I have question. With the talk of coolers, water, hvac, & fans on the forum, does a heated cpu temps rise after shut down or not?
1. This may sound funny, but I do automotive designs & we know that the temps raise sfter the engine is shut down & design the correct upper limit temps to off set this.
2. If I take a roast out of the oven & put in a meat theromoeter in it on the kitched table the temp will rise as well. I'm sure there other examples out there also.
The question is: If you are crunching the cpu up to almost its upper limit on temp, & without a cool down period, will it burn up? My reason is the fan heat sink would be rather hot at that point as well as the cpu, I think. The temps would seem to have to raise some more. I could be completely wrong in my thinking.
How do you eat a elephant? One bite at a time!
More about :cpu heat
May 30, 2001 9:05:15 PM
Hmmm...Sounds logical...When you shut off the power, the fan would go off and stop cooling the CPU, so the temp will rise. I think though that the Tbird's "upper limit" is 90C, so you'll never ever be running at close to that, unless you like looking at a blank screen, thinking about your burning CPU. What you'll usually run at OC'd is hopefully a max of 60C, and if that rises a couple of degrees, you'll definately be OK. The CPU won't be under load, and will cool off rapidly because it can handle "spikes" into the upper range, no prob.
It's raining outside, and my lawn has grown a foot overnight!
You're absolutely right. The heat is transferring around the object for quite a while after the energy source is removed. You'll find the temp of the exterior is cooling while the centroid temperature is rising. Temperature differentials exist in every material, be it solid/liquid/gas. The density and specific heat of the object are the controlling factors on the maximum temperature reached.
That processor heats up after you turn off the fan, but since it's such a small mass, it only rises a few degrees centigrade. You can check this out yourself with a small temp probe. Passive cooling drops the temp back down in less than 30 seconds. If you're running anywhere near the limit (90-95C), most likely you're melting the transistors when the cpu is on, not to mention shutdown.
If you're running a aircooled peltier, you will heat the bejesus out of the processor when you shut down. You'd have to wire up a AT power supply (or rig an atx) and leave the cpu fan after shutdown, but then you'd have to hotwire the leads for the cpu fan header on the motherboard so it doesn't go nuts when you turn on the power. If you're watercooling, you can leave the pump on for about 15 seconds to dump the heat.
Very good point, I used to be a mechanic and know the probs that afterboil can cause, but in the case of cpu's, (not sure about peltier setups), I shouldnt imagine its the same, once the power is off, it cant heat up any more, of course there is the issue of the heat thats "already there" but Im sure that the average heatsink is enough ti dissipate this heat. (mines not blown up yet!!)
That's what I though. I'm not trying to burn up my chip, I was just doing some thinking about some cooler designs that I have ratteling around in my head, & as a designer I'm just getting down the parameters in my head to do a design. Don't know what will come from it, but I'm always looking for the better route.