17

My housemates don't want me setting up a home server because of fire hazard, I've seen very little, if anything, pointing to desktops causing fires, just laptop batteries (Can overheating laptops cause fire?). I've got this system all set up with Proxmox and if that was in vain... (I accidentally started the OS setup with the installer flash drive on a USB 1.1 port so it took FOREVER)

I guess the question is... if I had a desktop on 24/7 unsupervised, would there be a non-negligible chance for it to catch on fire assuming stock CPU cooler, no overclock, PSU not near full capacity, dusting/compressed air regularly, and standard maintainence?

Improve this question
15
  • 5
    I looked around, and there seem to be a lot of people asking the same question (e.g.), so it is notable.
    – Oddthinking ♦
    Commented Jan 20, 2015 at 4:01
  • 2
    This should be relevant here. I also leave my home desktop computer on almost 24-7 with no issues for the last 4 years.
    – Ian Gallant
    Commented Jan 20, 2015 at 12:49
  • 2
    @Flimzy: But most offices have dozens to hundreds of standard PCs, many of which seem to be left running 24/7. And of course, automatic fire suppression systems protect the valuable data center from fires from any cause, not just those started by the computer equipment.
    – jamesqf
    Commented Jan 20, 2015 at 19:13
  • 4
    I think that unless your roomates are willing to forgo all electronic devices (TV, Microwave, stove, etc) because of the fire danger, they are being unreasonably paranoid that a computer is going to start a fire. There's probably a greater chance that a cell phone battery will cause a fire than a computer.
    – Johnny
    Commented Jan 21, 2015 at 17:44
  • 4
    I'll add that I managed IT for a 500 person company for 5 years - during that time we had zero fires caused by computers -- 75% desktops, 25% laptops, most kept on 24x7. We lost one server room power supply - smoke only, no visible fire or damage outside of the server, we had one small fire due to paper falling on an under-desk space heater (which was specifically against policy), several break room fires in the microwave and toaster oven, and one accidental fire suppression discharge when movers knocked off a sprinkler head).
    – Johnny
    Commented Jan 21, 2015 at 17:53

1 Answer 1

Reset to default
18

Do desktop computers catch fire often?

http://www.nfpa.org/~/media/Files/Research/Fact%20sheets/office_fact_sheet.pdf

In 2007-2011, there were an estimated 730 reported U.S. home structure fires involving office equipment per year... [of these] Computers [accounted for] 48%
In 2007-2011, there were an estimated 630 reported U.S. non-confined non-home structure fires involving office equipment per year... [of these] Computers [accounted for] 36%

So a little calculation: 48% of 730 + 36% of 630 = 350 + 227 = 577 computer fires per year. [Note that this category is a bit confusingly worded, I'm not sure what constitutes "office equipment" but I couldn't find computers in any other category, so they may all be accounted for in this one.]

So computers seem to catch fire on average about once every 15 hours in the US. (2007-2011)

Is the risk non-negligible?
http://www.nfpa.org/~/media/Files/Research/NFPA%20reports/Occupancies/oshomes.pdf

During the five-year period of 2007-2011, U.S. fire departments responded to an estimated average of 366,600 home structure fires per year. ...

Leading Causes of Home Structure Fires: 2007-2011
Cooking equipment 43%
Heating equipment 16%
Intentional 8%
Electrical Distribution and Lighting equipment 6%

A little more math:
43% of 366,600 = 157,638 US home fires caused by cooking equipment
48% of 730 = 350 US home fires caused by computers (see above)
350/157,638 = 1 : 450
You are 450 times as likely to have a (reported) US home fire from cooking equipment than a computer.

Maybe you could just eat out one more day per year instead of cooking, this should satisfy your safety conscious room-mates by mitigating the added risk of an extra computer.

Improve this answer
8
  • 3
    The statistics you referenced: are they for desktop computers, or are they for all computers (including laptops, and/or including rechargers)?
    – ChrisW
    Commented Jan 24, 2015 at 14:01
  • @ChrisW It's hard to tell. Explicitly not telephones, answering machines, "computer related", or "other office". I'll see if I can track down a better explanation.
    – Adam Phelps
    Commented Jan 24, 2015 at 14:07
  • @ChrisW I dead ended on that particularly, need to pay for more info. But I found an older report (1999-2003) that had similar metrics for a more explicit "computers", "computer monitors", "computer printers", "unclassified computers" and "mobile computers" were mentioned. So I think the above probably refers to all computers. nfpa.org/~/media/Files/Research/NFPA%20reports/…
    – Adam Phelps
    Commented Jan 24, 2015 at 14:18
  • 2
    "All computers" is OK: it means that the number you referenced is an upper bound, i.e. that the number or desktop computers is equal-to-or-less-than (i.e. no greater than) the number you referenced.
    – ChrisW
    Commented Jan 24, 2015 at 14:39
  • @ChrisW and servers, and desktops running server software, and old dusty laptops running server software in closets with newspapers on top and a tank of gasoline in the corner.
    – Adam Phelps
    Commented Jan 25, 2015 at 3:47

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .