I have a Com Key 416 primary set with an unknown operational history, and would like to verify its standard features. Thing is, I *don't* have a 91A connection block.

I assume (possibly incorrectly) the primary can be operated without the 91A by just giving it AC, and putting T/R on pins 1+26.

I've done this, and green lights work when I pickup the line, speaker and DSS/ICM modes work, and I can dial and make calls.

But.. Hold and Ringing don't work:

1) Putting a line on hold just seems to drop the call, and the green light simply goes out, it doesn't wink or give an indication the call's on hold. When I go back to the line, call is definitely dropped.

2) When I try to call the line from a cell phone, it doesn't ring, and green light doesn't flash, but if I pickup the line, I can take the call.

Regarding #2, I've verified:

* 4 switches under the DSS faceplate all in the ON position (to enable ring),

* Volume is set to '5' (SPKR and ICM work, and I can DSS the set and hear myself talk)

* Verified the LINE 1&2 vs LINE 3&4 connector on the logic board is in the correct LINE 1&2 position

I've tried using other lines, with same results.

An odd symptom: when I call the set from the cell, not only doesn't it ring, but after about two rings, the CO drops the call with an busy "error" signal; on the cell's earpiece I hear the CO give a few audible rings, followed by a double-busy signal, then disconnect.

I take this odd behavior it to mean the set is doing "something bad" to the line when a ring voltage comes in, causing the CO to drop the call after a few rings.

If I pick up the line during the first few CO generated rings I hear in the calling receiver, I can take the call normally. And after I hang up, I can re-seize the line to get dialtone, and make a new call.. so it's hanging up properly, and not keeping the line held.

I've tried swapping T/R, but that's not it, because then I get dialtone but can't dial. So I've ruled that out.

I'm guessing maybe the absence of the 91A is causing this? Or should Hold + Ring work without it? (seems everything else does)

If I need a 91A, are there any schematics for it such that maybe I can rig it to work? I'm an EE guy, so depending on what the 91A does, I might be able to fake it. (From the pic in the manual, it seems possibly simple enough.) Otherwise I guess I might have to track one down.

Thanks for any help/advice.

In coming CO lines for line one goes on 24 and 49 the violet brown pair, not the white blue pair. With just a primary I don't think you'd need the 91A, but I don't have that manual. CO's went on v/br and v/sl pairs of the primary sets.

EDIT: After thinking about this a bit the 91A only gave you access to the v/br and v/sl anyway. I remember putting the CO's on at punch blocks and not using the 91A or B.

The set isn't doing anything "bad", it is just wired incorrectly as Bill mentioned. Violet/brown for line #1 and Violet/slate for line #2.

Greg, while not likely, you may have damaged it by connecting the CO line on the protected side of the set's circuitry and exposing it to ringing voltage. Only proper connections will tell.

Bill, you are correct with your assumption regarding the 91 block. All it really did is to isolate the violet/green through violet/slate pairs so that the CO lines could be fed in at the wall outlet if desired.

Ahh, the little legend at the *bottom* of the table in the com key manual.. missed that.. where it says T1 on 49 and R1 on 24 for line 1/3.

Yes, I'll try rewiring for that, and will cross fingers that the ring voltage didn't nuke anything in the phone. I'll bet the parts are durable enough ..will follow up when I try that after dinner.

The overall design of the ComKey 416 system, with others, was to protect the general public from the dangers of central office line voltages. I can't say for sure, but I believe that the intent of the 416 was to eventually be a user-installable system.

True, the hardware is very rugged and resilient to improper wiring. My fingers are crossed that no harm was done. From what you have described, the extent of your testing/troubleshooting was limited enough to where no harm may have been done. Had it been left connected like this for dozens of calls, then there may be room for concern. Something tells me that if a 1983 vintage phone lasted this long, you should be fine.

Please check back and let us know how you make out.

Yep, works great now when hooked up correctly.

Wow, I haven't heard that ring in years.

Thanks much for the fast help.

Nothing blown.

I've noodled with Com Key satellite sets over the years as a curiosity, and based on the hybrid and touchtone circuits in them, they really appear to be POTS designs with all the usual precautions for CO line voltages.

Although I'm sure there's front end circuitry to prevent ring voltages from reaching the Com Key voice paths, still when a call is in progress, the voice path circuits are exposed to raw CO line voltages, so it would have to be somewhat impervious to brief wacky voltage swings from the CO during line transitions; ie. pickups during ringing, rats chewing at lines, lines crossed, lightning, etc.

