luce@aurs01.UUCP (J. Luce) (01/02/91)
I run a BBS using a 2400 baud modem (I have tried 2 flavors of modem with the same thing occurring). The Caller gets line noise garbage and the BBS sees nothing but a clean line. I have called my own BBS on a voice line in the house and have seen it myself. The BBS phone line has it's own jack and own inhouse wiring. Well, almost, there is a 2nd line on it used for a 2nd modem at 1200 baud. The line used to split at the terminal box on the house and 1 line went to the Phone company stump on one outdoor undeground cable and the other on another. I thought the bridge from 1 cable that terminated on another block and brought over to the BBS was faulty so I had them put the second line onto the same cable hung off the same terminal block. Well, it appears to have gotten worse since then. Any thoughts? This does occur even when the 2nd modem is OFF. And as I said, it is only the caller that sees the noise, not the BBS... Please respond via this newsgroup as mail *NEVER* reaches me. Thanks.
floyd@ims.alaska.edu (Floyd Davidson) (01/03/91)
In article <59403@aurs01.UUCP> luce@aurw46.UUCP (J. Luce) writes: >I run a BBS using a 2400 baud modem (I have tried 2 flavors of modem >with the same thing occurring). The Caller gets line noise garbage and >the BBS sees nothing but a clean line. I have called my own BBS on a >voice line in the house and have seen it myself. The fact that it is in one direction only indicates one of two things. It is either a relatively narrow frequency range that is interefering or it is occuring someplace where you circuit is a four wire circuit. Cable noise is almost never a restricted frequency range source. Sometimes crosstalk with another modem might give you that, but on a residential line it is unlikely. Your local house wiring is all part of a two wire circuit. Any wide band noise burst or crosstalk with a voice circuit will affect both the transmit and receive. Since you have tried more than one modem and still have the problem it is fairly clear that the phone company has a problem. If you are on copper cable all the way to the switch it is most likely the line card in the switch. However if you are some distance from the switch it may be on some form of cable carrier between you and the switch, and the cable carrier is 4-wire (a separate channel for transmit and a separate channel for receive). If you are getting garbage that is totally random it likely is a line card or a channel card (switch or carrier). If you get repeating characters (like: {{{{{{{{) it is a T-carrier system that is not syncronized (clock slips). Since you get the noise when you call from your other line, and you more or less indicate that almost every call is noisy, I think you can eliminate everything past the line card in the switch. But if some calls are clear and others are noisy, then you do need to note where the noisy ones come from and where the clear ones come from and look for a pattern. If noise is only when it is a call local to your line switcher, but clears up if the call comes from another line switcher, then the problem is past your line card. In that case it would be in the "network" or whatever they call the switching matrix on that particular switch. All of the above assumes a 4-wire switch (eg. digital). If you are on an old mechanical switch then it has to be some kind of carrier. When you talk to the phone company get a tech on line or at least ask the clerk on the trouble desk to have a tech call you. Floyd Disclaimer: I don't work on line switchers, only trunks. -- Floyd L. Davidson floyd@ims.alaska.edu Salcha, AK 99714 paycheck connection to Alascom, Inc. When I speak for them, one of us will be *out* of business in a hurry.
gt5302b@prism.gatech.EDU (R. Steve Walker) (01/07/91)
In article <59403@aurs01.UUCP> luce@aurw46.UUCP (J. Luce) writes: >I run a BBS using a 2400 baud modem (I have tried 2 flavors of modem >with the same thing occurring). The Caller gets line noise garbage and >the BBS sees nothing but a clean line. I have called my own BBS on a >voice line in the house and have seen it myself. > Try BallCo Inc. at (404) 979-5900. They sell line noise eliminators for modems. It works.
floyd@ims.alaska.edu (Floyd Davidson) (01/07/91)
In article <19239@hydra.gatech.EDU> gt5302b@prism.gatech.EDU (R. Steve Walker) writes: >In article <59403@aurs01.UUCP> luce@aurw46.UUCP (J. Luce) writes: >>I run a BBS using a 2400 baud modem (I have tried 2 flavors of modem >>with the same thing occurring). The Caller gets line noise garbage and >>the BBS sees nothing but a clean line. I have called my own BBS on a >>voice line in the house and have seen it myself. >> >Try BallCo Inc. at (404) 979-5900. They sell line noise eliminators for >modems. It works. Except that he does NOT have line noise, so it won't be filtered out. Notice he always has a "clean line" in the direction of the BBS, and "line noise garbage" toward the caller. That isn't caused by noise on the line. The line is a two wire device which has two way transmission. Noise on the line will garble traffic on both ends, although it may be more significant on the receiving end of the line that actually is causing the noise (lower signal to noise ration for the recieved signal). What he is experiencing is one way interference. That is a character trait of "4-wire" circuits where each direction has a separate transmission path all to itself. It is probably a bad analog-to-digital converter in the line card at the switch. Line filters just won't do much for it. Floyd -- Floyd L. Davidson floyd@ims.alaska.edu Salcha, AK 99714 paycheck connection to Alascom, Inc. When I speak for them, one of us will be *out* of business in a hurry.
gundrum@svc.portal.com (01/08/91)
> The BBS phone line has it's own jack and own inhouse wiring. Well, > almost, there is a 2nd line on it used for a 2nd modem at 1200 baud. I know this may seem stupid, but try tesing the line with the second modem physically disconnected from the line. We spend a lot of time in our office daisy-chaining modems. We have occasionally noticed cheaper modems causing noise on the line when they are turned OFF. I don't know enough about the technology to say why this happens, but it is easy enough to check out, why not give it a shot? _______________________________________________________________________ Any statements made by this account are strictly based on heresay and should be assumed to have no intelligence behind them. (No, that does not mean they have the approval of management.) gundrum@svc.portal.com
luce@aurs01.UUCP (J. Luce) (01/09/91)
In article <0B1gV2w163w@ersys.uucp> ersys!davem@nro.cs.athabascau.ca (Dave McCrady) writes: >gundrum@svc.portal.com writes: > >> > The BBS phone line has it's own jack and own inhouse wiring. Well, >> > almost, there is a 2nd line on it used for a 2nd modem at 1200 baud. >> >> I know this may seem stupid, but try tesing the line with the second >> modem physically disconnected from the line. We spend a lot of time >> in our office daisy-chaining modems. We have occasionally noticed >> cheaper modems causing noise on the line when they are turned OFF. >> >> I don't know enough about the technology to say why this happens, but >> it is easy enough to check out, why not give it a shot? >> > > I had a similar experience that drove me nuts until I remembered that I >had the modem in my Toshiba laptop connected to the line. Taking it off >solved it ... would appear some equipment doesn't like to share... > > > Dave McCrady | davem@ersys.uucp >+1 403-454-4054 (voice) | ersys!davem@nro.cs.athabascau.ca >+1 403-454-6093 (public USENET) | decrwl!alberta!aunro!ersys!davem The second modem is on a different pair of wires inside the same 4-wire cable. So, I cannot believe that when the modem is shut off that it creates a problem :) OK, 1 other thing (BTW, Thanks FLOYD for the reply, it has caused the following to come to light): After harrassing SoBell to get me off my current line card, I found I was on a SLC-96 attached to an AT&T FP-20 (??? not sure). Anyway, it is the oldest switch in the Raleigh area and is ANALOG. Since the move to a new frame and a new line card, the problem is the same. Supposedly I am on copper all the way to the switch (I find it AMAZING no one knows for sure...). OK, let's assume I am, is the T-clock still a valid concern. Since the change of line card, the noise persists but now is of repeating characters and not just a jumble of characters. Thanks for the help...
floyd@ims.alaska.edu (Floyd Davidson) (01/10/91)
In article <59420@aurs01.UUCP> luce@aurw46.UUCP (J. Luce) writes: [...] > >After harrassing SoBell to get me off my current line card, I found I >was on a SLC-96 attached to an AT&T FP-20 (??? not sure). Anyway, it is >the oldest switch in the Raleigh area and is ANALOG. Since the move to a >new frame and a new line card, the problem is the same. Supposedly I am >on copper all the way to the switch (I find it AMAZING no one knows for >sure...). > >OK, let's assume I am, is the T-clock still a valid concern. Since the >change of line card, the noise persists but now is of repeating >characters and not just a jumble of characters. > >Thanks for the help... Wow this is really getting interesting. I'm not familiar at all with the various kinds of line carrier or line extenders. Or for that matter with ATT switches. I don't know what an SLC-96 or an AT&T FP-20 are. I find it more than amazing if no one knows what kind of a loop you are on. More like the individuals you are talking to don't want to bother looking it up. One thing needs to be clarified. Is this *always* and only in one direction? Say like in a ten minute period how many hits do you get in each direction? The single character being displayed sounds very much like clock slips on a T1 carrier. Usually it is '}}}}}}}' or '{{{{{{{{' I can't remember which. With maybe on odd character at the start or end of the string. If you are getting a pattern like that and it is an analog switch, then you must have some kind of a digital loop extender on your line. Once again, it still sounds like a 4 wire circuit problem, not noise on your 2 wire drop. Floyd -- Floyd L. Davidson floyd@ims.alaska.edu Salcha, AK 99714 paycheck connection to Alascom, Inc. When I speak for them, one of us will be *out* of business in a hurry.
