Are energy efficent PSUs really worth it? - AVS Forum
Forum Jump: 
 
Thread Tools
post #1 of 16 Old 12-26-2008, 10:26 AM - Thread Starter
AVS Special Member
 
bjmarchini's Avatar
 
Join Date: Sep 2007
Posts: 3,977
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I need to pick up a powersupply for a new PC build that I am doing. I am a very experienced builder and understand Powersupplies.

I am just wandering if these 80 Plus Certified units are worth it economically.

If I install 500 Watt PSU that is >80% then I am wasting about 100 watts. If I went with a 600 Watt to compensate and it get lets say >65%, then I would be 210 watts. This is a difference of 110 watts.

If I left this running 24/7, I am coming up with the following math under full load.

110 watts x 24 hours x 30 days = 79200 watts / 1000 = 79.2 Kwatts

I think I am at $.15 per kwatt so that is a savings of $11 per month.

But this is under full load. I would say that it would it would only be under full load for about 15% of the time at most. The other 85% it would be relatively idle with an occasional drive access.

So I coming up with an actual difference of $1.65 (15%) and $4.25 (85% at half power).

So that is actually $6 per month. And this is assuming that the efficiency doesn't drop over time. Does it?

Anyone else have any thoughts?

HD DVD: 45 SD DVD: 350 BR: 120
PCs: 12, Mame Arcade:1, HD HTPC: 1, WHS Server: 1, HD A3: 2, HD-A30: 1

bjmarchini is offline  
Sponsored Links
Advertisement
 
post #2 of 16 Old 12-26-2008, 10:48 AM
Senior Member
 
CSMR's Avatar
 
Join Date: Mar 2004
Posts: 248
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 4 Post(s)
Liked: 12
Quote:
Originally Posted by bjmarchini View Post

If I install 500 Watt PSU that is >80% then I am wasting about 100 watts. If I went with a 600 Watt to compensate and it get lets say >65%, then I would be 210 watts. This is a difference of 110 watts.

I have no idea what you mean by "compensate" here.
You are assuming that the power draw of your components is 500W: you have some multi-processor, multi-graphics card system with attached storage farm?
Quote:


So that is actually $6 per month. And this is assuming that the efficiency doesn't drop over time. Does it?

I don't think it does.
Quote:


Anyone else have any thoughts?

80+ is just one indicator, what is important it the efficiency at the power level you are running.
Cost of power is just one disadvantage of an inefficient supply. The more power it wastes, the more heat that needs to be got rid of.
CSMR is offline  
post #3 of 16 Old 12-26-2008, 11:01 AM
AVS Special Member
 
nm88's Avatar
 
Join Date: May 2003
Location: San Francisco, California
Posts: 2,685
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by bjmarchini View Post

Anyone else have any thoughts?

Your calculations are all wrong. With a typical consumption of 200W, you are drawing about 250W for an 80% efficient supply and about 308W for a 65% efficient power supply, a difference of about 42.5 KWh per month. This is assuming you're running the system all the time, which you shouldn't be anyway if you care at all about power consumption.

Remember, you're not only saving money, but polluting less when you use more efficient equipment and put them in a sleep state when not in use.

And more efficient power supplies produce less heat, and are therefore easier to cool.
nm88 is offline  
post #4 of 16 Old 12-26-2008, 11:28 AM - Thread Starter
AVS Special Member
 
bjmarchini's Avatar
 
Join Date: Sep 2007
Posts: 3,977
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by nm88 View Post

Your calculations are all wrong. With a typical consumption of 200W, you are drawing about 250W for an 80% efficient supply and about 308W for a 65% efficient power supply, a difference of about 42.5 KWh per month. This is assuming you're running the system all the time, which you shouldn't be anyway if you care at all about power consumption.

Remember, you're not only saving money, but polluting less when you use more efficient equipment and put them in a sleep state when not in use.

And more efficient power supplies produce less heat, and are therefore easier to cool.

even at 42.5 that is about $6 per month... so I am not that far off. And as you said, that is 100% usage. 500 is a peak and not normal usage.

I am not so much worried about the environmental impact. The only difference seems to be a slight reduction in kilowatts. I am already green enough in my eyes with a fuel efficient car, the green bulbs and such. I am just looking at it economically. I would consider it for an HTPC, but this is going to be a ventilated desktop that will also hold my media HDDs (dual function) so I am not as worried about heat buildup.

HD DVD: 45 SD DVD: 350 BR: 120
PCs: 12, Mame Arcade:1, HD HTPC: 1, WHS Server: 1, HD A3: 2, HD-A30: 1

bjmarchini is offline  
post #5 of 16 Old 12-26-2008, 11:48 AM
Senior Member
 
ShagnWagn's Avatar
 
Join Date: Nov 2007
Posts: 474
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by bjmarchini View Post

I am just wandering if these 80 Plus Certified units are worth it economically.

So that is actually $6 per month. And this is assuming that the efficiency doesn't drop over time. Does it?

Anyone else have any thoughts?

