Eth6 signal issues (EMM drop)

[Crash2009]

I find it strange that these signals are mostly so high (+), when the reason I added the amp was because some frequencies were showing -13,-14,-10, etc.

Something doesn't add up. -13 plus 8 equals -5 and all the signal strengths you posted in the code section are higher than zero. So this means that either you didn't start at -13 prior to the amp, or the amp is pumping more than 8, or there is something screwy with your Channelmasters.

[cKuR6]

Or maybe adding the mocha filter has something to do with it.

Right now, everything works, so I'm not too keen on messing with anything. Some day though, I'd like to use all 6 with no fear of pixelation though. I've been on dillaudid and muscle relaxers for two months and barely able to move. Having back surgery next week. Once I heal, I have a few more ideas to try.

1. Remove amp and run app. Maybe the filter will show itself to have been the cause of the present abnormality.
2. Swap everything but the eth6/TA to the modem side of my inbound line.
3. If amp removal puts everything back to out of spec neg readings, ill run two lines to the area. One will feed the cm6004 for moca and the other will feed the eth6/TA.

I will also check stress on the eth 6 connection as mentioned in that thread. Inforgot to mention it earlier, but banks for lining the app there. I had looked everywhere for it and your post finally led me to the right place.

[Crash2009]

I've finally obtained mocha filters, and added the splitter to the inbound TA & 6 Eth connections as requested. Thus, I added in the amp (+8) twc tried last time they were out, and put a moca filter on the cable signal line going to the TA & ETH 6.

Hopefully, this is just a typo. Why didn't they put the filter on the incoming line before the first split? Like what is recommended by Cisco in the following document.

http://www.cisco.com/c/dam/en/us/td/doc ... 1235_B.pdf

[erkotz]

I've finally obtained mocha filters, and added the splitter to the inbound TA & 6 Eth connections as requested. Thus, I added in the amp (+8) twc tried last time they were out, and put a moca filter on the cable signal line going to the TA & ETH 6.

Hopefully, this is just a typo. Why didn't they put the filter on the incoming line before the first split? Like what is recommended by Cisco in the following document.

http://www.cisco.com/c/dam/en/us/td/doc ... 1235_B.pdf

That filter installation is for a different purpose. If you are using MoCA at home, you should also have one there, but putting it directly before the InfiniTV in this case is correct.

[Crash2009]

That filter installation is for a different purpose. If you are using MoCA at home, you should also have one there, but putting it directly before the InfiniTV in this case is correct.

Assuming you only need 2, why do you need 1 on the main in and 1 in front of the ETH/TA?

[erkotz]

That filter installation is for a different purpose. If you are using MoCA at home, you should also have one there, but putting it directly before the InfiniTV in this case is correct.

Assuming you only need 2, why do you need 1 on the main in and 1 in front of the ETH/TA?

They are doing different things.

The one at the feed to your house prevents MoCA from leaking outside of your house to the Cable TV network. It also reflects the signal back into your house (improving performance).

The one at the InfiniTV blocks MoCA from getting to the InfiniTV. Many devices (not just the InfiniTV) can have trouble with interference from MoCA signals due to their high signal strength - I'm going from memory, but the MoCA beacons were really strong - something like +50dBmV.

[Crash2009]

[the MoCA beacons were really strong - something like +50dBmV.

Sounds like the +8 amp is un-necessary. MoCA is all ready "amped". As you say, MoCA needs to be knocked down or filtered for devices that only need -12 to +12.

Any idea why -13 plus +8 does not equal -5 http://www.thegreenbutton.tv/forums/vie ... 375#p82375 Does this mean there is a leak?

[erkotz]

[the MoCA beacons were really strong - something like +50dBmV.

Sounds like the +8 amp is un-necessary. MoCA is all ready "amped". As you say, MoCA needs to be knocked down or filtered for devices that only need -12 to +12.

MoCA has nothing to do with the cable TV signal (other than it uses the same wire). Amps do not normally amplify MoCA, and may actually block it.

Any idea why -13 plus +8 does not equal -5 http://www.thegreenbutton.tv/forums/vie ... 375#p82375 Does this mean there is a leak?

It doesn't - something changed during the measurement. Possibly you are comparing different QAMs. Except for losses, the math should work as you expect (0dBmV signal + 15dB amp = 15dBmV signal. 15dBmV signal run through 2-way splitter (-3.5dB) gives you +11.5 dBmV per port.

[Crash2009]

Thanks for explaining. I can't think of anything else to ask. I guess we just have to wait for ckur6 for more info.

[cKuR6]

Erkotz is correct in his explanation of the reasoning for the placement of the filters. I can verify that it strengthens the moca signal. My master bed echo was displaying a considerable amount of network issue notifications. Anytime I ran the network tuner though, it displayed full green through the HDTV range. Since adding the filters, I've not experienced a single pop up or network issue. Echo still freezes and reboots though.

