Talk:Panel switch

Information and queries

I have access to a functioning panel switch at the Museum of Communications. Though I can't guarantee any sort of timeline in the future, I will be there on 20-Mar-2007. Drop me a line if you want pictures or audio recordings! The Slimey 23:09, 3 March 2007 (UTC)

Splendid. I only worked a few weeks in a panel office in the late summer of 1969, learned little, and forgot most of that. Mostly they had me pulling wires on the MDF. See what kind of detail you can get on the clutches and the, darn it, I've forgotten the name but the hand that grabs the terminals. Senders might be nice, too, especially if they've got a mix of flat spring and wire spring like I saw in 1969. Oh, and the trunk test panel with jacks. Lots of interesting parts. Jim.henderson 00:28, 4 March 2007 (UTC)

I've posted what pictures I took at [1]. There is a wide and narrow shot of a sender (I think that it's a sender ...). In this office, originally installed in Seattle in 1923, all the relays are flat-spring. Also a wide shot of a district selector frame, and some (dark) detail on the selecting hand things. Unfortunately there wasn't much light and my camera isn't that good at macro photography. If you think any of these deserve to be on Wikipedia then tell me which and I'll upload them. :) —The preceding unsigned comment was added by Slimey.limey (talk • contribs) 16:40, 26 March 2007 (UTC).

The last picture, of the selector frame, is good and I'd like to add it to the article. The first two don't indicate what kind of subsystem it is, but they are the only pictures we've got of flat spring relays, so one or the other would make a good addition. The others show poorly on my screen and presumably on others. Thank you; two pictures that actually illustrate something clearly, out of five, is not too bad a record. Jim.henderson 03:56, 28 March 2007 (UTC)

I've uploaded it. I think I have a better image, so I'll go look in my images directory (uncategorized :( ). The Slimey 07:16, 28 March 2007 (UTC)

Oh no, I don't know how to include a Wikimedia Commons image in Wikipedia! It's at [2]. The Slimey 07:20, 28 March 2007 (UTC)

So, it's in the article now, and a nice shot. I didn't know how to create a picture link either, so I went to the telephone plug article, stole the picture reference from there, pasted the panel picture's name in place of the plug picture's name, and deleted the size limitation so any user will see whatever size is set as default in "Preferences". Perhaps you would also like to put in one of the pictures of flat spring relays, but anyway, thank you for the picture. Jim.henderson 05:13, 1 April 2007 (UTC)

Can someone explain what a junctor exactly is? I've created a stub article called "junctor", please fill one or two paragraphs into it. I've wikified about 6 different pages (1ESS, Number One, Number Five, Panel, etc), none of them really clearly explains what a junctor is. Mdrejhon (talk) 23:23, 27 June 2008 (UTC) The Panel office does not use a junctor persay. Rather what holds the originating side of the call is a frame called the District frame which if maxed out could have 60 selectors, 30 per side. This frame also detects answer supervision from the trunk used to make the outbound call. This answer supervision is detected by the reversal of the battery and ground on the trunk. Hope this helps. Panel Switchman (talk) 21:17, 21 December 2008 (UTC)

As I understand it, a junctor is simply a path between frames in the same office. I only know about crossbar junctors, not panel or ESS. In fact I have never heard of them when talking about panel offices.

You may want to read Survey of Telephone Switching. It's not perfect but it explains a lot of background like that.

Crossbar and panel offices also have links. Survey of Telephone Switching defines the word as it applies to panel. In crossbar offices, the word means something different. It's a path between the primary and secondary switches of a pair. The reason for these pairs is interesting and is not explained in Survey of Telephone Switching. They allow you to build a bigger switch out of smaller ones. Say you have two 2x2 crossbar switches P1 and P2, and two more S1 and S2. Then you connect a row of P1 to a row of S1, a row of P1 to a row of S2, a row of P2 to a row of S1, and a row of P2 to a row of S2. You've built a 4x4 switch. That's a toy example, but when you build a 100x100 switch out of 10x10 ones it becomes more interesting. Dpeschel (talk) 00:07, 14 July 2008 (UTC)

I just erased ", differing in that the caller had to dial the entire number before connection setup could begin" from the article but I would like people's comments about that. Does anyone know of a real panel office that needs seven digits to start setting up a connection?

