bgoodbody
join:2002-05-08
North Truro, MA
 ·Comcast
| Signal Levels
Is Downstream S/N too low?
Is upstream Pwer too low
Downstream
Freq/Power: 699.000 MHz 4 dBmV
Signal to Noise Ratio: 36 dB
Modulation: QAM256
----------------------------------------------------------
Upstream
Freq/Power: 29.000 MHz 29 dBmV
Channel Type: DOCSIS 2.0 (ATDMA)
Symbol Rate: 5120 kSym/sec
Modulation: QAM32
--
- Bill |
|
jimbopalmer
Tsar of all the Rushers
join:2008-06-02
Greenwood, MS
·Windjammer Cable
1 edit | »homepage.ntlworld.com/robin.d.h.···goodlevs
Both look good
--
I tried to remain child-like, all I achieved was childish. |
|
beachintech
There's sand in my tool bag
Premium
join:2008-01-06
The Beach,US
clubs: | reply to bgoodbody
SNR is fine, Upstream TX is below spec. You will be susceptible to any noise issues on the node at that low of a level. I would have a tech come out and fix it.
--
Tech at the Beach.
I speak for myself, not my employer. |
|
Chip
Premium
join:2001-12-23
Connecticut
| said by beachintech  :SNR is fine, Upstream TX is below spec. You will be susceptible to any noise issues on the node at that low of a level. I would have a tech come out and fix it. A low upstream tx could indicate a very clean upstream path. There's no way to tell without the upstream SNR
--
The three great strategies for obscuring an issue are to introduce irrelevancies, to arouse prejudice, and to excite ridicule--Bergen Evans |
|
beachintech
There's sand in my tool bag
Premium
join:2008-01-06
The Beach,US
clubs:
·Mediacom
| said by Chip :said by beachintech :SNR is fine, Upstream TX is below spec. You will be susceptible to any noise issues on the node at that low of a level. I would have a tech come out and fix it. A low upstream tx could indicate a very clean upstream path. There's no way to tell without the upstream SNR I never said it wasn't. I said it was below where a tech would want to leave it. If any noise were to come into the system, modems running at a low TX level will be the first to see issues and lost connections.
--
Tech at the Beach.
I speak for myself, not my employer. |
|
mjr
Premium
join:2003-09-18
Bethlehem, PA
| Upstream power is regulated by the headend. The headend tells the modem how much power to put on the cable. The fact that the upstream level is lower than typical just means that there is less loss between the poster's modem and the headend. This is GOOD. I do not believe +29dBmV is out of spec. Can you reference the spec which says it is?
I don't think there's anything to fix, at least from what was provided by the OP. (The OP also didn't indicate whether he is having a problem.) |
|
beachintech
There's sand in my tool bag
Premium
join:2008-01-06
The Beach,US
clubs:
·Mediacom
| Any reference I have is field experience and working with cable every day.
Running at a low TX will make your connection susceptible to any fluctuation in noise on the line between your modem and the cmts. Keeping it above a 38+ on the return will generally keep you away from this situation and allow you a more trouble free experience. When you are down around 30, it won't take much in the way of noise to knock your modem offline. You are very close to the noise floor on the node, if you went any lower or the noise came up on the line, your modem would be drowned out / not "heard" by the headend/cmts any longer. (borrowed from another post I made)
--
Tech at the Beach.
I speak for myself, not my employer. |
|
mjr
Premium
join:2003-09-18
Bethlehem, PA
| Well, we can agree to disagree. As I stated before, the headend sets the modems transmit power. If it were too low, the headend would simply request more power, and the reverse amp would gain up, and provide it. I don't believe you understand how the reverse path works. You stated in your post above "Upstream TX is below spec" should have been qualified by "based on what my own experience of what the spec should be." There is only one spec when it comes to cable modems, and it's called DOCSIS, and it's defined by Cablelabs.
To the original poster, if you're having a problem, please let us know what the problem is. If you're just inquiring about whether the levels look ok, you have a few responses from a few different people to make your own determination. My opinion is expressed in my previous post.
Good luck. |
|
ftthdbm
@videotron.ca
| reply to bgoodbody
My first idea is the upstream power is a little bit low. Don't think he will have problems if the return patch is clean, but it's hard to tell cause we don't have the upstream signal to noise ratio. We don't know the cable plant. Maybe the plant is upgraded and very clean, after all, they use 32qam modulation.
"The easiest way to improve the carrier-to-ingress ratio is to make certain that all upstream transmitters in modems and set-top boxes are operating in the upper end of their transmit power ranges."
"Forcing transmitters into the upper end of their power ranges maximizes carrier-to-ingress ratio at the point where most ingress gets into the plant (i.e., at the subscriber drop).
»www.cedmagazine.com/Article-reve···n-1.aspx |
|
Cable Tech
join:2002-02-18
Vincennes, IN
| In my experience and in our plant 29db would be too low at the modem. I like to see at least 35db of higher. mid 40's is ideal.
Having "the lower the better" attitude isn't always better. It is way too low to the noise floor. The CMTS will of course tell the modem to range even higher output if it cant "hear" it, but that low to the noise floor, how many times does it not hear it? Several times a minute?
