Home Forums Networks & Cross-Platform Networking Cable Modem Failure Rate

Over the years I have used a variety of broadband services including ISDN, FiOS, and DSL, but never before have I had cable internet service. Two years ago, I gave up on DSL because the local cable plant is too antiquated to support more than about 3 Mbps and switched to the only other locally available option, cable at 30+ Mbps. I had never had an outright broadband modem failure until I switched to cable, but with cable the modems are lasting roughly one year before they fail. I installed the third cable modem this morning. All of the modems have been powered through a Tripp-Lite UPS and the external cable is fed through a surge protector as well. In the interest of full disclosure, the previous cable modems have all been Motorola. The new one is a Cisco.

My questions are

1. what is the average life expectancy of a cable modem?
2. If my experience is typical, why are cable modems so short lived compared to DSL, FiOS, and ISDN?
3. If my experience is atypical what cable modems would you recommend for long term reliability?

My cable modem is the one supplied by my ISP (Shaw) and it's been working for 11 years. It's a Motorola Surfboard Model SB5102.

FWIW, earlier this year I was looking into cable options, and had tentatively picked the Motorola Surfboard 612x DOCSIS 3.0 cable modem in case I'd switch. Part of that 'decision' was the positive user feedback. I haven't switched yet, but have found no reason since to change that view.

Same as everything else including properly sized protectors. Decades.

It is difficult to say why because you did not even list what has failed. Even details of the symptoms were not provided. So every answer will only be wild speculation.

Well you mentioned a UPS and protector as if that would do something. Some of the 'dirtiest' power comes out of a UPS in battery backup mode. A protector too close to electronics and too far from earth ground can even make electronics damage easier.

You had DSL. Which means a best protector was installed for free by the telco. Was cable connected to the same earth ground? If not,then protection is missing. And that adjacent protector may make electronics damage easier.

if all goes well, you'll need to replace your modem simply to keep up with the software at the central office, before it actually FAILS.

I was an 'early adopter' here, and one time when I put in a service call for poor performance, the guy that showed up saw my modem and chuckled, "wow we haven't put any of those out in YEARS" and he got a new one off the truck and things worked much better. The modem I was using was several "docsys" versions behind and couldn't support some of the most recent bells'n'whistles they'd recently deployed, for which I was only getting performance hits as the most obvious issue.

Symptoms of failure are as follows. All lights on the modem go out and cannot be restarted. aThere is no signal output from the modem. Trying different power bricks of the same rating had no effect. Using a different power outlet had no effect. The outlet and power supply work with other devices. The only solution found by either the cable technician or myself is replacing the dead modem.


I am not sure what you mean by ..a best protection…, but having worked for the telephone company as a central office foreman, and several years experience as a ground communications/electronics officer in the USAF in charge of among other things building and maintaining base telephone systems, my experience would lead me to question the validity of what I believe you mean by that statement.


I have no idea what ground the cable is connected to at the originating end, it is many miles from here. However in my home, the cable comes into a surge protector built into the UPS and the UPS in turn is grounded by the house wiring into a copper rod driven four or five feet into the earth. I suppose it could possibly be set deeper, but at that depth it exceeds local building code requirements and going deeper would require dynamite to blast through the 18 to 30 feet of limestone that underlies the surface. The telephone wire no longer comes into the house and is terminated on a pole about 30 feet in the air and probably 20 feet from the house. Telephone service is provided by another cable modem (also protected by a UPS in case of a power failure) attached to a coaxial cable that Ys off of the cable running to the internet modem. It has not failed.

You said you had DSL service without failure. Where was the telephone wire then? Why did you not have damage then on DSL? Because the phone line was connected to superior protection that connects to earth ground; not safety ground. Or was DSL not connected in this residence as your post otherwise implied?

Does not matter where the coax cable is grounded at the other end. Being at the other end, the ground may as well not even exist due to some basic electrical concepts. Electronic concept that must be understood to prevent damage to your 'not DSL' modem. Impedance alone says why the far end ground is irrelevant. And why the ground where it enters a building is critically important. If not properly earthed, modem damage may be expected.

Summarized as superior protection is even defined by Mil-Std 419. The solution has been standard for generations. But these are only solutions to a problem that is not yet and first must be defined.

Now, how do these cable modems fail? Do lights go on and then immediately go out? Or do lights never illuninate when powered. Which lights illuminate if any lights illuminated? Did these modems just stop working for no reason? Or when some other event occurred simultaneously? Did they stop working and then work again periodically? And what part inside failed? Facts that help identify the failure located elsewhere. If you were doing electronics in the USAF, then that last question (especially useful information) could be answered.

You have no reason to believe a deeper ground is a better ground. Superior earthing is even discussed in Mil-Std 419. Does not require 18 to 30 feet deep rods. If working in COs, then you know a wall receptacle safety ground cannot perform an earth ground function. And that protectors, routinely installed in all COs, connect to earth ground; not safety ground.

You have modem failure. If you believe it was a surge, then the surge was incoming on AC electric and outgoing destructively through one modem to earth via the coaxial cable. Or incoming on the coax. Chooses a best connection to earth via one modem. That one modem then acting as a protector for the other modem. Just some examples of why a surge might do damage.

So what changed? You did not have DSL damage. Now have cable modem damage. Did other damage occur to other appliances (even dimmer switches) in this or previous years? All examples of a search for a common factor. And why other items (ie DSL modem, dishwasher) are never damaged.

None if this should be considered a solution. Everything are examples to help identify why previous damage occurred. Solutions come later.

There are several devices in this house each connected to its own individual UPS including an iMac, Mac Mini, and a cable phone modem. None of them have had any problems. There are other devices connected to the same UPS as the cable internet modem including two Airport Time Capsules and they have not had any problems. There has been no damage to any other electrical device including the washer, dryer, dishwasher, television sets, satellite TV converter boxes, inkjet printer, laser printer, label printers, cordless telephones, light switches, etc. Other than the normal replacement of burned out light bulbs (which has almost completely ceased because I am using LED replacements) the only device(s) that are failing are cable internet modems. As to what has changed the only related change occurred about two years ago I switched from using AT&T DSL and phone to using Charter cable internet and phone.

One thing has changed that is related to the cable internet service is since the original installation, Charter has increased the connection speed from the original 8Mbps to 16Mbps and now to 30Mbps with no change is cost to me. (Happily, speed tests always show a download speed roughly 10% greater than the advertised 30Mbps which is a welcome change from AT&T DSL which never achieved the full 6Mbps I was paying for.)

I have no reason to believe the modem failure was a result of a surge. If that were the case I would not have posted my question in the first place. I was simply trying to determine if the rapid failure of the cable modems was common or not. Ryck, Alternaut, and Virtual1 have answered that question to my satisfaction.

Update.

I talked to a Charter technician who indicated they have been trying different modems because of an abnormally high failure rate on the Motorola modems they had been using. Apparently I was not the only one experiencing failures.