I do not believe we are going to get to the bottom of the inconsistency, until and unless, I can put things back how they were and run the app. I broke the rule in troubleshooting by performing multiple changes at one time.
I installed the filters, split the signal into the TA and eth 6, and added the amp all in one swoop. I am in too much pain to have done it all in steps, barely made it through the re tightening of all the connections. It all works at the moment and that is good enough until I am fully capable again.

The only possible thing that makes sense of the math not adding up is the eth6 not being fed by the TA any more. I read somewhere that the Cisco attenuated the outbound coax to the eth6.

Finally, they (I'm assuming you meant TWC) did not add the filters, because the moca setup is mine. The techs tried to grasp my setup by they had never dealt with cable cards or moca before. They knew of cable cards but seemed completely uninformed about moca. These inconsistencies seem to be the norm for my new house. When they first installed the amp the signals were great. Mostly around -3. Within a few hours they were +30-50 on some channels. Removed the amp and all was normal again.

[Crash2009]

Pm me if it blows up again while you are recovering. I'll give you a password to the web server. There are about a 100 channels including Discovery and History. Get well soon.

[cKuR6]

Nice, thanks a lot. They moved my surgery up to Monday. Should be a 3-4 day hard recovery, then slowly back to normal.

[Crash2009]

Since Erkotz so eloquently told me I don't know "jack" about MoCA, I decided to do a little research. The MoCA video's at at Extreme Broadband explain it the same way he did. You have to register to see the training video's. My favorite, so far, is Cable 101 Passive Troubleshooting, which explains some of the causes, cures, and symptoms, relating to the items listed in the following screen cap. I still don't know jack about MoCA, but my coax cabling skills have been improved.

http://www.extreme-broadband.com/training.html

[cKuR6]

cool, maybe ill check some out during my recoup.

tonight, one of my wifes shows failed to tape on ch 1131. WMC claims there was no signal at the scheduled time of recording. there are 5 other recordings showing successfully recording. so here we have another instance of the 6th tuner failing. i'll be pleasantly surprised if my other recordings are pixelation free.

i've tuned into 1131, which should be using the same tuner that the recording would have used.... and it is playing just fine.

Code: Select all

Signal       SNR Frequency channel#;channel#;... 
8.5 dBmV 37.1 dB 93000 3;18;19;22;26;82;115;175;177;240;359;371;802 
8.5 dBmV 38.3 dB 99000 9 
6.9 dBmV 39.0 dB 111000 991;994;996;1007;1010 
6.9 dBmV 39.0 dB 117000 995;1004;1013 
[color=#FF0000]8.6 dBmV 39.9 dB 129000 1131[/color];1201;1204;1901;1902;1903 
8.6 dBmV 38.6 dB 135000 1150;1350;1904;1905;1906;1135 
8.3 dBmV 39.4 dB 171000 1301;1302;1908 
8.5 dBmV 40.4 dB 189000 1151;1203;1224;1909;1910;1911 
6.3 dBmV 38.3 dB 537000 11;21;23;24;28;31;32;40;42;113;130;176;211;301;302;350 
6.5 dBmV 39.4 dB 543000 4;10;16;20;27;36;43;54;58;59;70;106;107;135;208;230;277;323;333;351;878;886 
6.2 dBmV 39.0 dB 549000 990;1006;1008;998 
5.8 dBmV 37.9 dB 555000 602;1103;1205;1603 
5.3 dBmV 39.4 dB 585000 12;39;44;45;51;63;64;112;131;150;151;205;207;352;884;889 
6.0 dBmV 39.4 dB 591000 1202 
4.9 dBmV 39.0 dB 597000 6;35;38;48;52;101;114;152;159;219;224;225;252;317;360;876;881 
5.3 dBmV 37.6 dB 603000 104;132;137;361;603;622;642;662;877;880 
5.3 dBmV 39.4 dB 609000 2;7;8;13;15;17;29;30;37;41;46;49;103;201;202;203;204;206;250;370;887 
4.2 dBmV 37.4 dB 693000 109;136;140;157;641;812;875;882;1209;1642 
5.4 dBmV 37.9 dB 519000 102;118;305;318;677;776;811;888;901;902;903;904;905;906;907;908;909;910;911;912;913;914;915;916;917;918;919;920;921;922;923 
4.8 dBmV 38.3 dB 753000 1250 
3.6 dBmV 36.8 dB 759000 1134;1207 
4.2 dBmV 38.3 dB 765000 47;55;61;65;134;153;185;214;228;251;276;278;353;391;883 
3.2 dBmV 38.3 dB 771000 14;33;34;50;53;56;57;60;62;73;74;76;133;155;209;215;216;220;254;308;309;316;354;355 

[Crash2009]

8.7 dBmV 40.4 dB 129000 1131;1201;1204;1901;1902;1903

Your signal is too hot.

