I was doing an uplift in an area. Node was set up as a 4x4. 1 of the 4 nodes had elevated noise floor and we had not even set foot into that part of the plant. I went to help a maintenance tech try to track it down since I was going to be working there the following week and wanted to know the cause so I could address it while replacing equipment. Was just a lose connector on a 2 way splitter feeding strip taps. 1 1/4 turn fixed it all. Avrg USSNR was low 20s and went to mid-high 30s. I later cut a new node in and 2/4 nodes had elevated noise. That time was just a faulty card at the headend feeding an EDR transmitter at the node. As someone whose job isn't fixing these problems but upgrading the plant and seeing it occur out of the blue it's always nerve racking as if it was something my crew did and often the resolution was the silliest thing. Really shows how sensitive the plant can be and yet it is still quite resilient in its own ways.
I find that most linear steps such as this one are plant-related. Irregular-shaped steps and impulse noise seem to occur more from drops - especially low-value deadend taps where there is little attenuation.
sometimes a new active will be put in and the person did not check docsis and balance the return and you end up with a raised noise floor. Or just a low upstream at the node. I have found low upstream set at an active then i balance return then all the upstreams on a leg of the active are high. Which means somebody did want to look further into a problem, such-as bad cable or passives.
Looks like overdrive to me, when the incoming DS signals are too powerfull on the amplifier first stage. Every amplifier manufacturer have a range how strong signal can hit the first stage, depending on channels numbers, type (analog, dvbc, docsis, specified by the manufacturer). If ur first stage gets overdriven cuz the signal is too strong, its makes a full spectrum noise. U can even see the first stage diplexer cutting range at the end of the reverse spectrum. If the day (and the cables) are warmer, the attenuation is bigger, so it can cause to dissapear cuz the signal strength on the input side is lower. If the time is colder, the cables are become colder too so the signal strenghts become stronger cuz more cold cable have lower attenuation (on absolute 0, every conductor is a superconductor) which can cause overdrive if the signal strenght reaches the max imput power with "x" DS channels Edit: sry for "necro" comment, but i "fight" these on Eurodocsis HFC everyday.
That’s still CPD. Levels being too high out of an amplifier can start to compress the sine waves. If an amp is amplifying too much it will cause cpd which comes from either corroded connections, bad terminators, or high levels out of an amp (I’ve also seen it come from loose connections and cracked hardline but this is less often and most likely electrical noise that looks similar or corrosion caused by the crack or possibly the center conductor keeps bending back and forth from the radial break causing it to stretch or partially break and cause reflections/echoes).
@@mrwashur1991 Yeah i know what u mean. But i meant when other things are good on the line, only the upstream SNR is affected (DS MER, -echo -ICFR are good) i have hundreds of photos of different type of ingress. I just meant, that a full spectrum ingress with that level, most of the time is caused by the amps. I only seen full range CPD caused by corrosion when there was also other problems too. I seen shorted cable ends, caused by physical and waterdamage, but those not only affected the Upstream range. With a TDR u can find those connection problems easily. Nowadays with the cheap chinese smart BS things are the ingress through lose connection become more and more common, thats true. the cheap led lights are the worst, they can create noise up to 100Mhz+
Just opened a 4th upstream on the node which runs my neighborhood. Internet has some bad common path problems from a feeder temp setup because some idiot ignored the locate flags and setup his fence post right through the dang cable. Now the we have HSD issues out the wazoo.
Those are the easy ones lol. Try finding 1 corroded fitting with 2 2 ways and 2 dc8’s between actives and the cpd isnt showing on the next active. Good luck touching the set screws lol.
Is there anything that stands out while looking into any other impairments when the CPD is not currently happening that can point us in the direction of possible causes?
Radial breaks so if your plant has a tagger there can be diode crystals where breaks are or if you know the relative direction you can replace terminators and check connection points/faceplates for corrosion.