tnixon@hayes.uucp (01/10/91)
In article <1991Jan9.075511.2269@ims.alaska.edu>, floyd@ims.alaska.edu (Floyd Davidson) writes: > Ten to one it is the line card. Either that or it isn't always > one way, just most of it. Probably the latter. It will be interesting to see if he gets any response from the telco. -- Toby Nixon, Principal Engineer | Voice +1-404-449-8791 Telex 151243420 Hayes Microcomputer Products Inc. | Fax +1-404-447-0178 CIS 70271,404 P.O. Box 105203 | UUCP uunet!hayes!tnixon AT&T !tnixon Atlanta, Georgia 30348 USA | Internet hayes!tnixon@uunet.uu.net
luce@aurs01.UUCP (J. Luce) (01/11/91)
In article <1991Jan10.042007.4227@ims.alaska.edu> floyd@ims.alaska.edu (Floyd Davidson) writes: +In article <59420@aurs01.UUCP> luce@aurw46.UUCP (J. Luce) writes: +[...] ++ ++After harrassing SoBell to get me off my current line card, I found I ++was on a SLC-96 attached to an AT&T FP-20 (??? not sure). Anyway, it is ++the oldest switch in the Raleigh area and is ANALOG. Since the move to a ++new frame and a new line card, the problem is the same. Supposedly I am ++on copper all the way to the switch (I find it AMAZING no one knows for ++sure...). ++ ++OK, let's assume I am, is the T-clock still a valid concern. Since the ++change of line card, the noise persists but now is of repeating ++characters and not just a jumble of characters. ++ ++Thanks for the help... + +wow this is really getting interesting. + +I'm not familiar at all with the various kinds of line carrier or +line extenders. Or for that matter with ATT switches. I don't +know what an SLC-96 or an AT&T FP-20 are. + +I find it more than amazing if no one knows what kind of a loop you +are on. More like the individuals you are talking to don't want to +bother looking it up. + +One thing needs to be clarified. Is this *always* and only in +one direction? Say like in a ten minute period how many hits do +you get in each direction? + +The single character being displayed sounds very much like clock +slips on a T1 carrier. Usually it is '}}}}}}}' or '{{{{{{{{' +I can't remember which. With maybe on odd character at the start +or end of the string. + +If you are getting a pattern like that and it is an analog switch, +then you must have some kind of a digital loop extender on your +line. + +Once again, it still sounds like a 4 wire circuit problem, not +noise on your 2 wire drop. + +Floyd +-- +Floyd L. Davidson floyd@ims.alaska.edu +Salcha, AK 99714 paycheck connection to Alascom, Inc. + When I speak for them, one of us will be *out* of business in a hurry. It is a CONTINUAL problem with occasional bursts AT the caller. The repeating character is not a brace of any sort, either a checkmark (IBM PC stuff) or a 2 letter combo that the extended PC char set assumes is what it is printing. Re: Filter. Someone (I think Toby Nixon) stated that there was a filter beast that I could get that could help the problem (maybe). I would like to try that also but no one has such a thing (filter AM radio signals, yes, but to help what is on the line, no way). So anyone who can give me info on this (either where to get 1 or how to make 1), it would be greatly appreciated. Nope, AT&T phone store, Radio Shack, and my local CompuAdd have no idea where to get one. I want to thank everyone for their help. I'll keep y'all updated. ------------------------------------------------------------------- John Luce | Life is the leading cause of death Alcatel Network Systems | ----------------------------------------- Raleigh, NC | Standard Disclaimer Applies 919-850-6787 | Mail? Here? Try aurs01!aurw46!luce@mcnc.org ------------------------------------- or luce@aurs01.UUCP
laird@chinet.chi.il.us (Laird J. Heal) (01/13/91)
In article <59426@aurs01.UUCP> luce@aurs01.UUCP (J. Luce) writes: >In article <1991Jan10.042007.4227@ims.alaska.edu> floyd@ims.alaska.edu (Floyd Davidson) writes: >+In article <59420@aurs01.UUCP> luce@aurw46.UUCP (J. Luce) writes: >+[...] >++ >++OK, let's assume I am, is the T-clock still a valid concern. Since the >++change of line card, the noise persists but now is of repeating >++characters and not just a jumble of characters. >++ >+ >+The single character being displayed sounds very much like clock >+slips on a T1 carrier. Usually it is '}}}}}}}' or '{{{{{{{{' >+I can't remember which. With maybe on odd character at the start >+or end of the string. >+ >It is a CONTINUAL problem with occasional bursts AT the caller. >I want to thank everyone for their help. I'll keep y'all updated. I am surprised that nobody mentioned the biggest problem with the aluminum wire. When someone noted that an intermittent-contact junction could cause