The Museum of Communications's switch is the one I'm familiar with, as is Slimey.limey, and it starts routing after you have dialed the three-digit office code or 0. The office code gets you a route out of the district frame. There are no office frames hooked up. The incoming and final frames are arranged as described in the article, so the sender needs seven digits to finish routing. Unfortunately there is no ten-digit dialing (the auxilliary sender isn't hooked up) so I don't know how it works.

You do need all seven digits to get a ringing or busy signal, of course, but that's not how I understand "connection setup" or "begin". Dpeschel (talk) 00:30, 14 July 2008 (UTC)

Hello All; Re the Panel switch: I have worked in Rotary exchanges in New Zealand c1960; the Rotary system was also developed by a group (under FR McBerty) in Bell Labs and was manufactured in Europe (Antwerp); the factory was sold to ITT in 1925 under an AT&T divestiture agreement. The Rotary system also used switches and sequence switches driven by row motors, but the switches were smaller (200 outlets). The linefinders served 60 lines (7A version) or 100/200 lines (7A1). Finally, some questions:

When was the first Panel switch (at Mulberry NJ) cutover?

Re the paragraph on “Telephone numbering”, while there may have been a city where manual offices or party lines required the use of five digits for an office, I would have thought that it would be preferable to have all office numbers of four digits only if possible?

Re “stepping” groups (PBX groups) of customer/line numbers, did the Panel system have much flexibility for grouping numbers (like crossbar/electronic switches where any number could be the first listed number of the group) or was the allocation rigid like step offices? While rotary offices could have 20 numbers in a group as against 10 in a step office, it was necessary to start a 20 number group on the first number picked by the final selector eg 48860 so that the selector would hunt over 48860, 48869 … until it found a free line. The article on the British Director telephone system suggests that Panel offices had greater flexibility because the call was register-controlled, but that did not apply to Rotary offices. Hugo999 (talk) 12:35, 7 September 2008 (UTC)

Once I get a good understanding of this Wikipedia I would like to add a section on the Link Frame and DSS (District Sender Selector). In Battery Cut Off offices the Link Frame and in Ground Cut Off offices the DSS switches sit between the District circuit and the Sender. They basically do the same job. When a caller goes off hook the District tries to find a idle sender. When an idle sender is found and is ready to receive digits it puts out dialtone. If after a period of time a Link Selector or DSS switch can not find an idle sender an audable alarm is generated and the associated isle alarm lamp comes on. The (Link/DSS) allow a chosen idle sender to collect the digits dialed and setup the outgoing call. Once call setup is completed the sender drops out of the connection and the District circuit becomes the controlling device. Panel Switchman (talk) 21:35, 21 December 2008 (UTC)

1970s photos

Hopefully I will figure out how to upload photos as I have some from a working Panel office taken in the mid 1970's. Panel Switchman (talk) 21:35, 21 December 2008 (UTC)

They will be very welcome. It's not terribly difficult to upload a picture once it's in your computer. Hit "Upload file" in the Toolbox on the left side of the Wikipedia page. I upload over a dozen photos per week, and send them to wp:commons so others can use them besides the English Wikipedia, but that's no big deal if you've only got a few. Jim.henderson (talk) 15:50, 25 December 2008 (UTC)

Finished adding photos of Panel CGO Line Finder brushes, rods and banks. Added photo of Panel CGO Line and CutOff relays found on Line Finder frames. Added photos internal view of a Panel Sender with one having a hand to show size. In the future I have others to upload. All photos were taken in the early 1970's. Panel Switchman (talk) 00:11, 7 January 2009 (UTC)

