LINE Port LoopCurrent question

[Mango]

I have been having severe audio problems with the FXO port on my OBi110, but not when a phone is plugged directly into the POTS line.  A tech from TELUS was here today and he found and fixed some damaged insulation (exposed copper) outside my house.  That didn't solve the problem, though certainly a good thing to do.

While I was testing the problem, I noticed the LINE Port LoopCurrent was 56 mA.  Then suddenly it dropped to 26 mA and the problem went away.  The tech mentioned he was going to check the CO, so perhaps he did something there.

tl;dr: Can anyone explain what the LoopCurrent is measuring?  Is 56 mA is actually a problem or just coincidence that I noticed it?

[SteveInWA]

It's measuring the electric current flowing through the two-wire analog telephone circuit between your phone and the central office switch port, or your ATA, if it's a VoIP line.  It should be between 23 and 27 milliamps (mA) when the phone is off-hook, and it should be zero or near-zero when on-hook.  56 mA is way out of range, and would definitely cause the severe distortion, and possibly damage your equipment.  Perhaps the telephone company tech fixed something in the CO that was causing the problem.

[Mango]

Thanks for confirming this.  Unfortunately the problem has returned; now the LoopCurrent is 62 mA.

[SteveInWA]

Did the tech test the loop current without the OBi in the loop?  That would be the obvious "is it this, or is it that" test.  If it's that high at the demarcation box, with all your premises equipment (phones and wiring) disconnected, then the telco has to fix it.  If it goes up when the OBi is off-hook, then you've got a bad OBi.  The same logic would apply to whatever else is connected in parallel to your phone line, like other phones, fax or answering machine, etc.

Note that you can't just hook up an ammeter to the two wires, with no telephones or other telco devices in the circuit.  You'll overload your meter and not get a reading.  You need a standard impedance load on the line to measure the current (the load is 600 Ohms in the USA, not sure about Canada).

[Mango]

Unfortunately he did not.  I only discovered the LoopCurrent after he left and the problem came back.

I just tested another OBi110; it also showed a LoopCurrent of 62 mA.  I've been testing with the OBi connected directly to the demarc, so the problem is theoretically not some other device.

[Mango]

Obihai's support responded quickly; unfortunately their advice was not to use the OBi with this POTS line

[drgeoff]

The OBi can only measure the loop current that is passing through itself.  There are only three things that can give a high reading for LINE port off-hook current.

1.  Faulty measuring circuit in the OBi.

2.  Faulty OBi LINE port.

3.  Too high a voltage being applied to the OBi's LINE port.  This ought to manifest itself as a too high TipRingVoltage for LINE port on-hook.  Normal for this would be say 49 +/- 4.

If this is a POTS line (metallic pair all the way back to the Telco CO) then there is very little possibility for any equipment the telco is responsible for to cause an abnormally high loop current.  (Abnormally low is a different matter.)

If possible try the following.  Get a line splitter or doubler or whatever they are called in the US.  It just puts two phones in parallel on the same line.  Use that to connect both an ordinary phone and the OBi110's LINE port to the telco line.  Plug a second phone into the OBi's PHONE port.

Make or receive a call with the OBi's phone using the telco line.  If the audio is distorted leave the OBi phone off-hook and take the other phone off-hook and listen to the quality in its earpiece.  Then put the OBi phone on-hook and listen again on the other phone.

Report what you hear.

[Mango]

Thank you very much for responding.  I checked TipRingVoltage.  It has been fluctuating between -51V and -53V.

This is a POTS line and is fed out of a CO.

I performed your test:

1) OBi phone off hook - audio severely distorted.
2) Phone directly connected to POTS line off hook - audio mostly normal, with slight buzz.  OBi phone slightly less distorted.
3) Other phone back on hook - OBi phone was severely distorted again.
4) OBi phone on hook and other phone off hook - audio completely normal.

I tested another OBi110.  The audio sounded different, but was still distorted.

[drgeoff]

That's puzzling.  Symptoms point the finger of suspicion at the 110 but two of them being faulty is less likely.

When you test with the second 110 are you also using a second PSU?

Having two 110s allows you to do another experiment.  You can connect the PHONE port of one 110 to the LINE port of the other 110.  The PHONE port of the first will act as a telco line in terms of providing CO battery voltage to the second 110.  See if the off-hook LINE current still fluctuates.  You can also compare the first OBi's PHONE port current measurement with the second OBi's LINE port current measurement.  They should be the same nominally.

