Kitz Forum

Broadband Related => Telephony Wiring + Equipment => Topic started by: konrado5 on September 07, 2014, 11:03:16 PM

Title: My cables
Post by: konrado5 on September 07, 2014, 11:03:16 PM
I've took more accurate photos of my cables. Is it all right? The horizontal cable entering into the white round box is phone cable. The large white rectangular box is doorbell. The doorbell cable cross with phone cable at 90 degrees. Furthermore, cable which is parallel to phone cable nearly the white round box is probably additional doorbell cable. Modem is connected to the phone socket by 10 meters long CW1308 (if CW1308 is polish YTKSY). Is phone cable connecting phone socket with white round junction box CW1308 cable?

I'm not sure if this post should be in this thread:
http://forum.kitz.co.uk/index.php?topic=14388.0

I've written about my cables, but I've took very poor photos some time ago.

(http://s25.postimg.org/3voj637l7/ca_o.jpg) (http://postimg.org/image/3voj637l7/)
(http://s25.postimg.org/5qjr6so4r/dzwonek.jpg) (http://postimg.org/image/5qjr6so4r/)
(http://s25.postimg.org/4hrqqsnqj/puszka.jpg) (http://postimg.org/image/4hrqqsnqj/)
(http://s25.postimg.org/xj59romy3/gniazdko1.jpg) (http://postimg.org/image/xj59romy3/)

Best regards
konrado5
Title: Re: My cables
Post by: konrado5 on September 09, 2014, 09:03:03 PM
Probably I use white-blue pair of phone cable from junction box to phone socket. It is noticable on following photo:
(http://s25.postimg.org/lyee5osvf/gniazdko2.jpg) (http://postimg.org/image/lyee5osvf/)

Best regards
konrado5
Title: Re: My cables
Post by: kitz on September 09, 2014, 09:45:13 PM
Wiring isn't exactly my forte unless it's something I'm familiar with... So I'll leave it to the other guys to comment.  sorry.
Title: Re: My cables
Post by: burakkucat on September 09, 2014, 10:03:49 PM
I have downloaded copies of those five images (and rotated three of them) for careful study.

Looking at images numbers 4 & 5 I can see that the flexible lead to the telephone and modem only uses the red and green wires. (I am assuming that there is a low-pass filter for the telephone and the filter, telephone and modem are connected correctly at the other end of that flexible cable.) The internal cable, used to connect the rectangular box with the circular box, looks to be standard 4-core, 2-pair cable and, I would it expect it to be equivalent to CW1308 specification cable. Assuming that they are two twisted pairs, the colouring of the wires' sheaths could cause problems. I see a blue wire & a white wire making up one pair and an orange wire & a white wire making up the second pair. It is the first pair (blue wire/white wire) that is being used. If I was being super-critical, I would mention that there should not really be any exposed metallic conductors showing -- look at the ends of the blue & white wires where they are connected under the screw terminals that hold the green & red wires, respectively.

Looking at image number 3 I can clearly see that the incoming service feed cable (with the black sheath) is a 4-core, 2-pair cable. Once again, the wire colours could cause a problem because they appear to be a red wire & and a clear wire as one pair and a blue wire & a clear wire as the second pair. It is the latter pair (blue wire & clear wire) that is being used and it is connected to the blue wire & white wire pair of the internal cable by means of two gel-crimps. The unused pair of the service feed cable (red wire & clear wire) and the unused pair of the internal cable (orange wire & white wire) are left unconnected. Again, if I was being super-critical, I would suggest that the unused orange wire & white wire pair in the internal cable should not have been left with exposed conductors, twisted together.

Looking at image number 2 it seems as if the service feed cable has been threaded through the existing hole which contains the wire for the door bell. There is not a lot that can be done -- unless you are prepared to have those two cables separated and a new hole created for the service feed cable.

Overall there is nothing particularly "bad" about that wiring and I can't see it being a cause for a degradation in the broadband service. I know that you have noticed that errors occur when somebody rings the door bell. One possible solution would be to fit a spark-suppressing capacitor/resistor series pair across the push button.
Title: Re: My cables
Post by: konrado5 on September 09, 2014, 10:10:27 PM
Thank you very much for your reply.
Quote from: burakkucat
I am assuming that there is a low-pass filter for the telephone and the filter, telephone and modem are connected correctly at the other end of that flexible cable.
I has not any phone any longer. I have only modem connected to the socket by 10 meters longRJ11 ended CW1308 0.5 mm diameter. Is it something which can be removed if I have not phone? In the past, I've had bridge tap from the socket to the socket in other room.

Quote from: burakkucat
know that you have noticed that errors occur when somebody rings the door bell. One possible solution would be to fit a spark-suppressing capacitor/resistor series pair across the push button.
I've also noticed errors when I turn off the lightning, when I unplug from the mains computer speakers. Phone cables are nearly the mains box (pink painted box) above the small white square connector.
(http://s25.postimg.org/5qjr6so4r/dzwonek.jpg) (http://postimg.org/image/5qjr6so4r/)

Quote from: burakkucat
Again, if I was being super-critical, I would suggest that the unused orange wire & white wire pair in the internal cable should not have been left with exposed conductors, twisted together.
What are the possible effects of exposed conductors?

