channel. CM as in ConnectionsMuseum or CuriousMarc, ha! I absolutely love both.
I remember doing a trace in a TXE4 crossbar exchange in the UK when an emergency services 999 test call failed. That was whole bunch of unwanted fun! Super video.
Another brilliant video Sarah. Your ability to take what could be quite a dry technical subject and make it fun and engaging is excellent. I certainly came away from this one with a huge deal of respect for the Bell System engineers who were working on these amazing pieces of equipment daily.
Sarah; I worked for about 10 yrs in a #5 office in So Cal, the number group and translator frame were located on the 1st floor by the IDF, first floor was SXS and T-Cxr. When frame people ran in new circuit orders, the last step was a Line Verification test, performed with a box that hung on the wall, and had a bunch of switches numbered 0-9. You could put in a telephone number, and it would tell you what LL, or put in a LL and it would tell you what telephone number on a panel lamp read out. Frame people did not have access to the #5 switch on the second floor, and were never allowed to touch the MTF. I remember this very well, because my first trouble shooting task after completing the 8 week #5 school, was to determine why the TRC was dropping almost continuous cards, turned out that someone had set the switches for a LL that we didn't have in that office. Also, frame people could dial a test number from the cable pair, and get an audible read out of what telephone number was on that ca pr. I think this was called automatic number identification.
We have one of those line verification boxes too, but I don’t think we have the interface circuit for it. I just do all the tests from the MTF cause we have no such restrictions ;)
Sarah, you are so COOL! I thoroughly enjoy your instructional videos. In the late sixties, as a WECO installer, I had the good fortune to be exposed to the No. 1 ESS. Having a natural curiosity to begin with, I wanted to know how this thing worked, and found myself driven to understand the machine. I was certainly aware of the #5 Crossbar, but had no idea how it worked. You do a fabulous job of explaining it in a way that's understandable. Thank you and keep doing what you're doing.
Thanks for the refresher course, much appreciated. On your trouble recording card, there is a punch called LVM. ( line verification match ) You should be able wire up the number group to drop a card when a call is terminated to the subscriber! It’s been 40 years ago. After you do a call trace a couple times it becomes easier. Again thanks for the refresher course.
That ending with the Trouble Recorder working was hilarious, especially since you had already dropped a couple cards. Me watching the vid "Oh, Sarah's got the trouble recording working again neat" then at the end "btw..."
Hey Sarah: Loved this, brings back good memories. I worked in a full 30-60 office, 60 line links, 30 trunk lines, triple trunk links,paired line links. The triples were 0,1,& 14 on vertical 0 of the line link juncture switch, the 2,3 & 21 vertical 1, etc. There was no 2nd juncture group, so tracing a call was straight forward. Level 0 left on the trunk link juncture was line link 0, level 0 right was line link 01 & so forth up to line link 19, line link 20 to 39 was on the extended juncture switch & 40 to 59 was on a second extended juctir switch. Easy to trace a call by memory. Common language was PTTWPAPTMG0. Cut over to 5ess in 1985. Was happy to send you a pack of trouble cards I had! L
It's been over 40 years at least since I had to trace a call through an Xbar switch and the #5 switches are wired a bit different than a #4 as we did not have that split in the middle of the switch then again we were a 4 wire talk path plus signaling thus switching 7 wires per trunk so I guess we needed more room but the idea was the same with the in-link, junctor, out-link arrangement and I always had that same dyslexic number switching problem too, always had to write each step down or I'd be starting over. I still now and then have a nightmare involving a trouble recorder if full overdrive during some massive malfunction. Our TR had bars you could fold down and just put the trash can under the thing and let the cards fall in as you were busy trying to fix the problem and had no time to babysit the trouble recorder from filling to the brim. Talk about Panic, had more of those than I want to remember. Xmas, Thanksgiving, and Mother's Day were mandatory work days, all hands on deck for Long Distance, we got hammered and stuff would break. Trunk selection was done in the Foreign Area Translator boxes, FAT Box. where cards were dropped to direct the Markers to interrogate an Outgoing Group of 40 trunks and if busy there were 3 alternate routes possible so that box would have to drop 4 cards in a row on one call to a busy cities trunk group so just physically pounded the boxes to pieces. We had to take off all the sides and put up as many big fans as we could find in the building as the magnets on the code bars were so hot they would burn up. I can remember punching those cards out on overtime, everyone hated that job, tedious, but Double Time and a Half would find me punching, could not pass up that free money. Oh we made so much money back in the 1970's. Many of us worked all available overtime back then. Could buy a brand new Porsche but had no time to drive the thing, so the wive made out car wise as I drove the old VW Bug to work.