For instance, when I've got the phone wired correctly to the CO on 49+24, I can see the CO's 48VDC on 1+26 even when the line is idle, and when I pick up, it jumps down to 6 or 7 in the usual way.

Where I think you could blow up Com Key is if you were to put CO voltages on the logic paths, or cross it with the 24 VDC supply voltages on 45 and 20. Definitely the logic board's circuitry appears to all be low voltage stuff; that little 7805 regulator tells me it's good ol' TTL 5VDC voltage levels.

Thanks again!

Oh, and I have a little Com Key web page over here:
https://seriss.com/people/erco/comkey416/

..which I'll probably flesh out over the next few weeks, now that I actually have a primary set to study. I included today's info in there as well. Comments welcome.

Glad to hear it worked out! Thanks for getting back to us and letting us know.

Sam

From your warning note:

Actually T1/R1 is the output of the key system lines. The incoming CO lines went on T/R.


By the way to the board.

That's good information, Greg. I'm glad that everything worked out for you in your quest. There are a few things that you'll probably want to verify though:

1. The other phone model numbers in the 830/2830 series were for a different family of ComKey systems, namely the 718, 1434 and 2152. These systems were completely incompatible with the 416 system or phones. They also used a KSU, and a mighty large one at that. The 416 system always used the 9XX/29XX series phones only.

2. An RJ21X is a bridged network interface, providing up to 25 two-wire POTS or similar lines on a punch-down connecting block. It is a common misnomer to refer to a 25 pair "Ampenol-type" connector as an RJ21X, but this is very much incorrect. The same thing applies to the improper use of the term "RJ45". People today believe "if it walks like a duck, quacks like a duck, then it must be a duck". There are literally thousands of wiring patterns that can be accommodated using a 25 pair connector, with USOC RJ21X being one of them.

3. You are correct when you state that the satellite sets don't contain much in the way of electronics with the exception of the amplifier. Mis-wiring CO lines into them on the white/blue obviously won't cause the satellite phones any harm, but back-feeding them into the protected interface circuitry in the primary sets could possibly cause harm to them.

Hi Ed, Bill,

Much thanks for your comments -- I've watched this board for the last few years, and I know you guys are the experts -- I'll make corrections.

Ed: Fixed, I think. I've corrected the T1/R1 stuff to use more precise references to the BSP signal names.

The Com Key manual's table is a little confusing to the layman; it shows 1+26 as "T(1)" and "R(1)", and pins 49+24 are "T1" and "R1". My table omitted the parens because they seemed to be used inconsistently (some had parens, "T(#)", and some didn't, "CA#").

Looking closely, it seems the parens were used only on the 1A2 style names (T/R/A/L), but not for the new "Com Key specific" names (CA#,DSS#). So I take it this was some sort of subtle visual clue.

I've tried to make this clearer by adding a footnote to my table. I should probably also include a scan of the original BSP table, as my table was re-arranged in the connector's pin order (my preference) whereas the BSP manual shows them in the 66block order. Thanks for the clarification!

Bill:

(1) Oops, I did seem to have the wrong description for my illustration; that should be a 981, not a 28xx. Fixed.

But regarding there being no 8xx/28xx's in 416 systems, are you sure? Maybe I'm missing something; that seems to disagree with BSP 518-450-105 (entitled 4A COMMUNICATION SYSTEM/COMKEY 416) which shows 2836 and 837 phones as being 416 sets:
https://seriss.com/people/erco/comkey416/comkey-8xx.jpg
https://seriss.com/people/erco/comkey416/comkey-28xx.jpg

From what I gather between 518-450-105 (which covers 8xx/28xx) and 518-450-106 (which covers 9xx/29xx), the two are both 416 systems and compatible; one notable difference being the lamp field; the 8xx/28xx have the lamps "in" the buttons, and the 9xx/29xx have the more common config of the lamps along the left edge of the line buttons. But both are KSU-less and according to the BSP's, compatible with each other.

I do want to get this right, so thought I'd better double check. Am I wrong on this?

(2) Yes, I'll adjust those references to RJ21, which is I guess an ethernet standard that came a long later.

But I have to say, if you're searching for these connectors and cables, searching for RJ21 seems to yield the best results from parts distributors, that all describe them as RJ21 (probably because thats what most of their customers are using them for these days).

So I guess RJ21 being back compatible with the phone company wiring is a plus for those of us still working with old business phones!

I originally had references to AMP type; when I worked for a 1A2 phone vendor in the early 80's (V-Band Systems in NYC) as a draftsman, our wiring diagrams referred to these as "Amphenol type" connectors which I think is consistent with the BSP terminology.

