Jump to content
videobruce

Can low signal to modem affect speed tests?

Recommended Posts

Can a lower signal level where the return path has a higher than normal transmit level cause a problem with download throughput as one so called 'tech' is claiming?

 

Running a broadband bi-directional amp, two taps & one splitter to the UBEE eMTA device (I hate called them modems), I have been able to get up to 15-18 Mbps d/l peak transfer on occasions, but that is rare, The majority of the time it only averages 4-6 Mbps on a 10 Mbps service level.

 

I can't access any internal menu in the device since TWC locks everything out which doesn't help.

Share this post


Link to post
Share on other sites

They lock you out of the modem? So you're not able to see http://192.168.100.1? If they block that, that's BS. It's YOUR modem, right?

Yes, low signal will definitely effect performance. There's actually a happy medium. Too high and it can sometimes have worse effects.

Downstream Signal to Noise Ratio (SNR) should be lower than 30 dB

Downstream power level should be between -15 and 15 dB but -8 to 8 dB is optimal

Upstream SNR anything above 29 dB is good

Upstream power should be below 55 dB - lower numbers are better.

Keep in mind that splitters drop the signal. If you have too much power, add a splitter.. dropping the levels will help in some cases. 2 way will drop 3.5 dB ... three way internally is two two way splitters. One leg drops 3.5 the other two drop 7... 4 way is three two way splitters. Thus dropping 7db on all legs. Not enough power? Amps can help but they can also hurt if not done properly. Make sure that it passes 1000+ MHz, same with splitters... If your amp is before the split right now.. try putting the splitter after the amp. And vice verse... sometimes switching that configuration can make a world of difference.

good luck!

Share this post


Link to post
Share on other sites

They lock you out of the modem? So you're not able to see http://192.168.100.1? If they block that, that's BS. It's YOUR modem, right?

 

Unlike conventional modems, I am locked out. Those figures mean little since the diagnostic screen isn't available.

It is a UBEE eMTA that I rent the TV portion (modem) for $6 a month fee  since they will not allow anyone to own one of these which makes no sense since they seem to be encouraging customers to buy a conventional modem. 

 

I like the concept doubling up the VoIP adapter and CATV modem in one though. It also does not run hot like my old DSL VoIP adapter did.

 

I ran speed tests for two hours, one way the regular configuration, the other way a straight feed (other than the three barrels replacing the amp and the four taps/splitters. I saw no real difference in download speeds running a couple of dozen tests from two different servers. Consistency varied widely for each test, but overall I can't say there was any real difference in d/l throughput.

 

Now, I have the system back up as I normally run it but placed a 10db pad in line. So far no apparent change, so I have at least a 10db leeway.

Share this post


Link to post
Share on other sites

Downstream Signal to Noise Ratio (SNR) should be lower than 30 dB

Downstream power level should be between -15 and 15 dB but -8 to 8 dB is optimal

Upstream SNR anything above 29 dB is good

Upstream power should be below 55 dB - lower numbers are better.

My modem status is below. Questions about your quotes - trying to be sure I understand what you're reeling us.

1.When you say Downstream SNR should be lower than 30 dB, I assume you mean that the noise is more than 30 dB down from the signal. So my 37-38 dB is good?

2. Downstream power says about 3 to 3.7 dBmv. That gives about 9.3 dB to about 11.5 dB. Does that sound about right?

3. My Upstream power is 45.8 dBmv, or about 33 dB above 1 mV, which is < 55 dB, so that's good?

4. For upstream SNR, I don't have a noise level so I don't know how to calculate the SNR dB.

Am I interpreting things correctly?

post-82979-0-92931400-1385174145_thumb.j

Hardware Information

System: ARRIS DOCSIS 3.0 / PacketCable 2.0 Touchstone Telephony Gateway

HW_REV: 5

VENDOR: Arris Interactive, L.L.C.

BOOTR: 1.2.1.61

SW_REV: 7.5.63.SIP.PC20.TW

MODEL: TG862G

Share this post


Link to post
Share on other sites

Here are complete details of my setup;

 

SLM; Tek/Tempo RFM151 Signal Scout (channel table edited to reflect the current assignments)

All levels using digital ch 117 (highest on my system)

Unit of measure; dbmV

Amp; CE Electronics 1GHz Bi-directional 0-18db gain set to +14db

RG6 except for a few RG59 jumpers

BT splitters and taps (taps used to equalize the level between devices where needed)

 

 

Level after the drop and before any CPE: +1 on digital ch 117 (+13 on analog ch. 2)

Level after the amp; +14db

Nine devices total

Lowest level; -3.5db

Highest level; -1.2db (eMTA)

 

Test run;

10-15 30MB download tests run as is and then again bypassing everything (straight run, only three barrels in between)

Servers in Washington and Seattle used.

 

Current test; system as is but with a 14db pad in line. This apparently is the "cliff" for the device. Anything below that causes it ti loose lock.

 

Results so far;

I can not see any real difference between the lowset signal level condition and a straight run into the eMTA. IOW's signal doesn't seem to make any difference on these speed tests.

 

Comments please. Are there any other tests I can perform?

Share this post


Link to post
Share on other sites
Guest
This topic is now closed to further replies.

×
×
  • Create New...