Hi Sarah Great video. I worked for NY Telco from 1961 to 1969 as a switchman. Our office had an A board, # 1 Crossbar CO, 2 # 1 Tandem Offices, N Carrier and T carrier. We added ANI in the late 60s. Of course we also had a main frame. I started as s frameman, went to Installation and then went to a # 5 CO for a cutover and a foreman took a liking to me and made me s switchman. I did the OGT for 6 months till I went to #1 char school. Did many a trace. Wish my CO was as quiet as yours. No one realizes why Telco people are deaf. Your videos are great and little by little I am getting my knowledge back. PS hated the trouble recorder
On TXK3 we were lucky enough to have an ID facility that would print out the call trace details on the incident recorder. However, to do it long hand quickly took a few years of experience. I also worked faulting microprocessor pcb's, but crossbar was the most difficult system I ever worked on.
I'm not quite certain which is more fascinating, trying to grasp the technology itself, which you are very good at explaining, or marveling at the grasp this technology obviously has on you! Wow!!
Another great instructional video. The more you show us, the more one can appreciate the geniuses in Bell Labs and Western Electric who figured out how to build the biggest machine in the world. Thanks much.
I very much appreciate the level of detail you provide in the videos. I've watched several of them thus far and have enjoyed each one. I've always been fascinated by machines, especially older mechanical systems such as the switches you talk about. You did a really good job of explaining how the machine sees the cross bar switches and calls get traced. I can only imagine how difficult that would be in a live central office. I look forward to watching more of your videos. Thank you for making them!
the value of "traver tr" and "vfy:ofc" are absolutely understood from this video. I have traced connections through SMAS starting at the Stage1 all the way through, but actually tracing a call on a Crossbar like this, wow. Thank you!
Excellent video, thank you so much for showing us this. I traced quite a few calls through a no.1 SxS years ago but not quite as complicated as no.5 xbar.
I LOVE your enthusiasm Sarah! I understood the panel switch tracing but the number 5 is just insane! Keep up the awesome vids :) Maybe Part 3 should be tracing on the 3ESS?
that is so amazing to see ..it would be easier to solve a skooby doo mystery than trace a call ..it is a network ..so wild to see ..so in a large city there would be multiples of these sites ..we all talk about mother earth ..but damn ..how about that maw bell :) and the stuff that melts my mind even more is ..they are freaking billing all that in real times as well ...i was a 70's kid and had no clue the tech it took to make that giant wall phone ring ..so kool to see it now :)
Sarah. I appreciate all the videos you have produced. Ii ask that you do one showing how a #5 handles the call from when a subscriber picks up the phone, through all the switches, to connect to another phone in the same office. Especially how the system decides which relays to use to connect the two phones. Thanks
This is a fun and informative channel. My exchange experience was in New Zealand. I ranged over manual, rotary (British system, I think), SxS (Stronger), NEC crossbar, that's enough for now. Do you want to do a trace thru a SxS switch to make an old man happy. Cheers, Richard
...and the police screech up to find a pay phone with a dangling handset. NOTE: Those born after a certain year will need to look up what a pay phone is.
Yay. Did some mods to a crossbar in the early 80's. Bit of a punishment job. A day on one step of the ladder only to move up a step and so on. It was also the time I left to go live on a croft.
Hey, im over in the uk, and i noticed something about telephone boxes in the 1990s, early 2000s I dont know whether this was a fluke or not, but the last 6 digits of the phone number of the telephone boxes/payphones, relates directly to the grid squares on an ordinance survey map, well, the ones i used did Just to give you an idea if you didnt know, we have an area code, which is now 5 digits long, followed by a subscribers number, which is 6 digits long, and the subscriber number for the payphones matched perfectly to the ordinance survey map So you could physically trace an actual location from a phone number, and a bit of area knowledge If someone called and they were lost, you could ask them what the number of the payphone was Then you could look up the area code to see which town they were in, flick through a map book, find that town, then use the subscriber number to pinpoint their location on the grid squares of the map
So the phrase, “Hurry, go trace this call!”, was the same thing as saying, “Hurry, get me some Unicorn blood!” It’s astonishing that people would sometimes come back with Unicorn blood.