Quote from: burakkucat
One possible solution would be to fit a spark-suppressing capacitor/resistor series pair across the push button.
What is spark-supressing capacitor?

Best regards
konrado5
Title: Re: My cables
Post by: burakkucat on September 09, 2014, 11:28:35 PM
I has not any phone any longer. I have only modem connected to the socket by 10 meters longRJ11 ended CW1308 0.5 mm diameter. Is it something which can be removed if I have not phone? In the past, I've had bridge tap from the socket to the socket in other room.

I don't know anything about the regulations regarding the connection of devices to telephone lines in Poland. All I can say is that without a telephone connected to the line, there is no need for a low-pass filter. Is the modem directly at the end of the flexible cable that is visible in images 4 & 5? (CW1308 specification cable is not flexible, it consists of solid core conductors and should be securely fitted to (say) a wall surface. The cable between the circular box and the rectangular box looks very much like CW1308 specification cable.)

Quote
I've also noticed errors when I turn off the lightning, when I unplug from the mains computer speakers. Phone cables are nearly the mains box (pink painted box) above the small white square connector.

I was wondering about that pink-painted circular cover, as it did remind me of the cover to a recessed junction box containing mains wiring. You have now confirmed that is the case. I would expect that somewhere there is a transformer to supply the low voltage for the bell? Unfortunately I am unable to make any other deductions regarding the bell circuit.

As you are noticing errors occurring when normal electrical items are switched off it seems like the telephone circuit has been installed too close to some mains wiring. The only real solution would be to have the telephone circuit's service feed cable moved, so that it does not share the same route as the door bell wiring and is clear of any other mains wiring.

Quote
What is spark-supressing capacitor?

To suppress the sparking that might occur when the door-bell push button is released, one could fit an appropriately sized capacitor and resistor, in series, across the contacts. See the attached circuit diagram. The capacitor is the component shown by the two parallel plates and the resistor is the component shown by the zig-zag line.
Title: Re: My cables
Post by: konrado5 on September 10, 2014, 12:08:28 AM
Quote from: burakkucat
Is the modem directly at the end of the flexible cable that is visible in images 4 & 5?
This cable visible in images 4 & 5 is  specification one twisted pair cable with RJ11 joints. It is on the floor. The modem is directly at the end of this cable. 
http://archiwum.allegro.pl/img/resize?img=http%3A%2F%2F10.193.20.130%2FimagesNEW%2Fac105465280842edd81ebe675b64f03fd346ad007a454482d92fb9fbed85b3d6&type=big
Quote from: burakkucat
As you are noticing errors occurring when normal electrical items are switched off it seems like the telephone circuit has been installed too close to some mains wiring
But I only see 1-2 CRC errors on fast path. I have stable connection with SNR margin 0.9 dB. What is possible effect of installing the telephone circuit close to mains wiring in addition to little errors while normal elecrtic items are switched off? Is it causes lower synchronization rate? I'm curious if I've had 0 errors on SNR margin 0.9 dB at 12 hours when I've had telephone circuit far away from mains circuit. Would shielded cable be helpful?

Quote from: burakkucat
I would expect that somewhere there is a transformer to supply the low voltage for the bell? Unfortunately I am unable to make any other deductions regarding the bell circuit.
My father says this type of doorbell doesn't need transformer. Old doorbells were connected by transformer. This doorbell is connected directly to 230V.

Quote from: burakkucat
Again, if I was being super-critical, I would suggest that the unused orange wire & white wire pair in the internal cable should not have been left with exposed conductors, twisted together.
What are the possible effects of exposed conductors?

Best regards
konrado5

Title: Re: My cables
Post by: NewtronStar on September 10, 2014, 12:14:30 AM
You can get wireless door bells which use the 433Mhz radio band, I have an 8 year old wireless doorbell and only needed to replace the batterys twice in those years of service.
Title: Re: My cables
Post by: burakkucat on September 10, 2014, 12:29:06 AM
This cable visible in images 4 & 5 is  specification one twisted pair cable with RJ11 joints. It is on the floor. The modem is directly at the end of this cable.

Ah, now I understand. It wasn't immediately obvious that the rectangular "box" is actually a socket with the aperture at the top.  :) 

Quote
But I only see 1-2 CRC errors on fast path. I have stable connection with SNR margin 0.9 dB. What is possible effect of installing the telephone circuit close to mains wiring in addition to little errors while normal elecrtic items are switched off? Is it causes lower synchronization rate? I'm curious if I've had 0 errors on SNR margin 0.9 dB at 12 hours when I've had telephone circuit far away from mains circuit. Would shielded cable be helpful?

