Does upgrading PSU to a higher efficiency reduce heat?

muddle

Supremacy Member
Joined
Apr 26, 2000
Messages
6,988
Reaction score
0
My seasonic x650 also doesnt spin its fan until it hit certain load. Coming to 5 years already.
 

annoymous

Supremacy Member
Joined
Jun 23, 2006
Messages
9,122
Reaction score
1
Just use until explode first, if never explode then no need to worry :D

fwah if explode alr all the parts will ki chia together?


It's decent la, just don't recycle it for your next build since it's using capxon caps which don't last long.

that time buy i thought the casing there say japanese 105 deg rated caps, yuan lai tio scam
 

crystalnox

Supremacy Member
Joined
Feb 2, 2006
Messages
8,735
Reaction score
2,326
fwah if explode alr all the parts will ki chia together?




that time buy i thought the casing there say japanese 105 deg rated caps, yuan lai tio scam
Primary side is Japanese, they never claim all are Japanese. :s13:

n04d2368a656070d6a45d74594a741cfe.jpg
 

killerslayer92

Senior Member
Joined
Nov 28, 2007
Messages
1,061
Reaction score
0
psu have a specific load at which they are most efficient. each psu is different. rated at 650w doesnt mean efficient at 650w output

analogy for illustration
650w load efficiency $1/w
600w load efficiency $0.50/w
550w load efficiency $0.30/w
500w load efficiency $0.50/w
450w load efficiency $1/w

maximum efficiency = 550w load

conclusion: dont buy too powerful psu, 20% extra can liao
 

crystalnox

Supremacy Member
Joined
Feb 2, 2006
Messages
8,735
Reaction score
2,326
psu have a specific load at which they are most efficient. each psu is different. rated at 650w doesnt mean efficient at 650w output

analogy for illustration
650w load efficiency $1/w
600w load efficiency $0.50/w
550w load efficiency $0.30/w
500w load efficiency $0.50/w
450w load efficiency $1/w

maximum efficiency = 550w load

conclusion: dont buy too powerful psu, 20% extra can liao
A more accurate illustration would be

650w load efficiency $1/w
600w load efficiency $0.95/w
550w load efficiency $0.90/w
500w load efficiency $0.95/w
450w load efficiency $1/w
100w load efficiency $1.20/w

The efficiency variation typically <10% from 20-100% load on those higher end psus. :)
 

Maeda_Toshiie

Supremacy Member
Joined
May 12, 2007
Messages
6,310
Reaction score
3
Two points that I would like to make:

1. Taiwanese caps cover quite a range. While their best are not on the same level Nichicon or Hitachi rated at 105, some like Teapo are not complete crap. Obviously avoid stuff like Fuhjyyu.

Of course, if you are from Badcaps, then by all means enforce the all Japanese cap rule on your purchases...

Note that if the advertising claim Japanese caps, it usually means only the primary, not the secondaries (eg. some Seasonic units, like mine). If all Japanese caps, the advertising will loudly proclaim that. If polymer caps are used, expect that to be mentioned.

2. Calculating cost savings from efficiency simply based peak load is misleading since most people do not purely game on the said machine (ignoring 24/7 mining, folding, and encoding here). If you want to "optimize", you must look at the typical usage pattern and the loads resulting from it.
 

Ultimate Chicken

Junior Member
Joined
Oct 1, 2009
Messages
83
Reaction score
2
Frankly, I'm not sure how the OP logic is supposed to work. A PSU mismarked because its from a crap supplier will seem inefficient, but that's only because its mismarked and is actually generating Y instead of X power - the missing power is simply not there and therefore won't ever become heat.

That being said, NOBODY has done any tests on which PSU causes the system to be less hot. >_>'
 

annoymous

Supremacy Member
Joined
Jun 23, 2006
Messages
9,122
Reaction score
1
Two points that I would like to make:

1. Taiwanese caps cover quite a range. While their best are not on the same level Nichicon or Hitachi rated at 105, some like Teapo are not complete crap. Obviously avoid stuff like Fuhjyyu.

Of course, if you are from Badcaps, then by all means enforce the all Japanese cap rule on your purchases...

Note that if the advertising claim Japanese caps, it usually means only the primary, not the secondaries (eg. some Seasonic units, like mine). If all Japanese caps, the advertising will loudly proclaim that. If polymer caps are used, expect that to be mentioned.

2. Calculating cost savings from efficiency simply based peak load is misleading since most people do not purely game on the said machine (ignoring 24/7 mining, folding, and encoding here). If you want to "optimize", you must look at the typical usage pattern and the loads resulting from it.

good info there :s12:
 

Mem2003

Member
Joined
Sep 30, 2003
Messages
440
Reaction score
1
1. AC power (wall) converts to DC power [Loss power: become heat]
1a.Depend on system required {[??? so higher PSU rated will run cooler ???]}
e.g. Use System 500w , same temperature
efficicency=80% {500/0.8}-500=125W are turned into heat
efficicency=90% {500/0.9}-500=55W are turned into heat

1b. If Actual system consume(draws) = 400W-to-600W


2. PSU rated at 500W can deliver 500W of power to the system regardless of its efficiency
2a. Buy PSU rated at which ??? : 500W; 600W, 800W, 1000W, etc
2b. Buy PSU efficiency which ?? : at 80+, 90+, etc
2c. Plot Chart on actual PSU Load level: idle, 25% 50% 75% 100%
2d. Plot measure Temperature level at difference PSU Load level
2e. Brand/ model/ etc
Affects .... .... ... system consume(draws) from idle, normal load to ....
2f. Ageing factor / ETC

=========================================
My pc with i5 3570 + gtx 670 is running really hot on my corsair hx520.
I am hoping by switching over to seasonic 660w platinum psu with more overhead wattage and higher efficiency will reduce the temperature.

I know that a higher efficiency psu will reduce heat. But can't seem to get any info on how much temperature it will reduce. Will the drop in temperature be an obvious difference? Is it worth the hassle to change?
 
Last edited:

a562140

Supremacy Member
Joined
Mar 12, 2004
Messages
6,412
Reaction score
3
psu have a specific load at which they are most efficient. each psu is different. rated at 650w doesnt mean efficient at 650w output

analogy for illustration
650w load efficiency $1/w
600w load efficiency $0.50/w
550w load efficiency $0.30/w
500w load efficiency $0.50/w
450w load efficiency $1/w

maximum efficiency = 550w load

conclusion: dont buy too powerful psu, 20% extra can liao

yah go look for such charts on review sites if possible.
 

escksu

Banned
Joined
Nov 24, 2011
Messages
8,391
Reaction score
3
Me using Seasonic X760 G now..... Good stuff.....

Don't know what coil whine, never heard it before.....
 
Important Forum Advisory Note
This forum is moderated by volunteer moderators who will react only to members' feedback on posts. Moderators are not employees or representatives of HWZ Forums. Forum members and moderators are responsible for their own posts. Please refer to our Community Guidelines and Standards and Terms and Conditions for more information.
Top