some of the symptoms I was sure it would be brought up. Aluminum wire was briefly tried in houses; there are not too many such houses left. There was this problem with electrical fires. It seems that the aluminum wire would expand with heat at joints, but being less ductile than copper it was consequently not as resiliant when it would cool. The result was, wouldn't you know, a gradual receding of the wire until the contact was so small as to create a great deal of heat - enough to start a fire. Telephone circuitry passes far less current than 120VAC, but the ringer voltage is certainly a periodic jolt, possibly enough to cause momentary heating. This would induce the gradual loss of contact which was described as a cause consistent with these symptoms. Take that wire to the recycling center. -- Laird J. Heal The Usenet is dead! Here: laird@chinet.chi.il.us Long Live the Usenet!
enger@seka.scc.com (Robert M. Enger) (01/14/91)
SLIC-96 is indeed a "carrier" based, pair-gain system. If you are using a SLIC-96, then you are NOT copper all the way to the CO. (slic stands for something like Subscriber Line Interface Concentrator). The SLIC-96 hubs upto 96 subscriber lines into a central point (usually a controlled environment vault, CEV). From there, 4 T1 circuits convey the info back to the CO. It is possible that real T1's are not evn used. Often, the T1's are themselves further multiplexed together and placed on a Fiber span. At any rate, all of this means that you are indeed in "4-wire" mode between your premesis and the CO. So, you may have problems wiht the interface card in the SLIC, etc. SECOND point. From what I've read of the problem, the situation is asymmetric: Outside callers dialing INTO your modem see the problem. The following experiment might provide an interesting data point: find a willing outside caller that REPEATABLY gets the problem when dialing into your modem. Then, try initiating an OUTGOING call from your troublesome modem TO the outside caller. This will reverse the usage of the low/high frequencies. If the problem reverses, then maybe it is a frequency dependent line degredation of some sort. Best wishes, Bob Enger enger@seka.scc.com
luce@aurs01.UUCP (J. Luce) (01/15/91)
In article <1991Jan13.183631.27081@europa.asd.contel.com> enger@seka.scc.com writes: > >SLIC-96 is indeed a "carrier" based, pair-gain system. [Stuff deleted] >and placed on a Fiber span. > >At any rate, all of this means that you are indeed in "4-wire" mode >between your premesis and the CO. So, you may have problems wiht >the interface card in the SLIC, etc. > True enough. > >SECOND point. From what I've read of the problem, the situation >is asymmetric: Outside callers dialing INTO your modem see the problem. > >The following experiment might provide an interesting data point: >find a willing outside caller that REPEATABLY gets the problem when dialing >into your modem. Then, try initiating an OUTGOING call from your >troublesome modem TO the outside caller. This will reverse the usage >of the low/high frequencies. If the problem reverses, then maybe it >is a frequency dependent line degredation of some sort. > >Best wishes, >Bob Enger >enger@seka.scc.com This is called Engineering stupidity on my part. Bob, that is something I should have done early on but did not (can you say 'Dumb' boys and girls? I knew you could). I did just that last night after getting this and Voila!! there was the line noise, BUT!!! Outbound once again. I saw the line noise (due to the other end echoing I think, because the crap showed up in the msgs I would enter). ::sigh:: OK, campers, what is next. Is there anything I can do to increase my outbound signal to basically 'pierce' the noise (you can tell I write s/w and not 'do' hdwe). Or is that a fantasy. Anyway at all I can take some control from my end. Oh, yes, a comment from the SoBell Sr. Tech ... "You will probably need a data line put in, but that is expensive...". Harrumph... Thanks once again for all your help, at least I ask intelligent questions to drive SoBell nuts... :) ------------------------------------------------------------------- John Luce | Life is the leading cause of death Alcatel Network Systems | ----------------------------------------- Raleigh, NC | Standard Disclaimer Applies 919-850-6787 | Mail? Here? Try aurs01!aurw46!luce@mcnc.org ------------------------------------- or luce@aurs01.UUCP
luce@aurs01.UUCP (J. Luce) (01/23/91)
In article <1991Jan20.204546.5398@europa.asd.contel.com> enger@seka.scc.com writes:
-John:
-
-You responded that the DIRECTION of data corruption DID NOT CHANGE
-when you reversed the roles of the tx/rx tones.