It is just bad practice and can be a source of errors to the broadband circuit.

Quote
My father says this type of doorbell doesn't need transformer. Old doorbells were connected by transformer. This doorbell is connected directly to 230V.

In that case, if I am correct that the incoming service-feed (the black cable that enters via the centre of the back of the circular box) is sharing the same hole (and route) as mains wiring to the bell-push, then it is extremely bad practice. In the U.K. it would be classified as dangerous and would not be permitted.

Quote
What are the possible effects of exposed conductors?

Another loose and bare wire might make contact. perhaps? It just is not "good practice" and looks as if whoever performed the installation could not be bothered (to do it right).

Please remember, I do not have any knowledge of the standards or the way such wiring should be connected in Poland.  :no:  I can only explain what I see with due consideration of the standards and practices used in the UK.
Title: Re: My cables
Post by: konrado5 on September 10, 2014, 12:37:47 AM
Quote from: burakkucat
It is just bad practice and can be a source of errors to the broadband circuit.
Is shielded cable bad practice? I suspect shielding of the phone cable would diminish interferences from the mains wiring.  Does installing the telephone circuit close to mains wiring lower synchronization rate? Is it causes other errors than these errors which I notice while the normal electrical items are switched off? Moreover, I hear crak in AM radio at about 612 kHz when I switch off the electrical items even if radio is far away from the mains wires.

When I was tooking off the cover of the phone socket, I've noticed my upstream SNR margin was jumping about 1 dB every second. Is it something wrong with the internal CW1307 joint of the cable in this socket?
http://postimg.org/image/lyee5osvf/

Phone socket with cover:
(http://s25.postimg.org/vydvzbtij/image.jpg) (http://postimg.org/image/vydvzbtij/)

What is probable the diameter conductor of incoming service-feed and internal CW1307 cable?

Best regards
konrado5
Title: Re: My cables
Post by: JGO on September 10, 2014, 09:20:58 AM
Konrado - I suspect it is "What is possible effect of installing the telephone circuit close to mains wiring in addition to little errors while normal electric items are switched off?  " NOT   "screened cable" that  burakkucat is talking about as bad practice.  There probably still is a IEE prohibition on any other wires in the same conduit as 50 Hz wires for safety. 

In your case it can conduct interference near the modem as well.  A filter in the modem mains leads will be more effective if the interference  from other nearby 50Hz wires can't bypass it.   Even if they are not in use for 50Hz, higher frequencies can ignore switches !
   

 
Title: Re: My cables
Post by: burakkucat on September 10, 2014, 02:26:16 PM
What is probable the diameter conductor of incoming service-feed and internal CW1307 cable?

CW1308 specification cable is typically between the two limits of 0.5 mm and 0.9 mm. The incoming service-feed cable would also be similar.

Quote
When I was tooking off the cover of the phone socket, I've noticed my upstream SNR margin was jumping about 1 dB every second. Is it something wrong with the internal CW1307 joint of the cable in this socket?

It would suggest that there possibly could be a loose connection or the RJ11 plug is not making perfect contact with the connectors in the socket. To eliminate a loose connection, just remake the two joints! Blue wire under the screw terminal from which the green wire link wire to the socket originates and white wire under the screw terminal from which the red link wire originates.

Quote
Is shielded cable bad practice? I suspect shielding of the phone cable would diminish interferences from the mains wiring.

I have no practical experience of using a shielded twisted pair (STP) cable in a telephony installation.

Quote
Does installing the telephone circuit close to mains wiring lower synchronization rate?

I would imagine it entirely depends upon the transceiver circuitry. In other words, how well the induced 50 Hz common-mode "hum" is rejected by the differential input stage of the receiver.

Quote
Is it causes other errors than these errors which I notice while the normal electrical items are switched off? Moreover, I hear crak in AM radio at about 612 kHz when I switch off the electrical items even if radio is far away from the mains wires.

Every electrical item will, when switched off from its power source, generate a small spark across the opening switch contacts. That is why switches are constructed to suddenly "snap" open, thus minimising the time that the opening contacts are still close enough together and allowing a spark to occur. There is nothing particularly special about the 612 kHz frequency. I usually check at 300 kHz, 303 kHz, 306 kHz and 612 kHz. There is electrical "noise" everywhere. The best that anyone can do is to minimise the effect of electrical noise. For an xDSL circuit, using good quality twisted pair cable and ensuring that both legs of the pair have good AC balance will be the first step.
Title: Re: My cables
Post by: konrado5 on September 10, 2014, 02:42:08 PM
Quote from: burakkucat
CW1308 specification cable is typically between the two limits of 0.5 mm and 0.9 mm. The incoming service-feed cable would also be similar.
From the socket to the modem I have 0.5 mm. I hope, CW1308 internal cable and external cable also have 0.5 mm. If I know in Poland typical twisted pair has 0.5 mm.
Quote from: burakkucat
I would imagine it entirely depends upon the transceiver circuitry. In other words, how well the induced 50 Hz common-mode "hum" is rejected by the differential input stage of the receiver.
Is mains wire noise noticable in my QLN? This is both upstream and downstream graph from DELT mode.
Quote from: burakkucat
e best that anyone can do is to minimise the effect of electrical noise. For an xDSL circuit, using good quality twisted pair cable and ensuring that both legs of the pair have good AC balance will be the first step.
Has internal CW1308 cable good AC balance? Is it possible to check it on my photos? What is AC balance?

