Could high audio level from the telco make the PBX user cutout on the TX side of a call. Outside call comes in on analog trunk ShoreTel user answers. Coversation is find at low levels but if the outside caller raises voice to medium high level the ShoreTel user cuts out when they speak at the same time. There are no other issues but this seemingly half duplex issue.
If the db level and loop current is to high on the line for the spec's of the Shoretel system wouldnt the high db level step on the TX side (ST user).

So I would like to reduce the db/loop current on the line. Wouldn't reducing the loop current reduce the db level also and or reducing db level reduce loop current.
I haven't tested the lines yet, but I think this is what is happening. What else may cause half duplex issues?

What is the loop current? Yes reducing loop current will reduce volume some. The clipping would be more of a condition of the IP phone system than the analog line. At least in my opinion. Test from DEMARC to prove this and let us know the off/hook and on/hook voltage and the loop current.

I agree that its probably the ST system. But I have to change the line to meet the VOIP system spec's, that is just the nature of the beast there is no settings I can adjust on the ST system.
I will update with test findings.

Thanks

If the loop current is high you can get loop current regulators from Sandman. If there are other places to get them I'm not aware of it. At least the constant ones. I have found those with different taps, but not near as handy. I'd certainly like to find a less expensive source.

Link to Sandman\'s regulators

Yeah I was looking at that site. I found the loop current regulator, echo stopper and the telephone line attenuator. I'm thinking the line attenuator looks like it will be my best choice. Since it reduces the db and loop current at the same time.
I just hope that is the issue?

^^^^ What Bill, said.


Is the analog signal and loop current really THAT high at the demarc? Did you test there as Bill suggested? ... And If you’re are going to attenuate for louder conversation what’s going to happen to normal conversation volumes or worse yet for quiet talkers? ... Just something to think about there.

Telephony is a a science that follows established electrical rules. There ain't any magic here. There are three ways to evaluate and solve telephone circuit problems.

1. measurements

2. substitution

3. hand-waving

As has been suggested, please measure the loop current. Then please tell us what it is. Then we will help you.

Can you substitute an incoming line using some other, proven, (read: non-voip nonsense) technology? Is there a POTS line within wiring distance of the KSU in the building (even if can be temporarily borrowed from a co-operative neighbor)?

Without knowing base-line factors, and comparing them to industry standards; and without eliminating the various parts of a suspect circuit, the rest is hand-waving.

An engineer should not suspect that turn of events. High loop current is not a linear problem.

“Loop current” would be measuring the DC component of what is found on a phone line. Georg Ohm figured out the directly proportional (linear) relationship of voltage and resistance back in the early 1800’s.

A voice signal transmitted over a copper wire is a sinusoidal waveform (AC) positioned on top of the DC component. So now if you meant the “voice signal” that is measurable in decibels, is a non-linear (logarithmic) scale, that would in fact be true.

I think I’ll stand by my concern, since in this case, it would seem an attenuator of any kind would be masking a symptom rather the correcting the issue.

IMHO unless he's right on top the CO or feed by carrier that's misaligned levels are not going to be an issue. I'd bet loop current isn't the issue either. I'm sticking with the problem lies with-in the IP system. Of course we won't know any of this for sure until we see the test results.

How can these IP developed systems all at once say the established requirements don't apply to us? At least that's the way I'm reading what mgere is telling us that he's being told by the IP systems tech support.

I’m still thinking you IMHO is correct, sir.

Disregarding long standing requirements and standards for equipment is happening on this side of the demarc too I’m afraid. It seems to be the way of the world when the CG’s and marketing people are driving the proverbial bus.

....................


I don’t know that this will specifically help the OP but typed this out for anyone else that happens to look here due to having similar issues and would like a little deeper discussion.

The DC component (loop current) found on telephone lines and the AC signal (voice) when present, is superimposed but the two elements are wholly separate. If you don’t believe me? … Place a call... Mute both ends so neither are transmitting or receiving any signal. Is the DC still flowing? Yup, sure enough is, the talk battery is still there.

On a POTS line, since the DC component is required to there to keep the line open at the switch, as well as power the transmitter and receiver of a telset, you can’t really “prove” that AC signals can exist without the DC on a regular POTS line… But there are a few examples of “special service circuits,” provided by telcos, that have no DC component but pass only AC signal over copper wires. Pretty much a thing of the past, but once upon a time it was done (i.e. old school analog data, hoot-n-hollar/junkyard lines, etc.) … And, I’ll leave the inherent corrosion issues on splice and sealing-current open to discuss for a different time, since this is already getting off track.

So, why is there even a conversation here (and in other posts) that seeming mixing and matching concerns with “loop current” and “signal?” ... It’s the nature of how the telco plant is constructed. The DC current reading can be VERY telling since a voice line is an “electrical circuit” wherein the conductors make a majority of the load and the battery is a fixed value…. That Ohm was a smart feller, right.

When questions of “too hot of a signal” come up, the follow-up question is often what’s the “loop current” at the demarc.... Why if they are separate things? A person can make some very accurate assumptions about “the loop” the particular voice service is being delivered on... Or how “the loop” was conditioned for service delivery.

If the signal is “too hot” and the DC current is high, one of two likely conditions is present… And both are telco issues. 1) The loop is very short and the equipment on the other end was installed at default settings that are designed for more average (longer) loops and needs to be padded down. 2) The loops is very long and the telco over compensated when conditioning for the higher resistance and “loop loss” of the extra cable.

I did test the loop current and found it to be 49ma and 45ma on the two analog trunks. I put the attenuation tester in line and dropped the loopcurrent by 4ma and the db by 4, but it didn't help. I also, just for shits and giggles, put a impedance matcher in line and that did nothing. The CO is probably 350 feet (crow flys) from the demarc of the place of buisness. I will call Verizon and see if they can get the loop current down on those lines. Im totally at a lose with this one.