Don’t pull power or Fwd. Don’t touch seizure screws. Only pull return pads. Track by monitoring return test point in each amp you hit or (invasive method) pull return pads. Any time you pull a return pad, watch for the elevation to drop and note the direction the cable goes in your plant. Go amp to amp until you don’t see it anymore, and then back track. Note that probing seizure assemblies will more than likely make the elevation disappear. It is important to monitor your noise as read by a spectrum analyzer in the headend (PathTrak), while you are walking up to and preparing to touch equipment. If you touch something and the elevation goes away, 9 times out of 10 the device you touched is what is causing your issue. Honestly? I just smack stuff until the elevation goes away, and when it does I cut out the last thing I smacked 😂 It’s effective 100% of the time.
@@Wav10001 When you find the line that cause the elevated noisefloor pull the last tap in line, this open line will create sort of an echochamber which causes visible cpd intervals when measuring at the amp. Then you take 150/cpd interval distance multiplied by vp (velocity of propagation) of the cable type. Usually (150/cpd interval)×0.88. This tells you the distance in "meters" to the problem going back from the open line you created. 😉
This is off the subject of the video, but we a being required to fix issues in mactrak before we can close a node health ticket. With mactrak we are finding that it is issues with individual customers, and there is nothing feeding back and impacting anyone else in the node. Have you or anyone else fooled with it? Thanks
Some companies do address average issues in a node and if several inside the home issues add up it could trigger an alarm for needing repaired. Unfortunately that can be a symptom to that type of reporting.
At the Headend there is a splitter on the signal coming in. It hits a product from Viavi (formally DSAM) and allows us to remotely log in and watch the spectrum analyzer of each node.
Find s/n via your docsis tools or just the height of the haystacks minus the noise floor Max height. There is no red so I guess you mean the yellow line? That's min hold, it's useful in some circumstances when looking for constant impairments in the floor. The black line is realtime readings.
I was doing an uplift in an area. Node was set up as a 4x4. 1 of the 4 nodes had elevated noise floor and we had not even set foot into that part of the plant. I went to help a maintenance tech try to track it down since I was going to be working there the following week and wanted to know the cause so I could address it while replacing equipment. Was just a lose connector on a 2 way splitter feeding strip taps. 1 1/4 turn fixed it all. Avrg USSNR was low 20s and went to mid-high 30s. I later cut a new node in and 2/4 nodes had elevated noise. That time was just a faulty card at the headend feeding an EDR transmitter at the node. As someone whose job isn't fixing these problems but upgrading the plant and seeing it occur out of the blue it's always nerve racking as if it was something my crew did and often the resolution was the silliest thing. Really shows how sensitive the plant can be and yet it is still quite resilient in its own ways.
I find that most linear steps such as this one are plant-related. Irregular-shaped steps and impulse noise seem to occur more from drops - especially low-value deadend taps where there is little attenuation.
sometimes a new active will be put in and the person did not check docsis and balance the return and you end up with a raised noise floor. Or just a low upstream at the node. I have found low upstream set at an active then i balance return then all the upstreams on a leg of the active are high. Which means somebody did want to look further into a problem, such-as bad cable or passives.
Good shit. That’s real. If you touch anything, it goes away and you think its cool. Two days later, the same shiiiit.
Looks like overdrive to me, when the incoming DS signals are too powerfull on the amplifier first stage. Every amplifier manufacturer have a range how strong signal can hit the first stage, depending on channels numbers, type (analog, dvbc, docsis, specified by the manufacturer). If ur first stage gets overdriven cuz the signal is too strong, its makes a full spectrum noise. U can even see the first stage diplexer cutting range at the end of the reverse spectrum. If the day (and the cables) are warmer, the attenuation is bigger, so it can cause to dissapear cuz the signal strength on the input side is lower. If the time is colder, the cables are become colder too so the signal strenghts become stronger cuz more cold cable have lower attenuation (on absolute 0, every conductor is a superconductor) which can cause overdrive if the signal strenght reaches the max imput power with "x" DS channels
Edit: sry for "necro" comment, but i "fight" these on Eurodocsis HFC everyday.
That’s still CPD. Levels being too high out of an amplifier can start to compress the sine waves. If an amp is amplifying too much it will cause cpd which comes from either corroded connections, bad terminators, or high levels out of an amp (I’ve also seen it come from loose connections and cracked hardline but this is less often and most likely electrical noise that looks similar or corrosion caused by the crack or possibly the center conductor keeps bending back and forth from the radial break causing it to stretch or partially break and cause reflections/echoes).