Very good, and thank you. In Commons yesterday I did a little housekeeping work, mostly just grouping your pictures into a new Category. Not everyone needs to know about Categories and other fine points of organizing Wikis as long as someone who has the material is entering it, and someone else who knows Wiki ways can Wikify it as it arrives. Keep 'em coming, if practical. If pictures become so numerous that they threaten to overwhelm the currently simple structure of the article, all you need is someone who knows how to restructure it. Namely, me unless some other experienced editor takes an interest in helping. Jim.henderson (talk) 05:28, 7 January 2009 (UTC)

Added "Panel Office Typical OGT" photo. It's location may want to be moved to a different paragraph in the future. When the photo is brought up there is a detail description of its primary functions. This text may want to be moved into the main body at some point. This photo has very high resolution. Panel Switchman (talk) 21:55, 7 January 2009 (UTC)

Yes, it's an excellent addition to the article even if it was uploaded directly in Wikipedia and would have been slightly more useful in Commons. Ah, it calls to mind the long ago times when we had hair and hope, and I was chained to the 5XB OGT, looking up cable pairs in the XRL and calling around town praying that some frameman would answer, and not the guy who always stuck his KS6320 orange stick into the lugs and swore he was giving a short. Chasing opens isn't really about Panel; it's about telephonic antiquity and perhaps we can make a new article like Switchman (umm, no, an article by that name already exists) to describe routine operations, back in the day of singing repeaters and flying dinosaurs. Jim.henderson (talk) 23:09, 7 January 2009 (UTC)

Yes, I agree to some of what you say but this OGT was in a Panel office. OGT's from SXS and Crossbar generally looked very different. That was one reason I kept the text of its operation with the photo itself. I think that photo came out very well. I have two other views of an OGT stashed away. Ah yes, Orange sticks, Flats, Needle Nose pliers, Heat coil pullers, Sandpaper on a stick (for really old cross connect wire with cotton, silk and enamal coverings), Wire loops, Wire cutters with that "notch" for striping the insulation. Singing repeaters, remember that test set used to detect singing repeaters? There are more photos to come. OK what do I have to do different when uploading to put a photo into Commons and then to reference it?? Panel Switchman (talk) 17:10, 8 January 2009 (UTC)

The Sender and Stuck Sender areas are getting kind of cramped for space. Any ideas how to reposition them?

I have additional photos of Panel Battery Cut Off Link frames and Panel Ground Cut Off DSS switches. These tie the Sender to the District during call setup. Panel Switchman (talk) 01:12, 9 January 2009 (UTC)

All righty it's coming along; lovely pictures. I moved the old crossbar pic to 5XB switch where it belongs, and arranged the others in one single column. This is an easy way to keep the pix from interfering with the section headers, though it has the lesser disadvange of separating pix from their particular texts, depending on browser font, screen width, and other variables. The picture a that shows a human being, I moved to near the top, regardless of what part of the article it's trying to illustrate.

Maybe when we see what pix we have, we'll want to use some of the other standard picture tricks, such a WP:GALLERY or two, or an alternate side scheme, or demote some of them to a {{Commonscat}}. This last method requires that the pictures be in Commons, so you should send all future pix there.

To get into Commons, you can click on a pic that's already working through Commons, for example the Seattle one at the top. That gets you to the Wikipedia Description Page for that picture, which has a link to its Description Page in Commons, which will get you to Commons. You might have to log in again at Commons, depending on your browser version. When you upload in Commons, a "Category" blank will be near the bottom of the form, and you should say "Panel switch" (no quotes) there. Goody; finally we Americans will be able to put up some competition against the profusely illustrated TXE article. Jim.henderson (talk) 02:14, 9 January 2009 (UTC)

Added photos of Panel BCO Link Frame and Panel GCO District Sender Selector and Panel Decoder Test Frame. I think they are in Commons. Panel Switchman (talk) 22:32, 9 January 2009 (UTC)

Updated initial paragraphs adding dates and true locations. Also added additional reference.Panel Switchman (talk) 01:37, 10 January 2009 (UTC)