You can also swap the two 110s over.  All observations should be the same as before.  If they are markedly different, something is wrong and probably exonerates the POTS line.

[Mango]

I redid tests with three OBi110 devices, and assigned some arbitrary numbers to them so that it is more clear about what is happening.

OBi110 #1, HardwareVersion 3.4: LINE Port LoopCurrent 62 mA, severe audio distortion.
OBi110 #2, HardwareVersion 3.4: LINE Port LoopCurrent 9 mA, very very quiet audio.
OBi110 #3, HardwareVersion 2.8: LINE Port LoopCurrent 62 mA, normal audio.

#3 has a damaged FXS port, so only works properly if the call is routed by IP to some other device.

Yes, I have tried multiple AC adapters.  I also checked them with a multimeter just now, and measured slightly over 12V.

I will try your other experiment later.

[drgeoff]

A lot of words at http://www.sandman.com/loopcur.html.  But what is genuinely useful and what is scaremongering?

[SteveInWA]

I read that long article by Sandman, and it makes sense to me.  Yes, he's selling hardware to mitigate or solve telco problems, but the advice is still sound.  I didn't find any FUD or misleading information there at all.

I don't understand the comments stating it's extremely unlikely that a POTS telco could be at fault.  Anything's possible.

The most definitive way to test this is to disconnect your premises (home, office, apartment, whatever) telphone wiring from the telco's service at their demarc box.  Assuming that you don't own or have access to a telephone "butt set" or a telephone line 600 Ohm load tester, use an ordinary telephone, preferably an old fashioned "2500" model desk phone, or some other phone that you can remove the case and disconnect the line cord.  Put a digital multimeter in series with the telephone line connections (inside the phone, disconnect one of the two wires leading to the modular line cord, then hook up the meter, one probe to the line cord side, and the other to the telephone.  Leave the phone off hook.  Set the meter to the highest current (amp) range.  Plug the phone's modular plug into the modular jack on the telco demarc box.  Read the current.  It should NEVER be as high as 62mA; that would fry electronics.  If it is significantly above 25mA, call the telco and demand that they look into it.

If it is within normal range, then you will need to troubleshoot the OBis.  Their built-in voltage and current measurement feature may be bogus or fried on your units, but I have a 110, hardware version 2.8, and it properly reports normal voltage and current on my POTS line.

[Mango]

drgeoff - thanks very much for the link.  I bought their loop current regulator and it performs as advertised.

Obihai kindly replaced the OBi110 that was under warranty.  I have installed the new OBi110 today behind the loop current regulator and it works perfectly.  It's somewhat unfortunate that the OBi110 does not self-regulate loop current, but since the device was replaced for free I can't really complain.

Hopefully this puts an end to my premature hardware failures.

[ProfTech]

That is all very interesting. My Obi is hardware 2.8. However there is a setting for the line port "CurrentLimitingEnable" that is UN Checked by default. The help screen says it will limit the loop current to 60 ma. Maybe the 2.8 hardware does a better job of handling excess current. My on hook line voltage reads 56 but the current only reads 29 ma with the box checked or unchecked. Doesn't change. Interesting find.

[sailing]

I will venture a guess as to what is happening. We are talking about a dc current so we can easily apply Ohm's Law. (If is was  an ac current, Ohm's law would still apply but it might complicate the explanation.) Ohm's Law is voltage=current x resistance. The voltage is the on hook reading with a voltmeter which is fixed. The current is dependent on the resistance. If you're close to the CO or if the wiring from the CO is larger then usual, the resistance will be lower. Hence you have too much current.

The original Ma Bell phones probably couldn't care less about currents that might be as high as 100mA. Electronic phones or adapters would be designed for a specific range. Currents over the expected range will cause clipping in the circuits, hence the distortion. Similar to overdriving audio amps.

Although this is a problem with today's electronics, the POTS and the Obi are not at fault. You just happen  to be the lucky one that has to deal with a worst case situation.

[Mango]

If you're close to the CO or if the wiring from the CO is larger then usual

Yes to both.  I'm a very short walk from the CO, and the wire - at least in the demarc where I have access to it - is larger than at any other house I've lived in.  It looks to be about 18 AWG.  Is that what you were referring to?

[sailing]

Yes. You have both conditions to help cause the high loop current.