7 ms·

I once had an issue where my phone line was completely dead, but I was still getting ADSL, just at very slow speeds. In the order of 1mbit. After the technicia

I once had an issue where my phone line was completely dead, but I was still getting ADSL, just at very slow speeds. In the order of 1mbit. After the technician came he told me that one of the two phone line wires had been broken off at the exchange. In other words, there wasn't a complete electrical circuit. The phone was dead but ADSL still works without even having a complete circuit. I don't even understand how that can work at all.

nomel 9y ago
Everything about a POTS operates completely from DC to audio band, from the DTMF tones [1] when you press a number to the signal that makes the phone ring (~90V@25Hz AC) [2], to the -48VDC power that makes the keypad light up and any electronics operate, backed up by massive battery banks and generators, in case of power outage. ADSL operates completely beyond the audio band[2]. Since it's entirely a high frequency AC signal, with no DC component whatsoever, a break in a wire will just act as a capacitor, reducing the AC signal strength. Since a broken wire is a terrible capacitor, the signal strength will be reduced tremendously, but in your case, not enough to drop the link. [1] http://www.etsi.org/deliver/etsi_es/201200_201299/20123502/01.01.01_60/es_20123502v010101p.pdf http://www.etsi.org/deliver/etsi_es/201200_201299/20123502/0... [2] https://en.wikipedia.org/wiki/Asymmetric_digital_subscriber_line#/media/File:ADSL_spectrum_Fritz_Box_Fon_WLAN.png https://en.wikipedia.org/wiki/Asymmetric_digital_subscriber_...
- Mister_X 9y ago
  Hmm, should have read your post first... Thanks for the more detailed explanation, and with that, my guess above, is on the money, interesting communication tech.
  - nomel 9y ago
    The most fascinating thing about any of these high-speed-over-crappy-wire technologies is that they all have very cool signal processing going on, with echo cancellation, crosstalk compensation, and even trading of signal response, all in an effort to design and transmit a sort of "anti distortion" distorted signal that, once passed through the distortion caused by the crappy wire, will results in the desired clean signal. For a funny example of how sensitive some of these links are, just flick a 100m ethernet cable, running at 10G, with your finger. If you're of normal luck, this will cause a temporary changed in the electrical characteristics of the wire, and that "anti distortion" distortion to be temporarily incorrect, causing a interruption in the data!
dwyerm 9y ago
I had the same experience, except it was my fault. I crimped a cable poorly. But I had no idea anything was wrong. I didn't even bother testing it because it Just Worked. The phone company (Centurylink) just one day called me, "Hey, we think we see a problem with your line, and we're sending someone over to check it. Is Saturday okay?" It's the most proactive I've ever seen a huge ISP be, and it really impressed me. Almost as much as getting nearly my full rated speed out of a single wire.
- toast0 9y ago
  I'm guessing they called you because the connection was so bad the compensation for your line was interfering with other lines that ran parallel to yours. It is nice that they called you, the standard procedure is to disconnect you and when you call to complain, come out and fix it (and bill you for it)
  - dwyerm 9y ago
    I think you're probably right. I was fully expecting to be billed, especially when the tech ended up climbing under my desk. Normally, they draw a bright line at the dmarc. But I cynically wonder if it didn't have something to do with my contract ending and the price nearly doubling, leaving me free to jump to the other provider. It worked, though. I'm still with them. It doesn't hurt that the other provider is Comcast...
jacquesm 9y ago
HF will jump that small gap easily enough due to capacitive coupling, POTS is a DC current to begin with as long as you're on-hook and very low frequency when off-hook so you're out of luck.
cdancette 9y ago
1mbit is my usual throughput.. I hope I'll get the fiber soon
Angostura 9y ago
Did the dog bark just before an e-mail came in?
- toomanybeersies 9y ago
  I head that story years ago from my father, who's a network engineer. I've always wondered if it's real, or one of those funny stories with a technical punchline. It seems feasible.
- chipperyman573 9y ago
  What story is that? Google didn't return anything.
  - haser_au 9y ago
    Search for "Berkeley Biff Dog" http://www.catb.org/jargon/html/B/biff.html http://www.catb.org/jargon/html/B/biff.html
  - razakel 9y ago
    >What story is that? Google didn't return anything. I'm not able to find a definitive source, but it could be the following, which I have also heard from a retired BT employee: "It's common practice in England to ring a telephone by sending extra voltage across one side of the two wire circuit and ground (earth in England). When the subscriber answers the phone, it switches to the two wire circuit for the conversation. This method allows two parties on the same line to be signaled without disturbing each other. Anyway, an elderly lady with several pets called to say that her telephone failed to ring when her friends called; and that on the few occasions when it did ring her dog always barked first. The telephone repairman proceeded to the scene, curious to see this psychic dog. He climbed a nearby telephone pole, hooked in his test set, and dialed the subscriber's house. The phone didn't ring. He tried again. The dog barked loudly, followed by a ringing telephone. Climbing down from the pole, the telephone repairman found: a. The dog was tied to the telephone system's ground post via an iron chain and collar. b. The dog was receiving 90 volts of signaling current. c. After several such jolts, the dog would start barking and urinating on the ground. d. The wet ground now completed the circuit and the phone would ring."
Mister_X 9y ago
Interesting, I had a similar problem, suddenly no land line, and crippled Internet, then after a week, dead, I had nothing. I used to work for a private TelCo, so checked all my inside wiring, then had a friend call in a trouble report, and it took the damn company 3 weeks! to clear the trouble, because their line test equipment indicated the line was fine. After reading your post, I'm guessing it was a barely broken/corroded part of a pair in an outside "B" (connection) box, as I live within 1/4 Mile of the ocean.