0.0 dBmV is a perfect signal.

The farther you get from 0.0, the closer you get to this.........

[Crash2009]

A while back we were speculating that there must be an amp and a bunch of splitters inside a multi tuner device. I downloaded this schematic and edited it like how I visualize the micro-circuitry. I figure "A" would be the coax in, "B" would be the amped splitter, "C" some more splitters. More speculation as to why these multi tuners are more sensitive to signal strength than a normal cable box.


I'm sure the ETH6 is more sensitive to signal strength (more splitters) than my Infini4. Here is an easy little project for you. Go into the tuner's device webpage, click on tuner 1, in the channel box enter your wife's channel (I think you said 1131 failed a recording), then click Set Channel. Now do the same thing for all 6 tuners.

Next go over to this post....http://www.thegreenbutton.tv/forums/vie ... 377#p80377 and get the code for the 6 tuner. Copy the code into a txt file then save it as a html. You will likely have to edit all the IP addresses in the file with your own address. Save it and run it, and you will end up with something like this 3rd picture. The reason I brought this up is so you can see the difference in signal strength from one tuner to the next. As you can see in mine, there is a 5 point spread between my best and worst. Whatever point spread you have might be the reason your wife's recording failed.

[cKuR6]

Alright, I've finally been mobile enough to do this. During the Holidays, there was no problem. Monday, when all shows were back on air, multiple recordings failed due to signal.

Running the app shows me all tuners, tuned to the station that failed twice Monday, show a 9.4,9.2,9.2,9.3,9.3,9.4. SNR = 39.9. Temps are 55.3-55.1C.

Surprisingly, not much variance. Across my lineup, signals now range from 8.4-11.1, too hot right? I'll try to remove the amp this evening, and check again.

[cKuR6]

whew, this is getting to be a hassle. On top of trying to get the correct wiring scheme, I've spent about 3 hrs dealing with TWC killing the cablecard at 2000 last eve, with a forced firmware upgrade......

So with the (15 db amp <- explains a little why the numbers did not match what we expected. it was 15, not 7) amp:
I have no pixelation. Frequently, at high record times, I recv the wmc blue screen, no signal error. Tuner 1 will frequently show a SL in the low teens and a very low SNR.



Without the amp, we see massive pixelation, frequent no signals, and other anomalies:




Continued next post......

[cKuR6]

No amp recording:



and theeeennnn......... here's what happened last eve, when they killed the card:





I don't know what to do at this point. I've gone through all the connections again, all are tight. Without the amp, just watching one - two tv's is sometimes a terrible failure. With the amp, we can watch live reliably, and record a few. It only bombs at prime time recording hour.

TWC is sending another tech (5th one if you count the Supervisor who came with the last guy) out to check the lines. I've asked for a cable card savvy, and mocha aware tech. I've been informed that can't be promised. So, I guess i'm just going to have to keep asking the tech when he makes his pre arrival call, if he is familiar, cancel and reschedule if not.

I believe I've at least disproved the potential splitters in the infinitv6 eth theory, as the potential signal culprit. You should see in the above post that when using all 6 tuners, the signals are steady across each. If the tech can't figure something out, then I'll try to obtain a 7db amp.

The one thing that still seems to point to the eth6 as the issue, is the fact that it only bombs when all the tuners are in use, and the signals are steady. My network is more than adequate. I have the requested 8GB of ram. i have a brand new 3TB, 7200RPM, non-green HDD.

[Crash2009]

The ETH6 is spec'd to run with a signal strength ranging from -12 to +12. While your signals are not "perfect", you are still within spec, and it should still run OK.

There is a totally different area that can cause "signal errors". I have no idea how to prove/disprove network saturation during a heavy recording schedule, but I can speculate that the network throughput specifications of the CM-6004 do not meet the system requirements of the ETH6.

CM-6004 = 270mbps ETH6 = 1000mbps

I'm not sure how many mbps each tuner requires during a recording, but if I do some simple math 1000 divided by 6 = 166. So based upon those numbers, CM-6004 would be unable to handle more than 2 recording streams. If this is correct, 332mbps of throughput is required, CM-6004 can only do 270mbps.

I would suggest taking a look around for CM-6004's Big Brother (if he has one), and install it between the HTPC and the ETH6.

I think we are flogging a dead horse.......until you upgrade.

http://www.solidsignal.com/pview.asp?p= ... r-(CM6004)

http://cetoncorp.com/products/infinitv-6-ethernet/

Here are some throughput test results:

http://www.smallnetbuilder.com/lanwan/l ... ed?start=2