High-wattage PSU on a Low-wattage Desktop -- Additional watts consumed? (math problem)
Tags:
-
Components
Last response: in Components
Upgrader3
May 16, 2014 7:20:41 AM
- Suppose that a desktop computer has a 400W, top-quality, 80+ Certified PSU [assume it is the best-quality, most efficient (consumer-level), 400W PSU readily available from Newegg, Amazon, etc].
- Suppose that the computer averages pulling about 200 watts [measured at the wall outlet] on an average day.
- Now suppose that the 400W PSU is replaced with an 800W PSU, and that is the only change made to the computer. . . [Assume that the 800W PSU is the best-quality, most-efficient (consumer-level) 800W PSU readily available from a vendor such as Newegg, Amazon, etc]
1) . What will now be the average power draw in watts [measured at the wall outlet] with the 800W PSU installed?
2) . If the 800W PSU were just a run-of-the-mill, mediocre-quality model (but still 80+ Certified), how much different could the answer [to question #1} be?
I'd like to see actual calculations, not just a qualitative answer.
Thanks for your answers.
- Suppose that the computer averages pulling about 200 watts [measured at the wall outlet] on an average day.
- Now suppose that the 400W PSU is replaced with an 800W PSU, and that is the only change made to the computer. . . [Assume that the 800W PSU is the best-quality, most-efficient (consumer-level) 800W PSU readily available from a vendor such as Newegg, Amazon, etc]
1) . What will now be the average power draw in watts [measured at the wall outlet] with the 800W PSU installed?
2) . If the 800W PSU were just a run-of-the-mill, mediocre-quality model (but still 80+ Certified), how much different could the answer [to question #1} be?
I'd like to see actual calculations, not just a qualitative answer.
Thanks for your answers.
More about : high wattage psu low wattage desktop additional watts consumed math problem
-
Reply to Upgrader3
Traciatim
May 16, 2014 7:27:50 AM
The power draw on the wall side isn't determined by the PSU capacity but it's efficiency.
If both supplies are 80% efficient and the machine needs 200Watts on the DC side then both cases the wall side will require 250 watts (200 / .8). In your example, 200 * .8, or 160 watts would be supplied to the machine while 200 came from the wall, 40 being converted to heat.
The trouble is that power supplies generally have an efficiency curve where they are more efficient somewhere near 50-75% of their rated capacity and it gets far less efficient at the very edges of their power capacity (especially at extremely low power usage). Generally it's not that much different for real world usage that you have to care all that much and just using it's peak efficiency rating is fine for most real world calculations.
If both supplies are 80% efficient and the machine needs 200Watts on the DC side then both cases the wall side will require 250 watts (200 / .8). In your example, 200 * .8, or 160 watts would be supplied to the machine while 200 came from the wall, 40 being converted to heat.
The trouble is that power supplies generally have an efficiency curve where they are more efficient somewhere near 50-75% of their rated capacity and it gets far less efficient at the very edges of their power capacity (especially at extremely low power usage). Generally it's not that much different for real world usage that you have to care all that much and just using it's peak efficiency rating is fine for most real world calculations.
-
Reply to Traciatim
m
0
l
IRONBATMAN
May 16, 2014 7:29:00 AM
Best solution
Traciatim
May 16, 2014 7:30:41 AM
Just as an example, if the 400 was 85% efficient and was measured at 200 watts from the wall, and you replace it with an 800 watt but 80% efficient supply you would do this:
200 * .85 = 170 Watts.
170 / .8 = 212.5 watts.
The 800 watt but less efficient supply would use 12.5 watts more than the more efficient supply.
200 * .85 = 170 Watts.
170 / .8 = 212.5 watts.
The 800 watt but less efficient supply would use 12.5 watts more than the more efficient supply.
-
Reply to Traciatim
Share
Upgrader3
May 16, 2014 7:41:01 AM
Traciatim said:
The power draw on the wall side isn't determined by the PSU capacity but it's efficiency. If both supplies are 80% efficient and the machine needs 200Watts on the DC side then both cases the wall side will require 250 watts (200 / .8). In your example, 200 * .8, or 160 watts would be supplied to the machine while 200 came from the wall, 40 being converted to heat.
The trouble is that power supplies generally have an efficiency curve where they are more efficient somewhere near 50-75% of their rated capacity and it gets far less efficient at the very edges of their power capacity (especially at extremely low power usage). Generally it's not that much different for real world usage that you have to care all that much and just using it's peak efficiency rating is fine for most real world calculations.
Thanks, this helps, but still begs the question of a specific quantitative answer.
1) . Can anyone name specifically, a circa-800W PSU that would likely have the highest (or among the highest) efficiency (of any 800W PSU) when operating at 25% of its rated-800W-capacity?
2) . And does the manufacturer of that PSU publish a complete efficiency-profile, showing how the efficiency varies from a 1% load to a 100% load?
3) . And what then is the efficiency when the PSU is operating at a 25% load?
-
Reply to Upgrader3
m
0
l
Doc Olsenenator
May 16, 2014 7:41:01 AM
Presuming a nearly perfect world for your question. There is a loss of approx. 20% to heat in converting the AC current to DC current so 20% of 400 watts is 80 watts loss to heat plus the 200 watts of the system = 280 watts measured at the plug for system 1. System 2 experiences the same approx. 20% loss due to heat too. 20% of 800 watts is 160 watts plus the system of 200 watts for system 2 = 360 watts measured at the plug. So you your difference is the additional 80 watts of waste heat of system 2 over system 1.
