Will the Schematic for a Dry Transformer Phone Line Interface Work?

[Phrak]

I need to know if the schematic of a "Dry" transformer phone line interface given on page 14 of application note 5 will work.

The application notes are here: http://randolph-telecom.com/articles-faq.html"

It has something that looks like a gyrator, but it's interfaced with a bridge rectifier.

Anyone familiar with telephone service interface circuitry?

 

[Phrak]

no takers?

 

[berkeman]

no takers?

I have a book about telephone electronics, but I need to get on a plane for the PF Mentor get-together. I'll take it with me and try to answer your question when I get there (they say they have WiFi access). Sorry for the delay.

 

[Phrak]

OK, berkeman.

Just now I decided to measure my own service (in the US). According to my research two wire should present nominally 50 volts and have a DC resistance of 600 Ohms.

I measured 22 volts unloaded and 15 volts when I put a 390 Ohm test load across it. ??

 

[berkeman]

22V unloaded is faulted, IMO. Is this at the phone line access box outside your house? If so, the local phone company usually has responsibility for the wiring up to that box. You should be able to call them to come test and fix it.

Where are you geographically?

 

[Phrak]

22V unloaded is faulted, IMO. Is this at the phone line access box outside your house? If so, the local phone company usually has responsibility for the wiring up to that box. You should be able to call them to come test and fix it.

Where are you geographically?

We don't have the best lines. For instance, we don't qualify for DSL, because we're to far from the field box. I have to use FIOS for the internet.

So, anyway, my own line isn't a very good test, but it may give me a worst case scenario, which is good. To get the numbers I pulled off the wall plate in the house. The calculated DC impedence of the service is 167 Ohms termination.

From what I've been able to piece together, so far:

US service is nominally 600 Ohms (DC) at 48 Volts. However one source says 200 to 400 Ohms. I'm to place a 150 ohm load across the line (One source says 200-300 Ohms.) to communicate an off-hook state to the exchange.

The load should also be inductive, which I can simulate with a gyrator.

I should expect an off-hook voltage of a nominal 9.6 volts. (One source says 5 to 9 volts.)

The transmit and receive bandwidths are both 300 to 3.3KHz. I think that means 3dB roll-off at 300 and 3.3K. Is that right? The modulation rides on the DC voltage offset.

The average level of the audio signal is "-9 dBm (275 mV)". I'm not sure if that means the tansmitted power, recieved, or both. In either case, I don't know what the impedance the 275 mV would be working into.

 

[waht]

22V for on-hook is too low unless there is another phone standard.

It's more likely that there is a shortcut, broken up splices that are barely held together, or corroded wires. If there is a tangled web of spliced phone wires somewhere in the house, cut everything, and redo all connections with new splices or an insulating tape. Also check up on all wall-plates in the house. If there is a sign of black reside on the copper, cut it out and redo it.

 

[Phrak]

22V for on-hook is too low unless there is another phone standard.

You're killin me waht. So I went outside and measured. It's still 22 volts, unloaded.

Besides, I don't care about my own service. I just have to verify a design.

 

[waht]

You're killin me waht. So I went outside and measured. It's still 22 volts, unloaded.

Besides, I don't care about my own service. I just have to verify a design.

Lol, phone wiring can be messy sometimes. But when you went outside to measure voltage on the line, did you actually disconnect the main line from the house and then measured its voltage or measured the voltage while the house is still connected?

 

[Phrak]

I'm on to you waht. You want me go a second session in the dirt, weeds and rust. If I come back and say it's still 22 volts, you'll have me busting into one of those green boxes with crowbar in one first and a DVM in the other.