We call that a modem that is "flapping", which is continuously ranging for proper output, and will cause slow and intermittent connections. |
|
AskATech
join:2009-10-16
Keansburg, NJ
| reply to bgoodbody
obvioulsy dont have a system map of your area but if u have a 29 trans it can very well be a maintence issue....transmit increases the same as the forward levels attenuates through passive devices...which is to say if ur cable goes thru a 2-way splitter foward levels would drop dow 3.5 db and ur return (transmit ) would increase 3.5 db.... a good way to approximate on wut the transmit should be outside is to add 21 to the tapplate so get some binoculars look up there if u have a 14K tap that trans out of the tap should be around 35db now there are instances were it can be lower or higher but this is a good starting point...so lets say u have a 4K tap then u are right around where its supposed to be 4+21= 25 + 3.5 (2 way split)=28.5 only if thats exactly how its set up....now just because the design of the plant is supposed to be that way and everything adds up properly doesnt mean its good for the modem...what i would do is put like a 6DB pad on it...that will bring ur trans up 6 and ur levels down 6 giving you a 35 transmit and -2 recieve (thats only if u have the 4K tap in front of ur house)/Any other type of tap plate id say there is probably a line problem there giving you a low transmit right from the tap because if something were bad inside the house (ie. bad fittings) it would usually cause ur transmit to increase and not in a good way |
|
cypherstream
Looking forward to the future of things.
Premium,MVM
join:2004-12-02
Reading, PA
clubs:
3 edits | reply to bgoodbody
If the upstream is too low, wouldn't the CMTS just request the modem to Tx a little bit higher? If it's working stable, then the CMTS most likely hears the signal loud and clear. If not it should say "Hey modem, can you talk a little bit louder, I can't hear you over all this noise!"
In my experience I've seen most modems in the 40's, but at a buddies house his is in the mid 30's, but get this, he gets 7mbps upload on Comcast at 16QAM, whereas in my house with a 0 downstream /49 upstream level I get about 3.5 mbps. |
|
Enjoralas
join:2009-10-21
Northampton, PA
| said by cypherstream :If the upstream is too low, wouldn't the CMTS just request the modem to Tx a little bit higher? If it's working stable, then the CMTS most likely hears the signal loud and clear. If not it should say "Hey modem, can you talk a little bit louder, I can't hear you over all this noise!" That's not really how it works. The modem transmits at whatever level is needed to overcome the "resistance" or loss on the way back to the CMTS. In other words, the level drops as it goes, until it is zero when it hits the CMTS. No matter what you put in it's way plant-wise, the CMTS will request that the modem increase it's power level until it reaches the equilibrium state where it will "use up" all the power before it hits the CMTS. If you put too many splitters between your modem and the CMTS, it will request a level higher than your modem can transmit and you will lose connection. So in that way, the CMTS *does* set the transmit level, but it's not a set value. It's a whatever it takes value.
Noise, on the other hand, doesn't care about anything. It is what it is. And when what it is become a value too high in relation to your transmit signal level, you have problems. The CMTS can't request more signal, because it doesn't really set the request, the plant does. So if the Tx level is 29, and you'd like to see 43 db Carrier to Noise or better, noise in the -15 range is going to be a big problem. Given the Funnel Effect and cumulative nature of the return path, that is easily reached. |
|
MadMANN
Premium
join:2005-08-19
·Comcast
| reply to mjr
said by mjr :Well, we can agree to disagree. As I stated before, the headend sets the modems transmit power. If it were too low, the headend would simply request more power, and the reverse amp would gain up, and provide it. I don't believe you understand how the reverse path works. I think he has a better understanding than those who think the headend controls a reverse amp in the field. An amp does not adjust itself to accommodate one individual modem. If it is adjusted, it adjusts for the whole leg of that part of the system after it and affects every single modem in that line. And it is definitely not remote controlled by the CMTS or anything in the headend. I am sure there a lot of maintenance techs who only wish this was true.
I will give you one thing, though: I do like the concept of your idea. If it really did work the way you say it does, it would mean a lot less work for those techs who maintain the plant and a whole lot less trouble calls for those modems which would automatically adjust their upstream levels to compensate for a rising noise floor. If you could market that, it would be revolutionary. |
|
mjr
Premium
join:2003-09-18
Bethlehem, PA
| Huh? The headend does exactly control a reverse amp in the field. The Reverse amp is IN THE MODEM. Duh. Read the DOCSIS spec before you start spewing. When the CMTS determines an individual modem needs to transmit more (or less) power, it sends an appropriate command to that specific modem. The demod chip in the modem, interprets the command and changes the output power of the reverse amp (IN THE MODEM!). It's really simple actually.
Nearly every modern communication system in the world has this feedback going for it today. Including your cell phone.. Did you know that the cell tower tells your cell phone how "loud" to transmit? Did you know this occurs many multiple times per second? Did you know the whole CDMA cell network would fall apart if this weren't possible?
Do some research. It's actually not that hard to understand. |
|
mjr
Premium
join:2003-09-18
Bethlehem, PA
| reply to MadMANN
Oh. And the power level isn't adjusting just for the noise floor. It's adjusting as much for the noise floor as to play nice with the rest of the modems on the system, and to not overload (distort) the reverse path amplifiers in the line extenders and the CMTS itself. |
|