Best regards
konrado5
Title: Re: My cables
Post by: Bodge99 on September 10, 2014, 03:04:24 PM
Hello,

Terminating solid core wire onto a screw terminal: As taught by PO telephones (this shows my age!).

The stripped conductor should be wrapped around the screw (pan head screw with washer assumed) in a clockwise direction. The "tail" should not cross over the conductor. This can otherwise cause conductor failure due to excessive localised compression of the wire.

The screw should then be tightened with the tail brought up vertically. The tail is then cut off flush with the screw head.

See the image for the BT 52a on http://www.telephonesuk.co.uk/connection_boxes.htm

This one is **nearly** correct, but gives the general idea. Actually I would say that the green and white wires are O.K.. No comment required on the lack of cable sheath inside the junction box!

Bodge99.
Title: Re: My cables
Post by: burakkucat on September 10, 2014, 03:11:33 PM
From the socket to the modem I have 0.5 mm. I hope, CW1308 internal cable and external cable also have 0.5 mm. If I know in Poland typical twisted pair has 0.5 mm.

From the pictures, I am fairly confident that your wiring is of appropriate diameter.

Quote
Is mains wire noise noticable in my QLN? This is both upstream and downstream graph from DELT mode.

I'll attempt to answer that by first asking you a question.  ;)  Would you expect to see it? One "tone" of an xDSL service is 4.3125 kHz wide. The mains frequency is 50 Hz. Think about the scale of those frequencies, it is 86.25 to 1 . . . (Would I expect to see anything? No.  :no:  )

Quote
Has internal CW1308 cable good AC balance?

Yes, of course. It is a stipulation of that specification.

Quote
Is it possible to check it on my photos?

No.

Quote
What is AC balance?

Just as one would expect to see the same DC resistance across the same length of the two legs of a cable pair, so one would expect to see similar AC resistance -- impedance -- for the two legs. (Along with other properties.) Good AC balance for two legs of a cable pair is just a way of ensuring that those two wires behave similarly. Remember an xDSL circuit operates in differential mode, so it is the difference in the signal carried on the two wires that is important. Having two mismatched wires (poor AC balance, non-similar DC resistance, etc) will result in the wanted signal being distorted at the receiver.
Title: Re: My cables
Post by: konrado5 on September 10, 2014, 03:41:09 PM
Quote from: burakkucat
I would imagine it entirely depends upon the transceiver circuitry. In other words, how well the induced 50 Hz common-mode "hum" is rejected by the differential input stage of the receiver.
Quote from: burakkucat
Would you expect to see it? One "tone" of an xDSL service is 4.3125 kHz wide. The mains frequency is 50 Hz. Think about the scale of those frequencies, it is 86.25 to 1 . . . (Would I expect to see anything? No.
1. Why would you expect to 50 Hz common-mode "hum" affect the ADSL service? What way, if it would be not noticable in QLN? Why does switching off can be heared in the AM radio on the significantly higher frequiences if mains frequency is 50 Hz.
Quote from: burakkucat
Just as one would expect to see the same DC resistance across the same length of the two legs of a cable pair, so one would expect to see similar AC resistance -- impedance -- for the two legs. (Along with other properties.) Good AC balance for two legs of a cable pair is just a way of ensuring that those two wires behave similarly. Remember an xDSL circuit operates in differential mode, so it is the difference in the signal carried on the two wires that is important. Having two mismatched wires (poor AC balance, non-similar DC resistance, etc) will result in the wanted signal being distorted at the receiver.
2. How can I check AC balance? Is it noticable in Hlog, QLN, SNR margin graph?

Best regards
konrado5
Title: Re: My cables
Post by: burakkucat on September 10, 2014, 04:41:38 PM
Quote
2. How can I check AC balance? Is it noticable in Hlog, QLN, SNR margin graph?

No, the graphs will not provide any information. Hand-held testers, such as a JDSU HST-3000c or an Exfo AXS-200/635, will have the means to qualify a circuit. They have an option to perform a Pair Quality Test (PQT).

Quote
1. Why would you expect to 50 Hz common-mode "hum" affect the ADSL service? What way, if it would be not noticable in QLN?

An induced longitudinal signal (common mode signal) could bias the differential input circuit and distort its operation, to the detriment of the wanted differential signal, for example. A well designed and optimality constructed differential circuit should be capable of rejecting a longitudinal (common mode) signal. But no circuit will ever be perfect. So it would be sensible to ensure that any influence of an unwanted signal is minimised.

