Toasty Laptop

March 2, 2004 4:30 PM

Why does my laptop run 10 degrees hotter when plugged into a ground fault circuit interrupter receptacle? The machine runs at 40-45C on battery, 45-50C plugged into a normal outlet, and 55-60C on the GFCI in my basement. Plugging into a surge protector doesn't seem to help. Why is this happening, and is there a hardware fix?
posted by PrinceValium to computers & internet (6 comments total)

The only reason I can think of is the interrupter, or something before it on the line, or maybe even a high-power device somewhere on those phases, actually being a filter or a power draw which clips part of the AC sine wave. This can cause the inverter in the laptop's power circuit to increase its duty cycle, which causes heating. The fix? If it's the entire phase pair that's affected, you may not be able to use any outlet on that pair (the pairs are distributed in the fusebox). If it's just the GFCI doing the clipping (though I don't think it's possible), don't use a GFCI.

In any case, the only reason a laptop would be doing that is different power parameters on that circuit.

55-60 C is scorching for a laptop.
posted by azazello at 7:43 PM on March 2, 2004

The laptops I've seen have an external black box for the power conversion. That would seem to eliminate the duty-cycle issue.

The laptop runs cooler on battery probably because it's going into a low-power mode, quite likely slowing the CPU and bus. Less work = less heat.
posted by five fresh fish at 8:15 PM on March 2, 2004

PrinceValium: Out of curiosity, do each of the three conditions you stated above take place in three distinct environments? Laptops do a good deal of heat dissipation through conduction, and I would speculate that the surface on which the machine rests plays a major role in its cooling. Do the lower-temperature situations generally involve setting the laptop on a good conductive (i.e. metal) surface, and the higher-temperature on a poor conductor (i.e. wood)? What about ambient temperature in the room(s)?

Also, charging batteries generally radiate heat. I would assume that, while plugged in, your laptop is charging, and therefore that heat is compounding the heat radiated from the machine components itself.

</unfounded speculation>
posted by Danelope at 8:29 PM on March 2, 2004

This is a Dell Inspiron 8500 2.2Ghz. It's pretty big and lunky, but good for wide-screen DVDs. :)

The surface is typically wood, with the exception of school where it's neo-formica.

The laptops I've seen have an external black box for the power conversion. That would seem to eliminate the duty-cycle issue.

Perhaps there is a problem with my AC adapter then? This is pretty curious. I have a 4 year warranty with Dell, so I'm not shy about getting stuff replaced.
posted by PrinceValium at 9:01 PM on March 2, 2004

You know what? I don't think you need to worry.

I'm pretty sure you have an external black power convertor.

When you're running on AC power, your computer is getting a juicy 20V DC power from the adaptor.

When you're running on batter power, your computer is getting a thin 11.1V DC

I haven't successfully googled ("problem" "power adapter" "gfci" OR "ground fault circuit") ... suppose I could add "ac adapter" to that. But I'll leave that to you. :-)
posted by five fresh fish at 11:38 PM on March 2, 2004

Well, for the real techies, here's how a GFCI is designed.

If the electricity passes through a transformer, I suppose it _could_ be modified if it were bad. However, your laptops power supply *MUST* be able to either handle the problem or blow the fuse (that is, assuming it's UL/CSA/CE certified). IOW, the external power supply should be getting hot, at worst, or supplying too little voltage, both of which are not going to damage your laptop, or cause it to run hot. There's *no way* a properly designed power supply should provide anything but the correct voltage on the output...

...which means that the difference between the two receptacles is due to surrounding environment, rather than power.

Now, if the AC power supply is built in (REALLY unlikely), then the power supplied to it could make a difference.

If you really care to find out more on this, check the voltage output of your power supply with a meter on both plugs. It should be identical. If you are worried about ripple (don't be), hook it to an oscilloscope.
posted by shepd at 3:19 AM on March 3, 2004