I'm a former field engineer for a US Cellular carrier.
I don't understand how a two hour outage is a panic event, both Cellular CO's and Cell Sites should have at least 3-4 hours of holdover time. Thats just the standard of design.
Like, what happens if I storm blows thru, knocks down the infrastructure, if you have heat related infrastructure failures, or some other just normal failure - power outages happen, for those sites in the states where long power outages are common they'll have a permanent generator on site.
> Europe has nearly half a million telecom towers and most of them have battery backups that last around 30 minutes to run the mobile antennas.
Which seems bonkers to me. Perhaps they deploy incredibly small batteries because they're engineered for a different outage profile than the US? Or perhaps the reporter is quoting something inappropriately -- traditional large sites with many hours, and many new small sites with no backup at all, might well average out to 30 minutes.
I think there's real work to be done on improving bring-up time when power returns, though. I worked on the EVDO rollout and rebooting a site after work would take 20-30 minutes because it's something that's only expected to happen once in a blue moon. I believe if they relaxed GPS timing tolerances a bit and optimized a few other things, this could be shortened to 5-ish minutes without a whole lot of work.
Just for some perspective: In my entire life in Germany, I experienced two power outages, both more than 20 years ago. The grid is rock-solid here; this might not warrant for huge battery capacity in general (even though 30 minutes does seem quite small).
The grid in the states is rock solid here, in general - certain cases not withstanding. Power distribution in the states is largely arial, which means deeply effected by large weather events, so while the power might be on - on the transmission side - it doesnt mean that the distribution side is in any condition to deliver any to you at that moment. I cant say my overall experiences with power are different than yours, I dont think I've ever had an outage longer than one hour, and I've only had four power loss events, each less than a half an hour in duration over the last 20 years.
The US has many extreme weather events - Hurricanes, Blizzards, Floods, Wind Storms, Ice Storms, Tornados, Wildfires (yes, thats sorta weather), Heavy Rain events, etc - that will often bring down trees onto distribution circuits. I may be wrong, but I have the belief (based on news reporting) that weather events which in the states are considered 'normal' would be 'extreme' in Europe.
The Samsung gear I worked on would take about 10 min to come back up, give or take. I know because I dropped a handful of sites when I did power testing - visual inspection was not always enough to verify if the power was wired correctly. Most of them came back up in 10 min.. well, except once there was one site that didn't come back because the Microwave didn't like the power hit, lol - that required another tower crew to revisit and hang a new RRU.
I just can't imagine power design is that different, I'm used to these cabinets even in urban areas having 4 strings of batteries in them, are they only building sites with a single string in europe?
I don't understand how a two hour outage is a panic event, both Cellular CO's and Cell Sites should have at least 3-4 hours of holdover time. Thats just the standard of design.
Like, what happens if I storm blows thru, knocks down the infrastructure, if you have heat related infrastructure failures, or some other just normal failure - power outages happen, for those sites in the states where long power outages are common they'll have a permanent generator on site.