Quote
Why does switching off can be heared in the AM radio on the significantly higher frequiences if mains frequency is 50 Hz.

It is due to the spark that is generated when the current being drawn by the load is interrupted by the act of switching off. The original radio transmitters were spark devices -- they radiated significant energy across the entire radio spectrum. There were true broadband signal generators, without any means of real tuning the signal to a specific frequency. (Perhaps you will find it interesting to read about Marconi's original radio experiments and his very high voltage spark transmitters.)
Title: Re: My cables
Post by: konrado5 on September 10, 2014, 05:13:49 PM
Quote from: burakkucat
No, the graphs will not provide any information. Hand-held testers, such as a JDSU HST-3000c or an Exfo AXS-200/635, will have the means to qualify a circuit. They have an option to perform a Pair Quality Test (PQT).
1. How is it possible the graph will not provide any information if bad balance affect quality of servie? Is it not raise attenuation? Is it not raise noise level? SNR on every distinct tone is affected only on power output of this tone, attenuation of this tone and noise of this tone.

Quote from: burakkucat
n induced longitudinal signal (common mode signal) could bias the differential input circuit and distort its operation, to the detriment of the wanted differential signal, for example.
2. What would be accurate affect? Higher QLN, higher attenuation, signal reflections?

3. Does spark cause little (1-2) CRC errors on fast path if someone has phone lines far away from mains circuit? I suspect it also cause CRC error because it is noticable in AM radio regardless where I put radio.

Quote from: burakkucat
Quote from: konrado5
Has internal CW1308 cable good AC balance?
Yes, of course. It is a stipulation of that specification.
4. Can CW1308 damage cause bad AC balance? Can loose joint cause bad AC balance?

5. Why does spark does not cause SNR lowering? It only causes errors.

Best regards
konrado5
Title: Re: My cables
Post by: konrado5 on September 11, 2014, 02:20:16 PM
When I do spark I has following BER at SNR margin 1.1 dB (1 CRC error).

Test Time (sec):       20
Total Transferred Bits:       319426560
Total Error Bits:       744
Error Ratio:       2.33e-06
Title: Re: My cables
Post by: boost on September 11, 2014, 04:36:27 PM
konrad... get a JDSU~ please :D
Title: Re: My cables
Post by: konrado5 on September 11, 2014, 04:57:43 PM
Unfortunately JDSU is very and very expensive. I'd like to test  my line accurately. Perhaps modem firmware modify would be helpful.

Best regards
konrado5
Title: Re: My cables
Post by: burakkucat on September 11, 2014, 06:01:44 PM
Just a quick note from The Cattery to say that I have posted everything, of which I can think, that relates to the wiring shown in the pictures.

If that wiring was in my home and I was not happy with my broadband service performance, I would rewire the circuit from the circular connection box to a convenient location for the socket. I would use a new length of the appropriate specification cable (CW1308 or better) and would probably fit a new socket, rather than reusing the existing one.
Title: Re: My cables
Post by: konrado5 on September 12, 2014, 06:51:46 PM
Thank you burakkucat. It is good idea. It seems the external black cable is not as nearly the mains wiring as the internal CW1308 cable. Admittedly, the external black cable is as high as mains wiring however it is at the other side, deeper in the wall.

What type of cable should be used to connect phone socket with modem? I have CW1308 with RJ11 ends but some people say this cable shouldn't be used with RJ11 ends.

Best regards
konrado5
Title: Re: My cables
Post by: kitz on September 12, 2014, 08:19:20 PM
Quote
What type of cable should be used to connect phone socket with modem? I have CW1308 with RJ11 ends but some people say this cable shouldn't be used with RJ11 ends.

You can use CW1308 with your RJ11's.   
CW1308 is better quality than CW1311/Flat cable or non-twisted which normally used with RJ11, if its crimped properly then its fine.  You could even use CAT5E..  as long as the cable quality is better then it shouldnt matter. 
Title: Re: My cables
Post by: konrado5 on September 12, 2014, 08:31:28 PM
Quote from: kitz
You could even use CAT5E.
I've heard CAT5e could give impedance mismatch if other cables are not CAT5E.

Best regards
konrado5
Title: Re: My cables
Post by: HPsauce on September 12, 2014, 08:51:17 PM
Conversely, I use CW1308 with RJ11 ends as LAN cable around my house.
Works fine for 100mbps (if you use the appropriate 4 pins) and is less visible. That speed is fine as it's for internet connections.
Title: Re: My cables
Post by: konrado5 on September 14, 2014, 05:30:20 PM
I'd like to swap the internal CW1308 cable pair: white-blue to white-orange. I see white-blue is nearer to the wall and nearer to the doorbell cables. How to do it? Will I need new joints?

Best regards
konrado5
Title: Re: My cables
Post by: boost on September 15, 2014, 06:11:31 PM
Have you considered removing the power source from the doorbell and checking what your QLN looks like after a resynch?