That is correct...
-
-If the problem were frequency related, then it would seem that
-the 'direction' subject to the errors should have reversed.
-Since it didn't, maybe we can assume that the problem is purely
-directional in nature.
My contention all along and 1 that was suggested by Floyd at Alaska
Com(?).
-
-You mentioned that you were feeding into a SLIC.
I was moved completely off the SLIC to a different frame. *NOW* the
SoBell people tell me I am on a No. Telecom DMS-100 digital rather the
aforementioned AT&T analog switch.
-If that unit is close by, perhaps its possible that you're
-overdriving its input. You might want to try padding down (REDUCING)
-the output level of your modem to see its effect.
I am quite a ways away and go through a 'box' containing T1 cards onto a
fibre down to the C.O. I am about 2 miles from the T1.
-
-The only remaining avenue open to you may be to perform a full technical
-analysis of your line. In theory, you should be able to get telco
-to dispatch a tech with a transmission test set to your premesis.
-They should be able to test for distortion at various signal levels/frequencies,
-noise on the line, and other degredations.
I have requested that and have gotten the old "We only check the center
of the freq. range, we can't do a sweep'.
-
-If you can't get telco to cooperate, you could rent the test equipment
-yourself and take a stab at the diagnosis.
-
-Good luck,
-Robert M. Enger
-CONTEL Federal Systems
-enger@seka.scc.com (Internet)
Also, I should note here that I am atteched inside to an RJ45S
connector now, the kind they use on 'data conditioned' line. I was told
(confirming Floyd's contention) that this is the only difference between
a 'voice' line and a 'data' line. Actually, there is another, a data
line is billed at business rates :)
I have been referred internally to some one 'who knows these things...'
and he is going to get a data-line knowledgeable sort on the case. Of
course that was a week ago and ... (you know, when cows can fly) :)
As for the test eqpt... being a s/w engineer, I am ignorant of a lot of
that stuff (and plan on staying that way <heheheh>).
Thanks, for the info. If you have more ideas, let me know.
-------------------------------------------------------------------
John Luce | Life is the leading cause of death
Alcatel Network Systems | -----------------------------------------
Raleigh, NC | Standard Disclaimer Applies
919-850-6787 | Mail? Here? Try aurs01!aurw46!luce@mcnc.org
------------------------------------- or luce@aurs01.UUCPenger@seka.scc.com (Robert M. Enger) (01/24/91)
In article <59468@aurs01.UUCP>, luce@aurs01.UUCP (J. Luce) writes: |> -You mentioned that you were feeding into a SLIC. |> |> I was moved completely off the SLIC to a different frame. *NOW* the |> SoBell people tell me I am on a No. Telecom DMS-100 digital rather the |> aforementioned AT&T analog switch. |> |> |> -If that unit is close by, perhaps its possible that you're |> -overdriving its input. You might want to try padding down (REDUCING) |> -the output level of your modem to see its effect. |> |> I am quite a ways away and go through a 'box' containing T1 cards onto a |> fibre down to the C.O. I am about 2 miles from the T1. |> Hello again: The "box' containing the T1 cards and a fiber connection IS THE SLIC. The slic not only contains T1 cards, it also contains subscriber interface cards. The subscriber (station line) interface card that you are connected to in that "slic" MAY BE BAD, or improperly configured. Another problem affecting your performance may be an improperly installed loading coil. Telco installs equalization coils on subscriber lines more than so-many miles in length. (the figure I got while asking around the office was about 1-1/2 miles, which would qualify your circuit). Coils are used to smooth the frequency response for voice use, but play havoc with the phase response of the circuit. As a side note, telco removes such coils from cable pairs used for Digital Dataphone Service (DDS) digital data circuits. I hope you get to the bottom of this problem. And when you do, please post/e-mail the cause/remedy info to us. Thanks! Bob --- Robert M. Enger CONTEL Federal Systems enger@seka.scc.com (Internet)