10050

Always good to get more good pictures. While were at it, "The panel system was designed to work with manual offices of up to 10,500 lines". Any idea where that odd number came from or whether it's correct? Jim.henderson (talk) 22:15, 13 January 2009 (UTC)

I will try and find out. Panel Switchman (talk) 20:53, 16 January 2009 (UTC)

From the TCI Yahoo Group SingingWires "The Bell Labs Record of Oct 1951 describing Basic AMA Central Office features states that the wide paper tape AMA system allowed party letters to be recorded on the tape as stations digit. It also mentions a called number structure digit that also allowed 5 digits with numbers above 9999 for the large No.1 manual B boards with up to 10,500 numbers. The only place that stations digits [eight digit] or over 9999 really worked was at the manual board using Panel Call Indicator pulsing. ".

I think the operative phrase is "for the large No.1 manual B boards with up to 10,500 numbers".

It was also mentioned by some that back in the day an Operators reach from her position could not exceed 10,500 numbers.

See Panel 'B' board below (not my photo)

http://www.flickr.com/photos/9479603@N02/1815411594/ Panel Switchman (talk) 20:02, 17 January 2009 (UTC)

The above photo is for calls from a manual office to a Panel office. I have to find one in the other direction. Panel completing to a manual office using PCI. Stay tuned.

Panel Switchman (talk) 21:22, 17 January 2009 (UTC)

Again from the TCI Yahoo Group SingingWires: The 10500 jacks come from 7 panels wide of 1500 jacks high. The jack panels were 20 wide so 5 high would give 100 numbers. The jacks were 309 type, the small diameter ones, so 100 numbers would be about 1.75" high and 7.75" wide, making 1500 numbers 26.25" high. Allowing an extra inch for mounting width would give 7 panels about 61" wide so the operator in the center had to reach 30+ " each side. The operator positions were close together and they overlapped.

Hope this explains it.

Panel Switchman (talk) 21:24, 18 January 2009 (UTC)

All righty; the number arises from the way the jacks were assembled into strips in the A or B board of a manual office. Pleasant if we could have an explanation of what was done when there were 500 more lines than phone numbers. Probably in the automated version either up to 500 lines were for originating traffic only, or else some sort of mechancial adaptation was set up for a busy line hunt to a line having no number of its own. Jim.henderson (talk) 14:57, 22 January 2009 (UTC)

Added photos of both types of 'B' Boards and edited text to interoffice signalling. These were the best photos I could find. Have feelers out to get beter ones.

The opinions from the old timers is simply that Panel could only complete calls with a 4 digit number into a Panel office and the lines above 10,000 were simply an atrifact of the operators reach although not all manual lines translated to machine switching. Go figure. Panel Switchman (talk) 23:32, 23 January 2009 (UTC)

Added link to "More Panel photos can be found at 'WECo Panel Office'". This is site I have setup. Panel Switchman (talk) 23:35, 27 January 2009 (UTC)

Lay reader

Before I say what I want to, I first want to commend all of those who have worked on this page (and the other 'phone pages,' as well), as I know most of you are real phone-tech & history lovers. I'm writing this, because, though I, myself love gadgets and stuff, I'm not single-focused on any in particular, and the brings me to a problem; a huge amount of this page (and other phone-tech pages/sections), are written 'simply,' only for someone who is themselves a phone-tech person. When I tell you I'm smart, I'm being polite, but, much of this requires a person to have a background in certain areas to understand it, and that's the problem.

See, Wikipedia, is an encyclopedia a general-knowledge tool for people - for laymen - to learn things.

If any of you are old enough to remember the days of real encyclopedias,' (my mom's still got the complete World Book encyclopedia 1974!), that was the one which most people are familiar (the other being the 'Britannica'), many of us as kids would use the World Books for our school papers (use being the operative word), and it was written so a youngster could understand.

The Encyclopedia Britannica, on the other hand, was written for a more mature audience - but, it was still understandable.