- Suppose that the computer averages pulling about 200 watts [measured at the wall outlet] on an average day.
- Now suppose that the 400W PSU is replaced with an 800W PSU, and that is the only change made to the computer. . . [Assume that the 800W PSU is the best-quality, most-efficient (consumer-level) 800W PSU readily available from a vendor such as Newegg, Amazon, etc]
1) . What will now be the average power draw in watts [measured at the wall outlet] with the 800W PSU installed?
2) . If the 800W PSU were just a run-of-the-mill, mediocre-quality model (but still 80+ Certified), how much different could the answer [to question #1} be?
I'd like to see actual calculations, not just a qualitative answer.
Thanks for your answers.
Upgrader3 said:
- Suppose that a desktop computer has a 400W, top-quality, 80+ Certified PSU [assume it is the best-quality, most efficient (consumer-level), 400W PSU readily available from Newegg, Amazon, etc].- Suppose that the computer averages pulling about 200 watts [measured at the wall outlet] on an average day.
- Now suppose that the 400W PSU is replaced with an 800W PSU, and that is the only change made to the computer. . . [Assume that the 800W PSU is the best-quality, most-efficient (consumer-level) 800W PSU readily available from a vendor such as Newegg, Amazon, etc]
1) . What will now be the average power draw in watts [measured at the wall outlet] with the 800W PSU installed?
2) . If the 800W PSU were just a run-of-the-mill, mediocre-quality model (but still 80+ Certified), how much different could the answer [to question #1} be?
I'd like to see actual calculations, not just a qualitative answer.
Thanks for your answers.
-
Reply to Doc Olsenenator
m
0
l
Traciatim
May 16, 2014 8:03:54 AM
Upgrader3 said:
Traciatim said:
The power draw on the wall side isn't determined by the PSU capacity but it's efficiency. If both supplies are 80% efficient and the machine needs 200Watts on the DC side then both cases the wall side will require 250 watts (200 / .8). In your example, 200 * .8, or 160 watts would be supplied to the machine while 200 came from the wall, 40 being converted to heat.
The trouble is that power supplies generally have an efficiency curve where they are more efficient somewhere near 50-75% of their rated capacity and it gets far less efficient at the very edges of their power capacity (especially at extremely low power usage). Generally it's not that much different for real world usage that you have to care all that much and just using it's peak efficiency rating is fine for most real world calculations.
Thanks, this helps, but still begs the question of a specific quantitative answer.
1) . Can anyone name specifically, a circa-800W PSU that would likely have the highest (or among the highest) efficiency (of any 800W PSU) when operating at 25% of its rated-800W-capacity?
2) . And does the manufacturer of that PSU publish a complete efficiency-profile, showing how the efficiency varies from a 1% load to a 100% load?
3) . And what then is the efficiency when the PSU is operating at a 25% load?
Some companies do release an efficiency curve chart. You can use the 80Plus spec they fall under though:
http://www.plugloadsolutions.com/80pluspowersupplies.as...
As an example, an 80Plus Gold rated supply will be 87% efficient at both 20% capacity and 100% capacity, but 90% efficient at 50% capacity.
The difference between 80Plus specs is pretty minimal though unless you are going to be doing something like a coin mining machine where you are maxing out your gear 24 hours a day. Generally going 80Plus Gold or Platinum are more expensive to buy and don't pay for themselves over the life of the supply.
-
Reply to Traciatim
m
0
l
Traciatim
May 16, 2014 8:05:40 AM
Doc Olsenenator said:
Presuming a nearly perfect world for your question. There is a loss of approx. 20% to heat in converting the AC current to DC current so 20% of 400 watts is 80 watts loss to heat plus the 200 watts of the system = 280 watts measured at the plug for system 1. System 2 experiences the same approx. 20% loss due to heat too. 20% of 800 watts is 160 watts plus the system of 200 watts for system 2 = 360 watts measured at the plug. So you your difference is the additional 80 watts of waste heat of system 2 over system 1. This assumes that you are using the maximum amount of power that each supply can deliver, not using it on the same machine.
-
Reply to Traciatim
m
0
l
Traciatim
May 16, 2014 8:25:25 AM
As for the request of specific power supplies that are very efficient and somewhere near 800 watt, just go to a shop that you can sort by 80Plus rating like NewEgg and here you go:
http://www.newegg.com/Product/Product.aspx?Item=N82E168...
http://www.newegg.com/Product/Product.aspx?Item=N82E168...
http://www.newegg.com/Product/Product.aspx?Item=N82E168...
http://www.newegg.com/Product/Product.aspx?Item=N82E168...
http://www.newegg.com/Product/Product.aspx?Item=N82E168...
http://www.newegg.com/Product/Product.aspx?Item=N82E168...
http://www.newegg.com/Product/Product.aspx?Item=N82E168...
http://www.newegg.com/Product/Product.aspx?Item=N82E168...
http://www.newegg.com/Product/Product.aspx?Item=N82E168...
http://www.newegg.com/Product/Product.aspx?Item=N82E168...
http://www.newegg.com/Product/Product.aspx?Item=N82E168...
http://www.newegg.com/Product/Product.aspx?Item=N82E168...
http://www.newegg.com/Product/Product.aspx?Item=N82E168...
http://www.newegg.com/Product/Product.aspx?Item=N82E168...
http://www.newegg.com/Product/Product.aspx?Item=N82E168...
http://www.newegg.com/Product/Product.aspx?Item=N82E168...
-
Reply to Traciatim
m
0
l
Read discussions in other Components categories
!