I Have a new installation of a 7100 with an Analog line attached. When trying to use the ERL tool, I get this message back:
CNL_NV7100_MEX(config-fxo 0/1)#test erl all auto-set
08:12:38 ERL.INFO fxo 0/1 erl measurement started, using impedance 600cCNL_NV7100_MEX(config-fxo 0/1)#
08:12:41 ERL.INFO fxo 0/1 impedance 600c - abnormal test termination
2014.06.17 08:12:41 ERL.INFO fxo 0/1 Did not auto-set impedance because of abnormal test completion
2014.06.17 08:12:41 ERL.INFO fxo 0/1 erl test terminated abnormally
I know we are connected because you can call in and hit the line. However you cannot talk over it due to buzzing and low volume. A 2500 set was plugged into the circuit and works fine.
We got it working. There was a line attached to the port, but there was a DSL that was not properly filtered on it. Also the ERL would not work until Blind Dialing was enabled on the circuit.
Double-check your wiring. It sounds like you might have a ground on the line. If you have a second FXO port, swap them to see if the trouble follows the line or the port. Bad hardware is a possibility if the 7100 has an internal short to ground.
We did swap the line to the second FXO port and i could not run the test on that either. We tested with an analog phone and hear no ground hum. There is a DSL, but there is a splitter at the NID provided y the carrier. Unfortuanly this system is installed in Mexico as a remote office, and can do little testing . Is there a way to locally loop back the FXO to a local FXS port to verify the hardware?
You could probably patch an FXO to FXS for testing, ideally on a second box just as a simulated line. It would be a tricky configuration on one unit. Do the people on site have a multimeter and can they verify -50 volts and ground on the incoming line when on-hook? It's possible that it is overvoltage and causing a MOV surge protector in the Adtran unit to fire.
Just for testing, try rolling tip and ring on the pair and see if that fixes it. Normally the Adtran gear isn't polarity sensitive but you never know.
There is no second box to test off down there. I can inqure about the
multimeter. How would the setup look for testing off the same unit?
Neither of the FXS ports are currently in use, and we can do whatever we
want with them. Does the ERL tool need to actually place successful
calls out to work properly?
I'm not sure exactly what it does, I've never used a 7100. Ideally, it would call a local central office test number that provides a silent termination, then run a sweep tone and measure what comes back. As Adtran has no idea where any specific unit will be placed or what test number to use, this isn't going to be practical. In practical terms, an ERL test shouldn't be required to turn up the interface and place calls. It would be useful for setting impedance to match local conditions.
If the ERL allows you to specify a destination number, then call a POTS line, answer with a standard 2500 set in a location as quiet as possible. The goal is to terminate the connection with something closely resembling a standard telephone that produces no signal of its own. Once terminated, the tester sends a signal. Ideally 100% of that signal is absorbed by the other end (your 2500 set in a quiet room). Any signal that is reflected back is measured, higher signals are worse. "Loss" is considered a bad thing generally, but in the case of "Echo Return Loss", the more loss (less reflected signal) the better.
The concept is similar to SWR on a radio circuit.
The ERL test and hybrid balance is going to be very subtle if noticeable at all to the person on the 7100. If out of whack it may be noticeable to the remote party as echo. It can also affect modem and fax transmission.
In your case, if you have clearly audible hum and low volume there is something else going on with the physical loop or interface.
When activated the ERL tool goes off hook, verifies loop current and dial tone (unless blind-dial is enabled), then generates white noise with a particular impedance value set, and measures the resulting echo level. When using the command test erl all auto-set, this process will be repeated with all the different impedance values. The value that results in the best ERL (highest value) will be configured on the trunk. This test cannot complete when calls are coming into or going out of the FXO interface, so it needs to be performed at a time when traffic is minimal. You can read more about the ERL Tool in this document:
I agree with jayh that this sounds like a line issue. Were you able to resolve it? If so can you come back to this thread to provide an update so others can benefit from the information?
We may be replacing this trunk with a SIP trunk, might not be an issue.
Plugging in a 2500 set produces no call quality issues on this line,
only when plugged into the 7100. Can I test the FXO interface by
looping back from an FXS?
To answer your question I would not recommend plugging an FXS into a FXO of the same unit as a reliable way to test. Were you ever able to get the bottom of this issue? Was it the line?
No resolution. I think they abandoned the use of the local circuit. We
are routing all calls through a SIP trunk to the main 7100. Are there
any tools available to see why we get "abnormal termination" of the ERL