Phone Line Issue w/Obi 110
sailing:
Sorry for not responding sooner but I don't get to read the forum everyday.
The on hook voltage is irrelevant. The off hook is where the problem would lie. 70mA seems awfully high but you didn't measure the voltage during off hook. Measure voltage and current off hook with a phone. Repeat with the Obi installed, with the phone plugged into the Obi off hook.
The POTS specification is a compromise to allow few problems under almost any circumstance, such as being too close or too far from the CO. Some lines will be just outside the spec but a phone will work regardless. My thinking is your setup may be operating in this type of situation. The phone company only needs to make the lines work with phones so they know how to do that. Electronic devices are designed to work within a certain spec. Once outside of it and they may have problems.
As of a couple of years ago, the integrated circuits in the Obi110 were from Silicon Labs and probably still are. I'll see if I can get the general specs for their POTS ICs. Then the above voltage and current measurements can be compared.
DanielG:
To Ostracus - I don't know how the A/D | D/A conversion works - that's news to me. As for the half/full duplex comment, we can unplug Ethernet all together and still have this issue for POTS calls. It's not related to Ethernet.
To Sailing - we'll test voltage/current again, but we just had a Verizon tech onsite who tested everything and he left saying the voltage/current was within specs and there's nothing else he can do to improve it further. Said it's totally ideal. Let me know what you hear back on the general specs for the Obi chips.
Daniel
Shale:
Yes, it is an interesting problem. Your duplicating the problem in more than one setup makes the more obvious potential causes improbable.
I presume that your distance to your switch is fairly short. As an experiment, I would try connecting the OBi110 line port with maybe a 1 K or 470 ohm resistor in series. I understand that a balanced pair of resistors would be better, but I think that the mismatch would not introduce a significant problem in a short loop. Or use 2 resistors. That does feel more more harmonious.
You may even have a breakout box that makes this easier than locating and cutting a cable that has real copper wires and tacking a resistor or two in place.
Why do I think this is worth trying? I think it will reduce the current for one thing. I suspect that you line from the switch is much lower impedance than it would be from a central office.
DanielG:
So here are the numbers from the latest test:
No Obi: 16vdc and 70ma
With Obi: 9vdc 70ma
All measured off hook
-Daniel
sailing:
I took a current measurement into a off hook phone on a POTS line. The current into the off hook phone was 40mA. 70mA seemed high so now I'm thinking maybe the input of the Obi is being overloaded. There is a current limit setting in the obi that limits the current to 60mA. Go into the Obi110 setup Physical Interfaces>Line Port>Port Settings. Uncheck the default for CurrentLimitingEnable and check the box to limit the current.
Another way to check this idea out is to plug one or two phones into the POTS line with the Obi. When talking into the phone plugged into the Obi, also off hook the other phones that are plugged directly into the POTS line. The other phones will shunt some of the current so the Obi will have less current to it. If the issue is too much current, then the Obi's phone should sound ok at this point.
Shale's idea with the resistors is another way to experiment but you have to choose values that will drop the current to around 40mA. 600 ohms should work although Shale's choices will work also.
Navigation
[0] Message Index
[#] Next page
[*] Previous page