If there was even the smallest positive change I would be binning it and removing the bell from the front door and replacing with a knocker.

Also, have you considered reterminating your external cable, temporarily perhaps, by using a short run of any cable you like directly to an RJ11 jack into your modem? You could literally drape it down that wall as a test? Check HLOG on resynch?

My paint skills are dire but I've tried to visualise what I mean below:

(https://dl.dropboxusercontent.com/u/20271204/dslstats/k5a.jpg)

In one fell swoop, you have built the case to palm this off to your ISP to investigate your dodgy pavement? Or... the new graphs present spectacularly smooth lines? :)
Title: Re: My cables
Post by: konrado5 on September 15, 2014, 06:28:15 PM
Quote from: boost
Have you considered removing the power source from the doorbell and checking what your QLN looks like after a resynch?
Unfortunately doorbell cable is not only mains cable nearly the phone cable. The round pink-painted box is mains box.
Quote from: boost
Also, have you considered reterminating your external cable, temporarily perhaps, by using a short run of any cable you like directly to an RJ11 jack into your modem? You could literally drape it down that wall as a test? Check HLOG on resynch?
Would I do it by cable with RJ11 on one leg?

Best regards
konrado5
Title: Re: My cables
Post by: NewtronStar on September 15, 2014, 09:11:32 PM
There are flight tickets at the departure desk gate 6 Heathrow in the name of Walters Wheelbarrow destination Poznan International ETD 16/9/2014 2:00am  ETA 16/9/2014 3:45 am  :D
Title: Re: My cables
Post by: boost on September 15, 2014, 11:35:33 PM
Incremental steps... just start the process :)

Pretend I drew two wires! Told you my paint skills were bad :D
Title: Re: My cables
Post by: konrado5 on October 13, 2014, 04:41:19 PM
I think about digging out of external black cable,cutting down about 20 meters and putting the phone socket in the other room. How to cut down the cable so as to conductors have the same lenght? Is junction box necessary? I think about connecting the black external cable directly to phone socket.

Best regards
konrado5 
Title: Re: My cables
Post by: burakkucat on October 13, 2014, 04:52:33 PM
Obviously we do not know if you should do it (only having experience of the network wiring within the UK) but if you are capable, then re-routing the incoming service pair away from the bell, its circuit and any mains wiring may prove to be beneficial.

Again (only based on experience with wiring in the UK) it would not be necessary to join the incoming service cable to an internal (CW1308 type specification) cable, if the terminating telephone socket is mounted on the wall through which the service cable will enter.
Title: Re: My cables
Post by: konrado5 on October 13, 2014, 04:53:57 PM
But how to cut down the cable so as to not change the diameter size? I'm afraid if I would cut down the some part of cable, the diameters in the cable would have sligthly different sizes. I think I would have extremally small BER if I would have the cables away from the mains wiring.
Title: Re: My cables
Post by: burakkucat on October 13, 2014, 05:15:39 PM
I would use standard wire "side cutters", such as these (https://www.google.co.uk/search?q=side+cutters&num=20&tbm=isch&tbo=u&source=univ&sa=X&ei=Tfo7VPHbBIrd7QbZ4QE&ved=0CEEQsAQ&biw=1280&bih=628) -- nothing special.
Title: Re: My cables
Post by: boost on October 13, 2014, 06:00:10 PM
Konrad, excellent idea.

I look forward to seeing your new HLOG and QLN when you've moved it away from the mains and any radiator pipes.


I am pretty bad at this kinda stuff so I try to get plenty of help and the right tools for the job.
Good luck.
Title: Re: My cables
Post by: konrado5 on October 13, 2014, 08:15:09 PM
"side cutters"="pliers"?
Title: Re: My cables
Post by: burakkucat on October 13, 2014, 09:21:38 PM
"side cutters"="pliers"?

No. Similar in design to pliers but with a pair of cutting edges rather than a pair of gripping surfaces.
Title: Re: My cables
Post by: konrado5 on October 20, 2014, 09:55:59 PM
I did it and I have slightly other Hlog and QLN. hlogdiagmode5.png is old Hlog. hlogdiagmode6.png is new Hlog. qlndiagmode5.png is old QLN. qlndiagmode6.png is new QLN. I have slightly lower attenuation and slightly higher attenuation at tones 187-289 and end tones. I have slightly higher QLN at the beginning of download. My synchronization rate is about 200 kbps lower. Unfortunately, I have still CRC errors when I turn off the light. It seems the impulse noise from mains comes to the router by router's power supply. Now I don't have phone installation nearly the mains wiring.