Sarah, you talked about the ringer signal how about the REN of each phone and how it relates to the operation of the ringing of the phones at one lacation?
Sarah, you do such a good job of explaining things. I had a little knowledge of tracing that I picked up from various switchmen over the years, but you tied it all together. And I’ve never seen a panel office in operation so that was all new to me. Now, for the big question: how has call tracing changed with the 5ESS? I’m guessing it’s just a matter of a few keystrokes on a terminal?
How long would the caller have to stay on the line to be successfully traced? There's a scene in a late 70s - early 80s Finnish TV show (we did and do have some differences in the telephone system) where the murder squad officers are chatting and hear a banging noise from the chief's room. Another bang and they open the door to find both parts of the local phone book lying on the floor and the chief talking very slowly and deliberately on the phone, with a scribbled note on his table: "TRACE THIS CALL". The line does have some additional equipment on it which makes it quicker to trace, but they still don't get it down to the subscriber line, only within a few possible exchanges. (This from memory, I may get some details wrong.)
Throughout this video, I kept wondering how billing for calls on these switches was handled. I believe even local calls were billed in the years this switch was a part of the telephone network, so I'm assuming there had to be something that collected at least some information about calls placed from each subscriber loop. Guessing it was far less information than you see in SMDR. Maybe a topic for another video?
Right there with you on the dyslexia thing. I'm a software developer and my colleagues have never quite "got" what's going on with the notebook and pen. How did you learn about all this stuff? Y'know considering it's age and your youth? It'd be interesting to hear how you got where you are today.
Another film reference for a call trace scene, in this case on a Step-by-Step switch: "Black Christmas" (1974). ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-tbK2GfN-E18.html
If there's one thing I've learned from these videos its that maybe my softswitches aren't so bad after all XD. It takes me longer to get connected to the correct network than it does to pull a CDR/trace out of our softswitches. Though having said that, it wouldn't be nearly as interesting as doing it this way.
Hated doing back traces in #5 X-Bar. Step was much more easy to do. Would be fun to see it done in a Lucent 5E or DMS 100 switch. On step the trap ckt. just had to ground the sleeve.
So I have never worked on telco/carrier scale switching but I have experience with early Lucent/Avaya Definity PBXs which if I remember correctly shared a lot in common with a small 5ESS. Tracing a call on those is just a matter of asking the computer to recall the call state, then everything just gets printed. I would assume the 5E is pretty similar (just at a much larger scale than a Definity) and likely the DMS 100 too. My how far technology has come, haha!
Sarah you lost me, my mind woke up from its trance when I saw you using an LED flashlight instead of that wonderful antique light that you used in episode one.
What would happen if after tracing the call you find out that it is one of the dozen subscriber lines belonging to a hotel PBX or commercial customer and they are uncooperative in further tracing the call to the correct extension at their premises?
Essentially, this implies the phone companies back in the day had employees in the exchange, who potentially blindsightedly followed with the authorities and participated in the tracing of calls at their request, without any regard for a warrant or judicial oversight. Honestly, I don't think things have changed. The government could easily initiate a trace without oversight and scrutiny, perhaps even more so given the laissez-faire attitude of governments with regards to oversight nowadays and the lack of human oversight amid the huge amount of data pouring through telecommunication systems just to claim a media victory and make themselves look good.
Right when I'm thinking that my dyslexic brain would be flipping and flopping with this, you tell us about your jumbled brain. You must love a challenge. My hat is off.
I'm surprised they didn't implement a way to just pull this info from the marker using the test frame, i.e. you enter the subscriber or trunk line and it'll drop a card telling you what is currently connected to it. It feels like that should have been totally feasible, but I guess it just wasn't important enough for day to day use to waste relays on it
That's basically it. It was much easier to train someone to do it, and it was done constantly for tracing trouble. When you got good enough at it, it was just like reading. The difference is, when tracing trouble, you have the luxury of time, and there is less urgency than "THE KILLER IS ON THE LINE NOW!" like in the movies. The marker knows when a channel is busy, but it doesn't know *why* it is busy or who is holding it occupied. That's a much harder thing to do. There are many problems that are "easy" to solve in one direction, like "set up a call on the first non-busy channel between point A and B. And there are many things that are harder to do like "scan each of the thousands of channels in the office and find the one that is connected to this particular trunk". That's the kind of problem you really need computers and memory to solve in a fast, cheap, and efficient manner.
