atcomsystems.ca/forum
When I was a kid I took apart a phone (probably a 500). For lack of a better term (i.e. what do you call it?) - the terminal block where all the connections join together?

Also, what is inside of that thing? And, since not all phones have it, why is it needed? I can see the need for a place to connect everything together... but that's about it.

Anyway, as I said, I took (rather tried to) take one apart when I was about 8. Pried up the tabs, but could only get it out about 1/2". Was immediately confronted with a bunch of goo, and decided maybe this isn't such a healthy thing to be getting on me. (This was the same time when rampant warnings were running around to not take apart a golf ball, as at the core was a substance that would burn, blind and cripple you).

Whats in there, why is whatever in there needed, and what's with the goo?

Thanks
Matt
It was the network. I never took one apart so I have no idea what was inside them.

I think You were a lot like me when I was a kid. I was always taking my toys apart to see what was in them. Sometimes I managed to get them back together. laugh
But it paid off I think, I scored 100 percent on the mechanical portion of the asvab test when I joined the Air Force.
Thanks! Network. Easay enough to remember.


Oh yeah.. we were alike. from TV sets to radios, anything that plugged in. Used my first pair of wire cutters on a live lamp cord. That was fun. They tell me I spent most of my toddler years sticking things in receptacles.

Then there was the time I got a plug fuse cause I thought it was a really cool looking light bulb. For about 1/10000000 of a second it really was. A little bright though.

I too had mixed results on getting things back together. But I'd say I had about a 65% success rate, most times actually fixing something. It really freaked people out when a 7 year old put in a car stereo or fixed their brake lights for 'em.

I don't remember my ASVABS... They were good enough to qualify me for a lot of the more technical ratings (Navy).. but sure wasn't 100%.

Never have understood Math.
When I was a kid, starting about age 9, I got a new Timex every Christmas. It would last until about the following November. When I was about 14, I decided to investigate why my latest watch had stopped. Got the back off and, son-of-a-gun, there was an eyelash in there jamming what I later found out was the balance wheel. Sooooo, I removed it. all 3 feet of it! It was actually the broken mainspring. I just slid the mess off the side of the desk into the trash basket and got a new watch for Christmas, as usual! John C.
lol

John - for future reference - there's this dial on the side you have to wind from time to time. Actually, if you pull out on it.... you can even set the time!!!!

smile
Matt, back to your original question, yes that was an early-vintage 425 type network. Many of the terminals on the top are blind (i.e: tie points), but five of them are associated with the balancing transformer. The two wires of the phone line go in on terminals C and RR. The three wires going to the handset go to terminals B, R and GN (actually two of the handset wires end up going to terminal R. The goo was in there to basically make the network much more resistant to shock if the phone was dropped. These things were so reliable that they were riveted to the base.

All phones have some form of network, just in various configurations. Western Electric had about five different ones, some to fit inside smaller phones, etc. They eventually stopped filling them with goo in the late 1970's. ITT and Sromberg Carlson switched over to a PC-board type network around that same time. At this point, phones were becoming disposable. A network that would last a lifetime was a great idea when phones were being rented.
Matt, real phones, such as all those based upon the circuitry of the 500 set, exhibit a trait called "sidetone." It is the effect that allows you to hear your own voice at a correct level, while conversing with the person at the distant end. It can be remembered by visualizing the "side" of your head, where the handset goes.

In generic terms, the 425-type WE "network" is a hybrid coil, or transformer, and a few capacitors. You need to see an older version of a BSP that shows the actual schematic, rather than a newer BSP that just shows the block diagram.

There are no phones that do not have some electrical equivalent of a network. The goo is silicone compound. Earlier ones had some sort of black crap in them. Later ones were open-air affairs, like the 4228-series, and their guts are visible.

The hybrid is arranged so that there is just the exact amount of sidetone so that your voice is reproduced in your receiver unit at a balanced level, taking into consideration the loop length and your local loop current. If you have too much sidetone, your voice sounds loud to you, and to compensate, you will find yourself speaking too softly. The person at the other end will not be able to hear you properly. Alternatively, if you sidetone is too low, you will find yourself shouting, and the person at the other end will assume that you're either very excited, or drunk.

Seasoned repairmen, working on trunk circuits, can tell if the trunk impedance is set properly, just by listening to the sidetone of a circuit, and they can do that without even establishing a conversation. They just blow softly in the transmitter, and listen to the white noise in their receiver. Too loud or too soft = impedance problems, or equalizers set improperly.

Newer el-cheapo phones occasionally have awful sidetone problems, and an old guy like me will remark on it. Nobody listens, or cares, but we complain anyway.

sidetone
Thanks guys!

Now I understand just a little bit more of this huge puzzle.

I take it that cell phones don't have sidetone, and that's why people yell into them, as they can't hear their own voice?