Best regards
konrado5
Title: Re: My cables
Post by: konrado5 on October 20, 2014, 10:00:55 PM
Hlog with one conductor connected - blue conductor - hlog39db.png
Title: Re: My cables
Post by: konrado5 on October 20, 2014, 10:51:21 PM
I forgot attach qlndiagmode5.png
Title: Re: My cables
Post by: konrado5 on October 21, 2014, 08:07:51 PM
The worst is 200 kbps lower downstream synchronization rate after cutting down the 20-30 meters of cable caused by higher attenuation at some frequiences. It is curious.
Title: Re: My cables
Post by: konrado5 on October 22, 2014, 08:21:42 PM
Could double RJ11 socket cause?  At the old installation I had single RJ11 socket. I suspect unused RJ11 connector can be little bridge tap (about 2 cm :D).
http://static.tme.eu/katalog_pics/9/b/a/9baa892871bfb621f7634b4e32b8838c/rj11gn-mx2.jpg
Title: Re: My cables
Post by: boost on October 22, 2014, 09:59:09 PM
I'm gonna say yes, as daft as I feel saying it :P

Shorter the bridge tap, the worse it is too, afaik?

A short while ago, I cut about 10 foot of cable off my old man's modem and installed a shiny faceplate. My reward? 3x increase in round trip times (11ms -> 29ms).
In doing so, I'd elevated the modem about 2 metres from it's original position on the floor and it was presumably picking up lots more RFI in the new position.
Title: Re: My cables
Post by: konrado5 on October 23, 2014, 02:53:09 PM
My circuit is contradictory with laws of nature. :) It is normal that attenuation is lower when circuit is longer. Shortening of my line about 20-30 meters and lowering amounts of joints caused that my attenuation (hlogdiagmode6.png ) around tone 200 and at the end of graph is higher (not lower). It is not possible that attenuation on some frequency is lower after the distance. It should be raised, not lowered. On my circuit after running of 30 meters my attenuation on some frequiences was lower. :)Unfortunately I see significant lowering of attenuation around 1.8-2.0 Mhz gap. My router now often enables singular tones of gap at SNR margin 0.9 dB eg. 430. It was not happening before shortening the circuit.
Title: Re: My cables
Post by: JGO on October 23, 2014, 04:44:05 PM
My circuit is contradictory with laws of nature. :) It is normal that attenuation is lower when circuit is longer.

NO !   This is what Power Engineers think because they don't deal with long lengths of cable (long in wavelengths;  at 50 Hz the wavelength is c 6,000 km)  so they get away with ignoring the inductance and capacitance of the cable.   Doing this at radio frequencies, including xDSL, gives wrong answers. Your results on attenuation seem plausible if you don't have one single cable of the same type all the way.
 I suggest you read a book on Transmission Lines for the full theory.
Title: Re: My cables
Post by: c6em on October 23, 2014, 06:04:22 PM
You'll be glad that this power engineer was taught extensive transmission line theory while at university.

Title: Re: My cables
Post by: konrado5 on October 23, 2014, 07:18:01 PM
I don't understand. Is it normal that some frequiences are more attenuated and some frequiences less attenuated if the line is longer? For example line is 1200 meters long. Attenuation at 900 kHz is 32 dB and attenuation at 1696 kHz is 40 dB. If line has 1230 meters, attenuation at 900 kHz is 30 dB and attenuation at 1696 kHz is 38 dB. Is it possible that 900 kHz signal travels 1200 meters and is 32 dB weaker than output signal and this signal is stronger (only 30 dB weaker) if it travels additional 30 meters? Furthermore, in the case of bridge tap, zeroing frequency (notch frequency) is depended on the lenght of bridge tap and not distance of socket to the bridge tap. I suspect that similar case is in the case of reflections caused by wire gauge changes. It should be not easy to change the reflected frequiences.

Best regards
konrado5
Title: Re: My cables
Post by: konrado5 on October 26, 2014, 08:58:45 PM
I took photo of my new socket.
(http://s25.postimg.org/z0ye9zq5n/new_socket.jpg) (http://postimg.org/image/z0ye9zq5n/)

Is second unused RJ11 contact bridge tap? Blue conductor is connected do two green conductors. White conductor is connected to two red conductors.

Now I have no circular box as in the old installation. I have external black cable connected directly to the rectangular phone socket with RJ11 connector. It implicates that at this moment external black cable is not connected by gel-crimps. Is it right?
(http://s25.postimg.org/4hrqqsnqj/puszka.jpg) (http://postimg.org/image/4hrqqsnqj/)

Best regards
konrado5
Title: Re: My cables
Post by: konrado5 on November 01, 2014, 02:54:49 PM
I changed my two connectors phone socket on one connector phone socket. I don't see difference.

Best regards
konrado5
Title: Re: My cables
Post by: burakkucat on November 01, 2014, 11:05:58 PM
I changed my two connectors phone socket on one connector phone socket. I don't see difference.

From that result you should therefore be confident that the circuit is as good as it can be. (I would have been very surprised if you had noticed any significant difference.)

