Recently, a thunderstorm fried my PC’s power supply and its motherboard. Although I wasn’t home at the time, I am pretty sure it was the storm and lightning-related surge of some type that did it, and not a random or coinciding failure, since a printer and cable modem on the same AC line were also zapped. Further, I know it was more than the supply, since a vendor-supplied replacement ($40, two minutes to swap it) did not bring the PC back to life.
I am not sure of the “mechanism” by which the lightning came into the system or what the surge (or whatever) looked like – that can only be done with the appropriate instrumentation connected at the time, of course. None of the other electronics on nearby circuits were affected, not even the cheap AC-line powered clock nearby which usually resets to 12:00 after even a momentary brown-out, when the nearby lights flicker for a moment.
OK, these things happen. In these situations, it’s customary to consider not having it happen again, even if it is too late this time. The power strip into which these items are plugged had a label proclaiming it has built-in “surge protection” but of unspecified rating. If I was going to get a separate surge protection device (SPD), I thought it made sense to do some research to assess what rating(s) would make sense.
Perhaps it shouldn’t be a surprise, but I came away confused. Why so? Certainly, surges and transients are inherently confusing topics. First, it’s not really a surge protector or surge suppressor, as they are called in mass-market consumer literature, a more-accurate name would be a transient protector. There’s also a difference: a surge arrester directs the higher-voltage transient to ground, while a surge suppressor absorbs the energy and re-emits it as heat.
Then there are three non-nomenclature issues: how much of this so-called surge protection do you think you need, how to get it, and where to put it? For the first question, numbers were all over the place. Blogs and sites of questionable credibility recommended 200, 400, and even 600 joules. I wasn’t looking to protect my equipment from a direct lightning strike to the house, just from line-related problems. None of them mentioned that it is not just the joules “energy” rating that matters, but also the time over which that energy is spread. There are industry and regulatory standards for that, such as UL1449, “Standard for Surge Protective Devices.”
On the other hand, the credible sources spent a lot of time discussing important issues of whole-site protection (house, office building, or industrial plant) which wasn’t my concern at this point. The basic options were protection by open circuit or by short circuit. This would be accomplished by voltage-breakdown devices (gas discharge tubes and spark gaps), voltage-limiting devices (metal oxide varistors or avalanche diodes), bandwidth suppressors (various filters, inductors and capacitors), and even isolation devices (opto-isolators and fiber optics – clearly not applicable for the problem) as the principle technologies. They also pointed out that a single protection device or subsystem located far from the device to be protected or its AC outlet may not actually be able to protect, given the nature of lightning surges and reflections.
Then there were several other concerns, assuming I even decided what to do. Credible sources noted that some types of SPDs do wear out from stress due to repeated “hits,” even those that were within their ratings; some suggested replacing the SPD every few years. Worse, some sources called out a report from NEETRAC (the National Electric Energy Testing, Research, and Applications Center) in Atlanta, GA which said that many surge protection devices failed to perform according to ANSI, IEEE, UL, IEC or other required specifications (sorry, I couldn’t find the report itself at their site).
Frankly, that last bit of bad news didn’t surprise me at all. It’s not news that there are lots of fake or substandard components and devices out there. How hard would it be for a vendor of SPDs – especially the cheaper ones you get at big-box stores, to substitute these after getting the UL or similar certification? After all, passives such as those in an SPD are easy to fake, and even dummy units will “work” just fine, until they are called upon to actually protect anything (it’s the same with batteries, of course).
The dilemma with SPDs is that the average user has no way to test the unit, so it’s easy to have false confidence in the one you are using. Further, if you could test your SPD, the tests are either stressful and degrade performance, or are destructive, depending on the SPD. Therefore, you must rely on the vendor credibility and assume (hope?) they do sampling tests of the materials, components, and end product. If the SPD fails to do its job, the vendor can simply claim the surge energy exceeded its ratings, a claim which is impossible for consumers to disprove.
So, what to do? I am still in a quandary, for sure. Until I decide, I’ll rely on the best surge-protection tactic I know: unplug the PC when a storm is imminent, or when I leave for an extended time and the forecast incudes thunderstorms. Perhaps this is not an elegant solution, but it is an effective one.
Do you have any experience with, or suggestions about, surge protection for individual units such as PCs in a residential setting, and SPDs in general?
Related Internal Content
Counterfeit Parts: It's Worse Than We Thought
Teardown: A19 LED bulb
Megavolt/kiloamp tests reveals extreme engineering challenges
Lightning surge protection: Two is better than one!
Importance Of Lightning Surge Protection Devices
The Surge Protection Device (SPD)
NLSI Recommendations for Surge Protection Devices