nomel 9y ago

Everything about a POTS operates completely from DC to audio band, from the DTMF tones [1] when you press a number to the signal that makes the phone ring (~90V@25Hz AC) [2], to the -48VDC power that makes the keypad light up and any electronics operate, backed up by massive battery banks and generators, in case of power outage. ADSL operates completely beyond the audio band[2]. Since it's entirely a high frequency AC signal, with no DC component whatsoever, a break in a wire will just act as a capacitor, reducing the AC signal strength. Since a broken wire is a terrible capacitor, the signal strength will be reduced tremendously, but in your case, not enough to drop the link. [1] http://www.etsi.org/deliver/etsi_es/201200_201299/20123502/01.01.01_60/es_20123502v010101p.pdf http://www.etsi.org/deliver/etsi_es/201200_201299/20123502/0... [2] https://en.wikipedia.org/wiki/Asymmetric_digital_subscriber_line#/media/File:ADSL_spectrum_Fritz_Box_Fon_WLAN.png https://en.wikipedia.org/wiki/Asymmetric_digital_subscriber_...

Mister_X 9y ago
Hmm, should have read your post first... Thanks for the more detailed explanation, and with that, my guess above, is on the money, interesting communication tech.
- nomel 9y ago
  The most fascinating thing about any of these high-speed-over-crappy-wire technologies is that they all have very cool signal processing going on, with echo cancellation, crosstalk compensation, and even trading of signal response, all in an effort to design and transmit a sort of "anti distortion" distorted signal that, once passed through the distortion caused by the crappy wire, will results in the desired clean signal. For a funny example of how sensitive some of these links are, just flick a 100m ethernet cable, running at 10G, with your finger. If you're of normal luck, this will cause a temporary changed in the electrical characteristics of the wire, and that "anti distortion" distortion to be temporarily incorrect, causing a interruption in the data!

dwyerm 9y ago

I had the same experience, except it was my fault. I crimped a cable poorly. But I had no idea anything was wrong. I didn't even bother testing it because it Just Worked. The phone company (Centurylink) just one day called me, "Hey, we think we see a problem with your line, and we're sending someone over to check it. Is Saturday okay?" It's the most proactive I've ever seen a huge ISP be, and it really impressed me. Almost as much as getting nearly my full rated speed out of a single wire.

toast0 9y ago
I'm guessing they called you because the connection was so bad the compensation for your line was interfering with other lines that ran parallel to yours. It is nice that they called you, the standard procedure is to disconnect you and when you call to complain, come out and fix it (and bill you for it)
- dwyerm 9y ago
  I think you're probably right. I was fully expecting to be billed, especially when the tech ended up climbing under my desk. Normally, they draw a bright line at the dmarc. But I cynically wonder if it didn't have something to do with my contract ending and the price nearly doubling, leaving me free to jump to the other provider. It worked, though. I'm still with them. It doesn't hurt that the other provider is Comcast...

jacquesm 9y ago

HF will jump that small gap easily enough due to capacitive coupling, POTS is a DC current to begin with as long as you're on-hook and very low frequency when off-hook so you're out of luck.

cdancette 9y ago

1mbit is my usual throughput.. I hope I'll get the fiber soon

Angostura 9y ago

Did the dog bark just before an e-mail came in?

toomanybeersies 9y ago
I head that story years ago from my father, who's a network engineer. I've always wondered if it's real, or one of those funny stories with a technical punchline. It seems feasible.
chipperyman573 9y ago
What story is that? Google didn't return anything.
- haser_au 9y ago
  Search for "Berkeley Biff Dog" http://www.catb.org/jargon/html/B/biff.html http://www.catb.org/jargon/html/B/biff.html
- razakel 9y ago
  >What story is that? Google didn't return anything. I'm not able to find a definitive source, but it could be the following, which I have also heard from a retired BT employee: "It's common practice in England to ring a telephone by sending extra voltage across one side of the two wire circuit and ground (earth in England). When the subscriber answers the phone, it switches to the two wire circuit for the conversation. This method allows two parties on the same line to be signaled without disturbing each other. Anyway, an elderly lady with several pets called to say that her telephone failed to ring when her friends called; and that on the few occasions when it did ring her dog always barked first. The telephone repairman proceeded to the scene, curious to see this psychic dog. He climbed a nearby telephone pole, hooked in his test set, and dialed the subscriber's house. The phone didn't ring. He tried again. The dog barked loudly, followed by a ringing telephone. Climbing down from the pole, the telephone repairman found: a. The dog was tied to the telephone system's ground post via an iron chain and collar. b. The dog was receiving 90 volts of signaling current. c. After several such jolts, the dog would start barking and urinating on the ground. d. The wet ground now completed the circuit and the phone would ring."

Mister_X 9y ago

Interesting, I had a similar problem, suddenly no land line, and crippled Internet, then after a week, dead, I had nothing. I used to work for a private TelCo, so checked all my inside wiring, then had a friend call in a trouble report, and it took the damn company 3 weeks! to clear the trouble, because their line test equipment indicated the line was fine. After reading your post, I'm guessing it was a barely broken/corroded part of a pair in an outside "B" (connection) box, as I live within 1/4 Mile of the ocean.