Folks, the work here shows you really are passionate, but, an encyclopedia is something which lets people of all levels of understanding get a chance to learn from, and as I said, I'm far from dumb, and not a kid, so, if I have difficulty trying to decipher some of this, how about a 12 year old? Most of the things I got interested in, happened when I was about that age, and if I were that age now, I'd probable find this go to source too off-putting to read, and that's a shame.

All I'm asking is if you - as a whole - could keep in mind that this is the World Book as well as the 'Britannica' of the 21st century, and in order to learn, it has to be accessible. I'm not a phone-tech expert, so, I can't help on this one, but, I think a point I just made - that it's both these encyclopedias - the simpler one, as well as the advanced one - but, combined, in one place is key to how this (and other articles) should be written - in such a way which a newbie can come and grasp the basics, but, also, someone more mature can come, as well.

I thank you for reading this.

Thanks, UNOwenNYC. I have taken the liberty of adding a section header as is usual when starting a new topic. Alas, telephone switching is one of those obscure topics which only a small number of people understand at all, and most of us lived and breathed it for years, hence don't notice how strange is the air we breath. This results in our output being clear to our own kind, while missing our true target. Yes, I am aware of our shortcomings in this regard, especially when comparing my 1943 Britanica to my own work of a few years ago. However, it's difficult. Indeed it's so difficult that in recent years I have cut back on my telephone writing, and now work areas where I can remember what the air tasted like when it was new to me. Anyway, these years of Wikipedia writing have taught me a bit about the art, and perhaps I can find some time to apply that learning to this article, this year. Incidentally, please remember to sign talk page posts with four tildes like this: ~~~~ Fortunately, my frequent encounters with Wiki newbies have helped me remember how the obscure topic of Wikipedia writing looks to the uninitiated. Jim.henderson (talk) 05:56, 1 April 2016 (UTC)

I know this is quite an old comment, but I wanted to chime in anyway. I strongly agree that this article is (was?) too technical. However (and this is an enormous "however"), telecom technology has always been an extremely niche field, with its own terms, used to express ideas that simply did not exist in other fields. We use words like "junctor, sequence switch, decoder, cutoff relay" and the like to mean very specific things. Even if I explained to you what those words meant, I'd then have to explain why those things exist in the context of the switch. I'm a professional teacher, and I give tours of this very equipment on a weekly basis at our museum. I've explained this stuff to lay persons, and people with degrees in electrical engineering. Everyone, everyone has difficulty understanding this stuff without background context. It took me a year to understand it, and I had access to a working machine, and all of the literature. Finally, take a look at the article Elliptic curve. The concept of an elliptic curve is used in cryptography, and is critical to the security of many of the websites you visit on a daily basis. I bet that no layperson could read that article and get anything out of it. That's because its a complicated mathematical concept, and in order to understand it, you'd first have to take several university-level courses in advanced maths.

However, I will work on simplifying and clarifying the article further :) neverether (talk) 20:00, 7 June 2022 (UTC)

Panel Locations and Timeline

I'm trying to get a good idea of what cities had panel and when they were installed or removed from service. I assume that during the 1920s and 1930s, most large Bell System cities (with the exception of Los Angeles) must have installed Panel, but I don't have specifics of where, when, and how long they remained in service.

1915 – Panel switch tested in Newark Mulberry, followed by a second office, Waverly. These were “B” incoming switches only, so customers still placed calls via the “A” operators for their exchange.

1921 – Omaha, NE installs the first Panel using both the A and B sides (serving subscribers directly).

1923 – Atlanta, GA Walnut opens as city’s first dial office, using Panel

1923 - Seattle's Rainier, Melrose, and West offices are the first dial offices, using Panel.

1929 – first dial switch installed in San Francisco: Ordway, a Panel office.

1960 - Atlanta Main (Murray 8), last panel switch in Atlanta, removed from service.

1974 - Rainier removed from Seattle, replaced with a 1ESS

1983 - Last panel removed from Newark, NJ

neverether (talk) 20:03, 7 June 2020 (UTC)

Reward (talk) 23:33, 11 April 2020 (UTC)