I'll always remember my boss, Leonid, who with his thick russian accent, had a strange way of pronouncing "Amphenol", it was more like "ArrmffeegnooOoolll", his tongue really twisted around those syllables. Good old Leon.. where is he now. Was a fun guy, designed phones out of thick steel, so the stock brokers we sold them to could beat on them with the receivers without breaking them. (NYC stock brokers were always pretty stressed out guys)

I recall there were other vendors for these connectors besides AMP; ITT and a few others I can't recall now. Leon and a few other guys in drafting would have heated debates about which vendor had the better connector; some used thick plastics (phenolics?), some thin, some metal, some had bad tolerances causing them to fit loosely or wear out quickly.

Would it be correct to call these "25 pair Amphenol-type" connectors as phone company parlance, and to indicate if one were searching today for these connectors/cables, one would have better luck searching for "RJ21"? I take it RJ21 basically embraced the phone company's wiring standard.

(3) Right, I'd agree.

I was thinking of doing a circuit analysis of the Com Key phones.. are there actual schematics for these? ie. not just block diagrams, but actual circuit diagrams showing resistor values?

I figure a schematic would be useful for fixing old phones (for those of us that wire them wrong!) and to see what design techniques were used. It'd be a challenge though, as there are some unknowns, like those custom WE chips on the logic board. My guess is they /might/ have pin compatible commercial equivalents, but it's hard to know.. might have to guess on those.

Greg, I can already tell that you are going to blend in well around here.

You are absolutely correct about the 83X/283X nomenclature. I have absolutely no explanation as to why they underwent such a dramatic change except perhaps to avoid conflicts with ITT and S/C set designs with similar numbering that were beginning to infringe upon each other. Who knows?

I also concede that my continuing efforts to right the world's wrongs with regard to RJ21 or RJ45 misnomers are fading fast. The IT cancer has taken over our industry.

There were so many custom WE components, all the way down to resistors. I've never seen a true schematic for their assemblies, but it is obvious that they must exist (or have existed). They claim that just about anything ever made by AT&T/Lucent/Avaya has documentation available on their web site. It might be worth taking a look. They appear to be fairly liberal about sharing what others would deem to be proprietary.

I wasn't going to say a word about the RJ reference. 50pin Amphenol connectors were around long before they came up with Registered Jacks. I knew Ed would set it straight.

Ed, thanks! Yes, computers have found their way into everything; phones, video/film production, typesetting.. nothing can stop it, it seems.

You're right about Lucent at one time being quite open about old 1A2 stuff.. not to be too pesimistic, but those days might be gone; I recall doing some heavy 1A2 related searches, and encountered many links pointing to Lucent's website for 1A2 data, but all those links were dead.

Seems like around 2004/2005 those links were active.

A quick prodding of the search feature at support.avaya.com only yielded "end of maintenance" letters dated 2006 for Com Key and 1A2 searches.

Might have to dig deeper. If anyone has leads, do tell.

But I can't get this feeling out of my head that either now, or years ago, some manager at Lucent/Avaya inherited a room full of yellowing schematics all hopping with silverfish, and decided "Out of maintenance? Out of this office!", and I envision a guy in a back parking lot at Lucent with his white sleeves rolled up, an empty gas can, and a metal drum full of burning papers, and a column of smoke reaching to the sky.. (shudder)

Hope I'm wrong on that..!

I might (emphasize MIGHT), have a Comkey book in the dreaded basement. As I never really worked on them more then 2 or 3 times I don't remember the last time I even saw the manual.

But I'll look. Maybe it's got something in there.

Sam

Well, I checked. All I was able to find was the Key System Services Manual Vol. III. It's got the block diagram, but not a "real" schematic. (BSP 518-450-105)

Sorry.

Sam

Thank you Sam for checking!

Yes, I figured what you had is the '105 document from vol.3.. when folks refer to the Com Key "manual", that seems to always come up.

I guess what I'm looking for is probably an internal document to WECO, which would be a schematic the likes of which would never be used in the field, but more for parts replacement/repair.

Thing is, the schematics for things like the 400D line cards abound and the 1A2 phones abound.

One possible avenue I haven't checked is the USPTO; it's possible the Com Key schematics might be documented there.. though Com Key is more of a complete system.. I'm not sure they'd cover the whole circuit, as that would cover too many details.

Certainly I've found a wide variety of line card schematics there. A year or so I had a wild hair to build a circuit board from scratch that would the job of an entire KSU; hold, ring gen, even intercom and music on hold. All relatively straight forward stuff from an EE point of view, esp. with the help of modern chips (and by "modern", I mean 1980's tech)

While studying 400D card variations (WECO, ITT), and studying the variety of patents (including my old boss's), I came across this one 1A2 line card design that really impressed me with its simplicity; patent #4093830; their use of optocouplers with bidirectional LEDs inserted directly into the POTS line was an amazing approach.