You could look on sandman.com. He has several devices that can be installed in line to reduce the loop current. I believe there is also a diagram to build a resister and capacitor circuit to reduce it.

While that loop current is on the high side it still isn't going to cause the problems you describe. Loop current around this area is constantly in the 50-60ma range. It does affect something's you just have to learn what. Actually those who manufacture the products should allow for this as the loop current does have a low requirement, but the high is to anyones interpretation. I found this out after fighting with Qwest for sometime. Read Sandman's article about loop current.

That's enough to burn out some trunk cards! Been there!

I have to disagree, justbill, I have had the same type of issue with loop current in the 40 to 45 milliamp range. I built a capacitor-resistor circuit, used it like a bridge clip between the CO and the switch. This dropped the current to just under 30 milliamps and the problem cleared.

oldfashioned,

Why the capacitors? I have seen this done with just straight resistors in series with the line, but I have never seen the circuit you describe. Can you describe it and/or post a schematic?

Enquiring minds want to know...

Jim
**************************************************
Speaking from a secure undisclosed location.

The capacitors would bypass the audio around the resistive attenuator pad so that the audio wouldn't be attenuated, only the DC voltage. I don't know if that's an advantage or not.

-Hal

You had a problem with voice clipping and it was cause by high loop current? That's a new one on me.

I opened a ticket with verizon we will see what happens. I have 30 ShoreTel customers and 3/4 of them use analog lines to some capacity, not a one has this issue. Could something in the CO cause this issue, like a bad card or something??

I'm at my wits end, I even ajusted the trunk gains in the database to see if that would help, nothing.

If it's not doing it from the DEMARC it's not Verizons trouble.

It's not out of the realm of reality that the high loop current is causing the problem. Esp. if there is a transformer interface to the C.O. line. But, as has been already pointed out, high loop current is not a concern of Verizon, or any other LEC. I suspect that if you take a DC voltage check on-hook, you'll find something in the neighborhood of 52 Volts DC, rather than 48 Volts DC. Remember that the LEC would like to have a high enough voltage out of the central office to support communications out several thousand feet. Resistance being what it is, and the need to maintain 5-7 Volts DC off-hook, they start out high. Your problems probably will go away by getting your off-hook voltage down to 5-7 Volts DC and your off-hook current down to 28-35 milliamps. The LEC is not going to do it! It's your problem, they're within their spec!

Try this, put a 100 ohm, 1/4 or 1/2 watt, resistor in series with the tip lead and another in series with the ring lead. This will drop off-hook voltage and current, although probably not enough. Next steps would be 220 ohm and 330 ohm.
You must put equal resistances in the t and r leads in order to maintain longitudinal balance, don't cheat. This may reduce audio too much, if so try a .47 uf 200 volt non-polarized (no plus or minus sign on the capacitor) capacitor in parallel with each resistor. All these parts used to be available at Radio Shack. But, they've been cutting their component inventory, so you may have to go elsewhere. DIGI-KEY would be a choice, but they have a $20? minimum and these parts won't be more than $5, even at R/S prices.

While I agree that you can lower current and voltage as you describe Lightning, I have never seen a condition affecting voice that couldn't be duplicated at the DEMARC with a butt set. I'm not talking about high loop current affecting the key equip, modem or whatever, just what the OP has discribed, voice clipping.

I call Verizon and had a tech on-site, he read the 50ma and said what do you want me to do LOL!!
Asked if he could walk to the CO and see if they could do anything about lowering the loop current, guess what they said, pack sand. That was to be expected, I had to try.

Can you old school guys tell me what I need to buy as far as components (resistors/capacitors) that will drop the current down and maybe the db too.
I at the point I will try anything, even if its a long shot.

Re-read the 2nd paragraph of my last post. Ask specific questions based on it. John C.

Ok, so I will get two 220 resistors and put them inline with tip and ring. Where is the best place to get those?? I will install those inline and post my findings.

If you're going to lower the loop current get the ones from Sandman. They keep a constant current no matter what the phone company throws at you. With resistors you'll have to experiment until you get what you want and the conditions could change.

I'm going to say this once more to maybe get through. High loop current will not cause clipping.

Bill has the best idea. Resistors are cheap, but things could change and throw all your hard work out the window.

Bill, I'll agree that the loop current itself is not directly the problem, it's the inability of the equipment to cope with it. It's a design problem.

There-in lies the problem. They are trying to reinvent the wheel instead of adhering to the established standards.

I carry a half a dozen of the sandman regulators and if loop current is even a little high I throw them on to see if they have any effect

they usually don't so I pull them off and continue looking for the cause

when they help the customer gets billed and I buy more .

Well finally this issue may get resolved today. I'm getting some help from ShoreTels escalation engineer. He has seen this problem before and he calls it audio clipping because of Echo (ERL). (I called it half duplex). He thinks the issue is with the DSPs in the ST PBX switch not having the ability to correct the echo (ERL) from the CO. Usually this happens out in rual areas where the CO is miles away from the customer.
The only confusing thing is that the CO is 300 feet from the customer. I have to install a couple of patch files on the switch to correct this issue, I will let you all know if that was the problem. What does ERL stand for anyway??

ERL=Echo Return Loss. A problem with 4wire to 2wire hybrids not a problem of POTS. But I have heard of Echo problems on VOIP lines. Once again because they refuse to follow industry standards.

I suppose what's next is putting talk back amps into circuits so people know they haven't been cut off. That's about a 1940's technology.

Well the new DSPs didn't work. So back to the drawing board. They are working on correcting the DSP issue and will release new ones to me shortly.