JGO's comment about transmission line theory (for radio frequencies up to ~ 2MHz) and considering the capacitance, inductance, etc, as well as the DC resistance of your circuit, may help to explain some of the observations that you have made.
Title: Re: My cables
Post by: konrado5 on November 01, 2014, 11:44:33 PM
Thank you very much for your reply. How to diminish CRC errors when doorbell rings, lights are turned off, electric saw is used ? It seems that impulse noise from mains gets to the modem by modem power adapter.
Quote from: burakkucat
JGO's comment about transmission line theory (for radio frequencies up to ~ 2MHz) and considering the capacitance, inductance, etc, as well as the DC resistance of your circuit, may help to explain some of the observations that you have made.
Unfortunately I am ignorant of physics. Could you explain it clearly? Is it normal that shortening the circuit causes that some frequiences are less attenuated but other frequiences are more attenuated? It seems to me that on every frequence attenuation should be lower than before shortening unless something was damaged.

Is it normal that QLN on my starting downstream tones is higher after shortening circuit?

Best regards
konrado5
Title: Re: My cables
Post by: NewtronStar on November 02, 2014, 10:53:58 PM
Thank you very much for your reply. How to diminish CRC errors when doorbell rings, lights are turned off, electric saw is used ? It seems that impulse noise from mains gets to the modem by modem power adapter.

I have done quite a lot of stuff around the house to lessen the effects of REIN but it's still get's through is it the internal broadband cable is it my mains wiring picking up REIN or the drop wire  :-\

You can only do so much to limit the effects of REIN and hang up your tools and feel proud with yourself for trying your best  ;)
Title: Re: My cables
Post by: konrado5 on November 03, 2014, 04:05:50 PM
Fortunately I have very little impulse noise (very short), only when my lights are turned off, doorbell rings. I'm curious what number of CRC errors I have if I would have no CRC errors from mains. Perhaps power strip with some kind of filter would be helpful.

Best regards
konrado5
Title: Re: My cables
Post by: burakkucat on November 03, 2014, 05:49:47 PM
There is an experiment that you could perform to see if interference is induced into and carried by the mains power supply to your modem/router . . . That is to borrow a 12V car battery or power tool battery pack and run the modem/router from the battery for a period of time.
Title: Re: My cables
Post by: konrado5 on November 03, 2014, 06:11:11 PM
1. Thank you for suggestion. I've seen thread about car battery in this forum. They write that car battery causes other type of interferences:
http://forum.kitz.co.uk/index.php?topic=7448.0

2. What do you think about issue in reply #51 ?

Best regards
konrado5
Title: Re: My cables
Post by: burakkucat on November 03, 2014, 06:30:19 PM
Sorry but I shall have to be brief --

1. A slightly discharged car battery (for example) will be close enough to 12V (if that is what your modem/router requires, of course). As I also mentioned, the battery pack from a power tool could be used. (I'm not sure who (without checking) but other kitizens have successfully experimented that way.)

2. Transmission line theory is something I know very little about. At best one needs to consider the (transmission) source, the terminating load (of the receiver), the resistance, the impedance, the capacitance and the inductance of all segments that make up the circuit. Undoubtedly complex numbers, differential equations and integrals will also be part of the theoretical mix. I tend to ignore the subject in detail and just regard it as a black-box effect. It is only when someone (like JGO) makes a mention of the subject do I realise that my simple DC circuit theory may not be giving the full picture. The (English) wikipedia page (http://en.wikipedia.org/wiki/Transmission_line) would be a good starting point for any understanding.
Title: Re: My cables
Post by: konrado5 on November 03, 2014, 07:39:36 PM
Is it surprise for you that I have slightly other Hlog after shortening the circuit. Hlog before shortening is hlogdiagmode5.png. Hlog after shortening is hlogdiagmode6.png. Do you see that some frequiences are more attenuated than before (around tone 200 and end of the graph)? Is it surprise for you that on my new QLN there is more noise at the start of downstream?
Title: Re: My cables
Post by: konrado5 on November 03, 2014, 07:40:18 PM
QLN
Title: Re: My cables
Post by: NewtronStar on November 03, 2014, 10:26:53 PM
Sorry but I shall have to be brief --

1. A slightly discharged car battery (for example) will be close enough to 12V (if that is what your modem/router requires, of course). As I also mentioned, the battery pack from a power tool could be used. (I'm not sure who (without checking) but other kitizens have successfully experimented that way.)

Its something I would be interested in doing, this is something I did in my early teens running a CB radio from a car battery and if it needed charged hook up the car battery charger and you can always use a multimeter to check the volts before the experiment  ;)
Title: Re: My cables
Post by: konrado5 on November 05, 2014, 09:22:28 PM
What about reply #57 ?
Title: Re: My cables
Post by: konrado5 on November 07, 2014, 01:11:41 AM
Now I have no circular box as in the old installation. I have external black cable connected directly to the rectangular phone socket with RJ11 connector. It implicates that at this moment external black cable is not connected by gel-crimps. It implicates also one less section of cable. Is it right?
(http://s25.postimg.org/4hrqqsnqj/puszka.jpg) (http://postimg.org/image/4hrqqsnqj/)

Best regards
konrado5