I’m having trouble with xfinity, aka Comcast. I lost internet last night and my modem reported good downstream numbers, but not so good upstream numbers. Upstream is the signal being sent from my modem to Comcast’s equipment. Comcast’s equipment control how much signal my cable modem puts out, but last night it was right at the edge of the maximum my modem is capable of providing.

I’m using a four way splitter and understand that each of the four taps is getting 1/4 of the incoming signal.

What happens with the outgoing/upstream signal from my cable modem? Am I correct in thinking the signal is not getting split and the only difference between the signal strength before being fed into the splitters and after the splitters is the sum of the insertion losses of each splitter?

I wonder if there’s a signal amp to boost the 5-42MHz range while leaving the 400-550MHz range alone.

What speeds are you getting???

can you try it straight through without the splitter as a temporary test? no splitter.

Depending on which splitter you have, they can have little transformers in them that tend to attenuate lower frequencies worse than the freq range they were designed for, so that might be a factor

All day yesterday until about 7pm I had no connection. My cable modem reported good downstream connections on 24 channels between 400MHz and 540MHz with SNRs between 39.5 and 42.

Upstream dBmv on 6 channels from 10.4MHz to 39.6MHz was 51.8. That’s high.

My understanding is Comcast measures the signal they are receiving on their end trying to keep it between -2dB and 2dB. Their equipment tells my modem to increase or decrease power so the signal is in range on their end. The problem yesterday was their equipment asked my modem to use more power than it can put out so the signal on their end wasn’t strong enough to maintain the connection. The most likely cause is a failed connection or wire outside like the last 6 times.

Today, my modem is showing 49.5 to 50 dBmv and we aren’t having any issues. I can wire the modem to the incoming coax before the splitters and see what effect that has on. That will eliminate all the connections and splitters to the outside of the house. I’ll do that before they come out between 10 and 12 on Wednesday.

as I stated above, some are rated for lower frequencies than others. Some 5 MHz, some are 40 MHz. Do you know which you have?

I'd still try it without the splitter so see what happens. As a temporary test.

Thanks for being more specific about the frequencies ‘cause now I get what you are getting at and it’s a really good question. One, that if none of Comcast’s people haven’t caught the last five times they’ve been out, I won’t even know what to say.

They should be passing through the upstream frequencies. I’ll check in the morning when the kids are at school. It wouldn’t explain the change in signal from a month ago, but it would certainly help if it were the case.

Signal info with splitter:



Splitter:



Signal info without splitter:



Both downstream and upstream are about 5.5dBmv stronger without the splitter. My initial assumption that the splitter wouldn’t affect the upstream signal is false.

With two TVs and the modem, removing the splitter permanently is not an option. I think I need an amplified splitter with an active return like the 5 port one here: Link.

looking at the specs for the splitter that you linked, I don't think it's going to work either. The bi-directional "voice" port that goes down to 5 MHz still has loss similar to what you have.

I was wondering about that port’s purpose. I don’t believe I have to use it for the cable modem just because it’s labeled Modem/VOIP. My understanding is it would be the only port that would still work if the unit lost power. The theory being the phone would still work as the cable modems with phone have battery backup built in.

I pushed the Comcast appointment to Monday and Amazon just dropped off an amplified splitter, so I should know in the next hour or so if an amplified splitter with active return will help.

Signal info from today before the new amplified splitter:



And after the new amplified splitter:



All the dBmv readings are about 5.5dBmv better with the amplified splitter than without, so it’s as if the modem was directly connected to the incoming feed instead of the previous splitter.

Looks like it does what it says it will do. The proof will be if I lose the connection or not from this point forward. Those number in the 50s from before the new splitter would only be getting worse with summer coming. The numbers from the earlier post were from a cooler cloudy day.

I bought a Commscope CSMF1APDU9VPI.