I even built the circuit just to see if it works, because there were so few components.. and it did! I had a 1A2 phone doing hold/ring on two lines without a KSU. (Had to make up the ring and lamp winking stuff, but that was trivial).

I'll try posting my final schematic; some of you guys (or your audience) might be EEs who like this kind of stuff. If I do, I'll open a separate thread, as that's off the Com Key topic.

The big thing that blew my mind about that patent was the brazen use of optocouplers with back-to-back LEDs directly in series with the POTS line, no other circuitry to interface the LEDs into the line -- literally, in series between the phone and CO. I was amazed the LEDs did not affect the voice frequency stuff, handled ring voltages and the 48VDC all perfectly.

It's a great way to do ring detect and hook state sensing. Plus, the logic of the circuit with their use of transistors is very clever, using very few relays that are /not/ specialty items. (WECO used some very custom relays where the coils were used as part of the line state detection.. hard to come by).

This circuit was the best I'd seen, from a simplicity point of view. Seemed a circuit even novices good build/understand.

A good way to view the patents is through google's patent search:
https://www.google.com/patents/down...p;sig=ACfU3U2mkG6jbom87KrBrjprYwojUvVDBA

(I find it better than the USPTO's website; I have a lot of trouble with USPTO's image viewing.. for some reason their image viewer is really funky on non-windows systems like linux and mac).

Does anyone remember the similar phones made by Amtron? They predated the WE ComKey 416's and the WE were copies of them. I assume that WE bought the licensing and rights to make them from Amtron, although I have never been able to find any evidence of this, not can I find any mention of them in the search places.

Maybe looking for patents issued to the Amtron company would yield something interesting.

I saw one installation of the Amtron sets, and the only difference in operation was actually an improvement over the way WE did theirs: When pressing an intercom DSS key, the Amtron called station emitted a zip tone, then cut through the voice from the calling station. The tone could be transmitted repeatedly, by re-pressing the DSS key.

This feature allowed the customer to use the system as a sort of "button & buzzer" operation, in case there was a task that someone needed to do, by pre-arrangment, instead of having to listen to a voice message.

Arthur:

I don't remember those, but I remember 1A2 sets that looked identical to ComKey 416 sets that were made by TPI (TelePath Industries in Roanoke, VA). I believe that they were associated with Valcom as some point.

They did not have the DSS key field, but otherwise, they looked identical. Perhaps more similar to a MET. The dial had speed dialing functions, as well as redial.

Yes, Ed, I certainly do remember those, or ones very similar. During my many years enslaved by Mother Bell, I often sinned and did some outside work.

Xtel also made something similar. The ones I remember were called the "Perfect Ten" series, and they looked nice, and were easy to install. However, they became known as the "Perfect Disaster" phones after a few months in service.

Their only downfall was the fact that the 10-button key was hard-wired (!!!) into the phone, instead of being on a plug or connector arrangement. What a nightmare! When a key failed (and they did) you had to replace the whole phone.

The reason they failed was that instead of using a decent quality phosphor-bronze release tab on the Hold key linkage, they used el-cheapo plastic, which wore out and could not be replaced.

They took a nice aesthetic design, and ignored the Bell state-of-the-art in the internal stuff.

It's interesting there were knock offs, and even more interesting that WE itself may have copied the Amtron set.

I do seem to recall seeing systems where pushing a DSS key gave some kind of special tone alert. The Com Key 416 primary I have here doesn't seem to do that; when I hit the DSS key for my own phone, the speaker immediately becomes a live amplification of whatever one says into the handset; no warning tone.

I wonder if they limited themselves to 10 DSS keys so that one could push all 10 keys with all 10 fingers to do the equivalent of a 'page'

It's interesting the DSS keys don't light up so you can see who is talking to you on the speaker. Otherwise I imagine one could be relatively "anonymous" with announcements..

Greg, they did make a DSS key assembly with busy lamps. It was an optional item of course, but it plugged in as a replacement DSS with no special wiring required. Those signal leads had all kinds of activity going on.

There was also an adapter that occupied the code for one DSS key that emulated the all-call function. It had a selector switch to choose the code and that was pretty much it. It connected via (you guessed it), a 25 pair cable just like any other station.

I can't believe that WE copied anyone either, but there are still arguments about some of the very early switching systems, and then the infamous "who made it to the patent office first" debate.