I know, right? I mean, I understand if it might cause a power surge to an electrical supply, but why would that crash a computer? Wouldn't the battery take over if the power went out? Just goes to show, I suppose, how fragile computers can be...
I know of one occasion where there was a computer on a UPS in an apartment where the electrical plug was on the same circuit breaker as part of the kitchen. An appliance in the kitchen caught fire which triggred the circuit breaker. The UPS alarm was triggered well before the smoke detector went off.
When I first got into industrial systems I worked at a company that made a lot of production line test equipment. Our chief engineer demanded that every machine that utilized logic more complex than relays had to have a Sola Constant Voltage Transformer. Of course, it did add to the machine cost, but those machines never suffered from power problems, other than total feed collapse. The CVTs were not that efficient, they still run too hot to touch, even today, but preventing both spikes and dips from getting into the sensitive circuits has avoided a wole big lot of headaches and grief. So possibly just using a CVT could be beneficial.
But I did work at one company where my UPS would sound it's warning every day a bit after three. It would happen on many of the ups systems on many of the computers. WE never did find the cause, but the systems paid for themselves in less than a year, in lost time avoided. So sometimes the workaround is cheaper thanbnan actual solution. "Long live the UPS"
A good quality UPS does a lot more than solve a power loss problem during operation--that's an aspect of their power conditioning function. It also, and most important, provides clean, conditioned backup uninterruptible power (UP) when the regular AC goes down. When I worked in the field, the main machines were great, big honking things for hospitals and the IT departments of major corporations. Now we've also got very low-end systems, as bob from maine describes (although those excessive conditions aren't normal where I live). These products are over-priced, not very capable, and not very useful. They're also extremely user-unfriendly, at least to purchase correctly, unless you happen to be an electrical and/or UPS specialist. After 25 years, I have yet to buy one for my home office.
I used to do computer controllered conveyor systems and my partner and I would work through the night and never have a problem. But everyday around 10 am, the computer would crash. We reasoned it was due to brownouts and finally were persuasive enough to get a UPS added and the problem went away.
I had another site that kept crashing and the plant managers swore up and down that they were using less power than they ever had and pointed to their terminals attached to their mainframe that power couldn't be the problem. I pointed out their mainframe was on the other side of a massive UPS and those were only terminals. Did they have any PCs in the plant? They said only on the worker punchclock and that the vendor had placed a line monitor on it. So I wasn't the only one suspecting power problems. The next morning on the cab ride to the plant I asked the cabbie if he knew anything about the area, He said yes, he remembered when the plant was the only building in this huge industrial park we were driving through. One UPS later, the problem was solved.
I'd come back to visit my home town and my aunt would handcuff me to her machine to fix it. I determined the power in her small town flickered enough to scramble her hard drive controller, but leave her CPU running just fine. One UPS later, I could come and just visit instead of spending my entire visit fixing her computer.
From what I have heard, we still have scheduled brownouts but they are carried out less obtrusively than in the past. I would imagine there was some type of drop in voltage or dip as Ann mentioned - it would be interesting to know the exact cause. While UPS's would solve the power loss problem during equipment operation - maybe the root cause could be addressed...good job making the connection between the computer's power loss and that billboard!
We have these wonderful strip-malls here in Maine that have one single power feed, they're usually on a fairly rural road and the 5 or so tenants which vary from high energy metal processing to insurance agencies to supermarkets, all in the same building. Firing-up a high energy plasma cutter and air compressor, a couple of refrigeration compressors and an annealing oven all at the same time tends to put more noise on the AC lines than any sane designer ever imagined. Now take your critical items like PC's in insurance and law offices, cash registers in supermarkets, wi-fi sites in book stores, and try to convince all of them they need UPS at THEIR cost. Our power distribution grid in some of these areas is still using #10 copper insulated with tar installed in the 1920's. It's asking an awful lot of a 21st century power-supply make by the lowest bidder to deal with that kind of noise.
I worked in power line analyzing/conditioning and UPS decades ago: it's a field ripe for Sherlock Ohms stories. Thanks for this one, although I agree with 3drob--what exactly was the computer's failure mode? In this situation, a dip of some kind may be more likely to occur after the billboard comes on than a spike or surge, unless a grid adjustment then causes a spike or surge.
Focus on Fundamentals consists of 45-minute on-line classes that cover a host of technologies. You learn without leaving the comfort of your desk. All classes are taught by subject-matter experts and all are archived. So if you can't attend live, attend at your convenience.