View Single Post
  #10 (permalink)  
Old 08-24-2008, 06:12 AM
~misfit~
Guest
 
Posts: n/a
Default Re: PSU temperatures

Somewhere on teh intarweb "VanguardLH" typed:
> ~misfit~ wrote:
>
>> Somewhere on teh intarweb "VanguardLH" typed:
>>> GT wrote:
>>>
>>>> Would a 650W PSU run hotter when powering a 484W computer than when
>>>> powering a 232W setup? Obviously case temperatures would affect
>>>> things, but if we could isolate that...
>>>>
>>>> Does a PSU run hotter when under more load or is its heat output
>>>> independent of the amount of power drawn from it?
>>>
>>> You cannot consume more power without generating more heat. More
>>> power means more heat.
>>>
>>> 1000 Watts = 0.948 BTU

>>
>> Indeed. However, the PSU doesn't 'consume' the power, it supplies
>> (most of) it to other devices which consume it.
>>
>> The question was really about PSU efficiency at different loads.

>
> True, but since efficiency is never 100% then the more power consumed
> by the target devices (mobo, memory, cards, drives, etc.) means the
> more power also consumed by the PSU. For a 600W PSU, if efficiency
> were 70% (a measure in difference between input and output power at
> some load), a load of 300W means the PSU is consuming 75W (300/.8 -
> 300, or 375W input power minus the 300W load). For a 400W load, the
> PSU eats 100W ( 400W/.8 - 400W, or 500W input power minus the 400W
> load). As the PSU consumes more power under a higher load, the more
> heat is produced by the PSU itself. Of course, assuming 80%
> efficiency is being generous since, as you said, that's typical of
> "good" power supplies.
>
> So back to the OP's question of will a PSU run hotter with a 484W load
> than with a 232W load, and whether efficiency were 70% or 80% for the
> PSU, my answer still applies. More power consumed means more heat.
> No matter what PSU the OP uses and irrelevant of the PSU's maximum
> capacity (i.e., max load), you cannot *consume* more power without
> generating more heat because. Efficiency is never 100% for the PSU.


All true (except the 70% typo, that threw me for a while <g>).

I was simply elucidating to answer the OP's second question; "is its heat
output independent of the amount of power _drawn_ from it?" You answered
with a reply talking about *consuming* power, I just wanted to cover
*supplying* power.

Cheers,
--
Shaun.

DISCLAIMER: If you find a posting or message from me
offensive, inappropriate, or disruptive, please ignore it.
If you don't know how to ignore a posting, complain to
me and I will be only too happy to demonstrate... ;-)



Reply With Quote