And I'm glad the goo was just silicone.
Arthur, how about the lack of sidetone on cellphones...
"Newer el-cheapo phones occasionally have awful sidetone problems, and an old guy like me will remark on it. Nobody listens, or cares, but we complain anyway." Amen to that, Arthur. But, I'll bet some day every one of 'em will say, "what'd that old phone guy say about sidetone?" When SWB started doing away with build-out coils and making the compenation in the SLICs, they never thought about G/S trunks, I guess. At least in the 1 C.O. town I lived in, the motel calls were ok to loop start lines, but any time a Motel called a Motel, EEEEEEEEE, or HOLLOW, or well you know. Since we took care of most of the motels in town, once I realized what was going on, I requested a pad of 1-2 Db on all the trunks of the motels we dealt with. BUT, we picked up all but one of the other motels in town, cause we were the only ones that could stop the problem. smile The others hadn't been padded down yet and just enough gain to EEEEEEE, when they called 1 of 'our' motels, or each other. But all the owners talk and good as well as bad gets passed around, pretty soon all but 1 began to understand that we were 'The Best'. It does my heart good to know that when they call my old outfit, they STILL want to know if and when I'm coming back! OW, I broke my arm patting myself on the back! smile John C.
Ah, If I just read before posting (From Arthur's wiki link):

"Many cell phones do not provide adequate sidetone. Usability experts believe this causes some people to shout or speak too loudly when using a cell phone. [1][2] Similarly, in military field operations in enemy territory, operatives can be encouraged to speak quietly by increasing the volume of the sidetone."
The goo is silicone compound. Earlier ones had some sort of black crap in them.

Circuitry like that is said to be "potted". It's used to protect the circuitry from water, moisture and tampering. That "black crap" was tar. Over the years many different types of potting compounds have been developed. The usual one today is a two part epoxy.

-Hal
Thanks Hal -

I'm just glad it wasn't PCB's..... and it sure kept me from tampering.

- Matt
Good topic for those without a proper telecom upbringing. As Arthur said the network is based on the hybrid circuit. They purposely make it unbalanced to provide the sidetone.

Matt, hopefully you will challenge us with qustions about queuing theory and nyquist sampling. You know that queing theory can be used to predict earthquakes, or the odds of obtaining a parking place.

Of course you all know my opinion about cell phones and their destruction of acceptible quality
I know about nyquil sampling... does that count? J/K.
Matt,

Nyquist states that you need to sample at twice the highest frequency to obtain a true representation. That being said, the phone system is based on transmitting bandwidth from 300-3300HZ where most power and telligable voice occurs. Some may say 3600HZ, but that neither here nor there. What really matters is that PCM circuits are at 8KHZ i.e. twice 4000Hz. 8 bit sampling at 8Khz gives you th 64kbps of a standard T-1 channel. Doing some math, 24 channels with this sampling rate + some signaling bits, gets you the 1.544 value.

Also that is why you see compact discs at 44KHz. Twice the highest frequency that one(not me) could hear.
And queuing theory?
Queuing theory is the reason that during a disaster when you pick up the phone you don't get dialtone.

I'm being simplistic of course, but Telephone Systems are built around the concept that they will NOT supply DT to everyone all the time.

It would just be too expensive

And the reason you can get away with it? Not everyone is (usually) on the phone at the same time.

Figuring out the right ratio of equipment (Trunks, DT registers, Time slots etc) to Telephone Lines is where queuing theory comes in.

It relies very heavily on statistical theories and practices. I did quite a lot of it when I ran the operation at the Bank of New York. Matter of fact when I took over as Network Manager it was probably 25% of my job.

An example:

You have one phone in your business - you need at least one line from the phone company. As you add phones and personnel you need to add more CO Lines, but not necessarily on a 1 to 1 basis. (Call Centers are the exception). The bigger your business gets the LESS CO lines you probably need as more and more of your business is carried on in-house.

At my location at the network hub of the Banks Private Network we had over 5,000 telephone sets. We had 180 DID (incoming) Trunks, about 180 DOD (outgoing) Trunks about 120 Long Distance Trunks and about 300 Tie Trunks to other locations on the network.

That's a ratio of less then 1 outside circuit for every 6 phones. We did have a very high ratio of time slots to extensions (24 slots for every 64 extensions) - that allowed for a high degree of internal calling.

Other sites on the network had similar ratios. (We had about 15,000 people on the private net).

Ask away. This is one subject I used to know quite a bit about. It should be a separate post though, probably under Telephone Services.

Sam
Cool. I can see how that would work. Then I guess you probably low ball it, and if the customer complains of having trouble getting a line, or grows, you just adjust upwards?

On the Carrier I was on, when in port, it was the inverse. 3,000 phones and probably 3 lines going out. Sometimes you'd spend hours trying to get an outside line. When at sea it was even harder. laugh
Matt, same at Thule, Greenland AFB. Program called Morale Calls. Lines were available after 7PM local time to New Jersey, collect from there. Call the base operator and get your name on the list. Call might go through ANYTIME!
However, being in the sub-cable head bldg, we had access to allll sorts of options. Easiest was taking a morale call circuit for maintenance. But, you still had the collect charges. I used to get a spare channel on the Tropo shot patched thru to Cheyenne Mountain and then get an Autovon connection to Richards-Gebaur, then ask for off-base number XXX-YYYY. I quit that when I hung up, got a ring back a couple minutes later. It was the operator @ Cheyenne Mountain, wanting to know why they couldn't release the call. I call the Tropo shot @ Cape Dyer and had them tear down their patch, Tech control on base and had them tear down their patch and finally, tore down my patch. For all I know, that operator STILL can't release the call! John C.
Bell typically lowballed it, at GTE we usually went a little better on the residential subscribers and a lot higher on the business network end of things.