@@mrwashur1991 Yeah i know what u mean. But i meant when other things are good on the line, only the upstream SNR is affected (DS MER, -echo -ICFR are good) i have hundreds of photos of different type of ingress. I just meant, that a full spectrum ingress with that level, most of the time is caused by the amps. I only seen full range CPD caused by corrosion when there was also other problems too. I seen shorted cable ends, caused by physical and waterdamage, but those not only affected the Upstream range. With a TDR u can find those connection problems easily.
Nowadays with the cheap chinese smart BS things are the ingress through lose connection become more and more common, thats true. the cheap led lights are the worst, they can create noise up to 100Mhz+
Just opened a 4th upstream on the node which runs my neighborhood. Internet has some bad common path problems from a feeder temp setup because some idiot ignored the locate flags and setup his fence post right through the dang cable. Now the we have HSD issues out the wazoo.
hi, what do the black and the red signals mean ?
Most the time I find a bad LE causing that issue. super pain in the ass
Those are the easy ones lol. Try finding 1 corroded fitting with 2 2 ways and 2 dc8’s between actives and the cpd isnt showing on the next active. Good luck touching the set screws lol.
Is there anything that stands out while looking into any other impairments when the CPD is not currently happening that can point us in the direction of possible causes?
Radial breaks so if your plant has a tagger there can be diode crystals where breaks are or if you know the relative direction you can replace terminators and check connection points/faceplates for corrosion.
Would love some tips on how to track it since it's so touchy...CPD is my biggest challenge..
Don’t pull power or Fwd. Don’t touch seizure screws. Only pull return pads. Track by monitoring return test point in each amp you hit or (invasive method) pull return pads. Any time you pull a return pad, watch for the elevation to drop and note the direction the cable goes in your plant. Go amp to amp until you don’t see it anymore, and then back track. Note that probing seizure assemblies will more than likely make the elevation disappear. It is important to monitor your noise as read by a spectrum analyzer in the headend (PathTrak), while you are walking up to and preparing to touch equipment. If you touch something and the elevation goes away, 9 times out of 10 the device you touched is what is causing your issue.
Honestly? I just smack stuff until the elevation goes away, and when it does I cut out the last thing I smacked 😂 It’s effective 100% of the time.
Haha William did a great job of answering the question for me.
@@thebasiccablechannel7464 yes he did. I forgot to thank him
What if you don’t have amps? Where can you find which tap the offender is plugged into?
@@Wav10001 When you find the line that cause the elevated noisefloor pull the last tap in line, this open line will create sort of an echochamber which causes visible cpd intervals when measuring at the amp. Then you take 150/cpd interval distance multiplied by vp (velocity of propagation) of the cable type. Usually (150/cpd interval)×0.88. This tells you the distance in "meters" to the problem going back from the open line you created. 😉
I hate this birthday cake!
Worst birthday cake EVER. Hopefully you never have one on your birthday!
@@zlcoolboy haha!
This is off the subject of the video, but we a being required to fix issues in mactrak before we can close a node health ticket. With mactrak we are finding that it is issues with individual customers, and there is nothing feeding back and impacting anyone else in the node. Have you or anyone else fooled with it? Thanks
Some companies do address average issues in a node and if several inside the home issues add up it could trigger an alarm for needing repaired. Unfortunately that can be a symptom to that type of reporting.
What hardware are you using for this video?
Hardware? A Dell Laptop.
Samsung Note 8 for recording.
Should have been more specific. In reference to your method of viewing the RF on siad laptop that is. How are you connecting to the cable
At the Headend there is a splitter on the signal coming in. It hits a product from Viavi (formally DSAM) and allows us to remotely log in and watch the spectrum analyzer of each node.
Pathrac (software)
hi, can any one please tell me what do the black and the red signals mean ? and hiw to defuine that S/N is about 20 ?
Find s/n via your docsis tools or just the height of the haystacks minus the noise floor Max height. There is no red so I guess you mean the yellow line? That's min hold, it's useful in some circumstances when looking for constant impairments in the floor. The black line is realtime readings.
How do you fix or find it?
Viavi meter, 2 hands and a couple wrenches lol
The dreaded RNF!!👎🏽