Now you get into things like CCS and Erlangs.

CCS=Call Centi Seconds. How many seconds an hour can each phone be off hook. There are 60 seconds in a minute and 60 minutes in an hour, therefor there are 3600 seconds or 36 Centi seconds in one hour. To allow each phone to be off hook anytime and never get a busy signal would then require that each phone have a CCS of 36. Bell used to spec a residential CCS of 2.5 on their COs. NEC and Northern Telecom (on their PBXs) went in for a CCS of 5 or a 6. GTE used to spec about a 10 or a 12 for their private networks. I'd have to pull the old priming charts and tables to be sure but those numbers should be close.

At the Bank, the SVP of Telecom told me he wanted 1 more trunk then he really needed. He never wanted people to get a busy signal, but he didn't want to pay too much.

You need to run endless traffic studies, find the "Busy Hour" and then work from there. Before PCs it involved pouring over page after page of green-barred printouts. And after you figure it out you've got to continually monitor it, because - things change!

I actually got written up in a trade journal because I created a PC based application that made the Network Managers job significantly easier.

But that was then.
Sam I remember doing traffic studies also, we had classes on traffic theory at SNET. If I remember right we used Poisson tables for DID trunks. Back then the phone company could dictate to you the number of DID trunks you had to have so as not to back-up the network.
Larry -

Poisson tables were for Erlang, right? I seem to remember that you could run your traffic tables with either CCS or Erlang. They were essentially two ways of looking at the same thing.

I learned them both but stuck to CCS - unless I was dealing with our overseas operations and then I would (curse and) convert to Erlang.

Sam
I'm somewhat familiar with the queueing theory.

Part of my job consists of deconstructing it to figure out at what rate (dials/second/channel) calls can be placed in order to not busy out a shelf and getting the customer fined for obstructing a public utility.

If it's not that it's trying to figure out at what point they start throttling you. They all do it but wont admit it. I've actually had to have a carrier distribute T's through their facility to prevent it.

Like Sam said, call centers are the exception to all things queuing theory smile
So, to answer your original question Matt, they put that goo inside the network enclosure to keep people from opening them. Does that answer your question in fifty posts or less? :rofl:

Sorry, I couldn't resist.
LOL - and it works really well!

It's amazing what one can learn from goo, isn't it?

And if ya told me when I was 8 I'd be asking about goo 36 years later, on a magical device that would let me communicate with some of the very people who were installing the goo container containers at that very moment... well, you know.

And further, that one day that same goo would be responsible for me learning a little bit about Nyquist, and Queing theories, I'd of probably said: "Damn, I wish I'd of torn this apart 2 years ago."

Ed, I think your post deserves another goo story, only let's really go off into the ditch this time.

Brother says his flourescents come on intermittently... I tell him it's probably a bad ballast. Course he doesn't know what a ballast is, so he's looking for a starter, but can't find one. I tell him better to let it be, a little about PCB's, how very unlikely it is they are present, but still .... leave it alone, I'll have a look at it.

So I go on over, flip the lights on, and sure enough, they're intermittent.

Take the cover of the fixture, and .... loads of black goo drip on down........ brother says: Oh! so that's a ballast!

No, that WAS a ballast. Now it's a goo dripping black box.

Take the cover off the other fixture (there were 2, 2 bulb strips), just to make sure... and whatta ya know, the goo is oozing there as well.

So I cut it out and bring it down, brother looks at it and says Hey! no PCB's! Says so right here.

So, my brother now knows a bit more about how flourescent fixtures work, what PCB's are, that a new ballast costs more than a new fixture, and how to replace a fixture. But, he doesn't know a darn thing about Nyquist, I can tell ya that much.

Regardless, all these good things come from our ever present friend: goo.
PCB's and leaking ballasts.... I shudder to think of the hundreds of leaking ballasts that I replaced in my full-time career as a sparkie. Not to mention the tons of "that white stuff" that I removed from boilers and steam pipes when working as a plumber's helper through high school.

Cracking into a phone's network sounds like small potatoes now.

Hey, you hijacked your own thread this time, so I'm not responsible for having replied.
The PCBs weren't in the "goo", which was the same as the potting in that old network. The capacitor, one of the two devices in that goo used to contain small quantities.

-Hal
Quote
Hey, you hijacked your own thread this time, so I'm not responsible for having replied
ROFL. I got busted.

Sure you didn't have a part time job as a cop in there somewhere?
© Sundance Phone System Forums - VOIP & Cloud Phone Help