Regardless of how much you use, why wouldn't you pick an efficient power supply over and inefficient one? Using your figure of $6/month, that is still $72 a year. You can almost buy another power supply with just the cost savings in electricity for one year!

You are only thinking of your pc by itself. Think of the extra energy the air conditioner uses to remove that extra "inefficient heat" from inside your home in the middle of summer...

An inefficient power supply also has to remove more heat than an efficient one, which means the fan has to spin faster (louder).
ShagnWagn is offline  
post #6 of 16 Old 12-26-2008, 12:00 PM
AVS Special Member
 
nm88's Avatar
 
Join Date: May 2003
Location: San Francisco, California
Posts: 2,685
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by bjmarchini View Post

I am not so much worried about the environmental impact. The only difference seems to be a slight reduction in kilowatts. I am already green enough in my eyes with a fuel efficient car, the green bulbs and such. I am just looking at it economically.

As just mentioned, there is little reason not to get an efficient power supply. It will generate less heat, be quieter, and be less likely to fail from overheating. It's a no-brainer.

If you really want to save money, use S3 sleep states, unless you are literally using your computer 24 hours a day. I save nearly $35 per month by just putting one desktop system to sleep when not in use.
nm88 is offline  
post #7 of 16 Old 12-26-2008, 02:51 PM
AVS Special Member
 
blue_z's Avatar
 
Join Date: Feb 2006
Location: So Cal
Posts: 1,031
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 18
Quote:
Originally Posted by bjmarchini View Post

If I install 500 Watt PSU that is >80% then I am wasting about 100 watts. If I went with a 600 Watt to compensate and it get lets say >65%, then I would be 210 watts. This is a difference of 110 watts.

Hi there

As CMSR and nm88 already pointed out, this math is all wrong.
A 500 Watt PSU that is 80% efficient will draw 625 watts from the wall outlet (at full load). A 500 Watt PSU that is 65% efficient will draw 769 watts from the wall outlet (at full load). That's a difference of 144 watts that you have to pay the power company.

The efficiency of the PSU drops when less than the full load is delivered. So if you install too big a PSU (e.g. put in a 600W PSU when all the components only use 200W) then you cannot achieve the advertised "efficiency".

Regards

PS the question of using an efficient PSU is moot if you buy one on sale (i.e. remove the price premium) as there have been a few deals the past few days.
blue_z is offline  
post #8 of 16 Old 12-26-2008, 04:00 PM
AVS Special Member
 
K-Spaz's Avatar
 
Join Date: Oct 2008
Location: Central Pennsylvania
Posts: 1,300
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
I think there's some misconceptions above.

First, the efficiency of a PS does not represent some load vs capacity ratio. It represents the amount of the input power from the wall that is available for use by the load.

A PS that is 80% efficient will take 600 Watts of wall AC power, and return to you 480 watts of useable regulated DC. Give or take. The additional watts are eaten up as heat. There is no free lunch. That supply will be rated at 480W. Now, the rating system for cheapo supplies vs good ones is very different. A good supply that's rated at 480w will supply 480w continous. A chinsy one will be rated at 480w and will do that for 5 seconds before it explodes.

Now, take a PS that is not efficient but is rated for high output like 600W, and it will use a lot of power generating heat. At a certain point, it will have a nuclear meltdown, or your system will. I've seen (just once) a PS that exploded in flames. It was in a dark room on a gaming system and it looked really cool. Moreover, it wasn't mine so I especially enjoyed it.

That was on a Athlon FX based system. The PS was replaced with a 80+% supply. The CPU temps on that system dropped 20C after installing the more capable supply. The cost of your electric bill is the least of the advantages to having a more efficient power supply. On a system that is overtaxed, and if you were using a chinsy supply, you will most likely pay for the new supply with savings in electric.

The hotter they get, the more heat they produce. Thus, yes, the less efficient they become as they warm up. I don't think age factors into this much though.

Components in a system with an overkill supply, tend to last a lot longer. Folks who build nice systems with good PS's usually get to throw the machine away in working condition when it's so old it's worthless.
K-Spaz is offline  
post #9 of 16 Old 12-26-2008, 06:21 PM
AVS Addicted Member
 
stanger89's Avatar
 
Join Date: Nov 2002
Location: Marion, IA
Posts: 17,501
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 178 Post(s)
Liked: 158
http://www.silentpcreview.com/article814-page1.html
Or any of their other PSU articles will be a good read...

See what an anamorphoscopic lens can do, see movies the way they were meant to be seen
stanger89 is online now  
post #10 of 16 Old 12-26-2008, 08:07 PM
AVS Special Member
 
blue_z's Avatar
 
Join Date: Feb 2006
Location: So Cal
Posts: 1,031
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 18
Quote:
Originally Posted by K-Spaz View Post

I think there's some misconceptions above.

First, the efficiency of a PS does not represent some load vs capacity ratio. It represents the amount of the input power from the wall that is available for use by the load.


Hi there

Not sure who you think has some misconceptions, but I assure you that I understand PSU efficiency. What I did misstate was the load point of peak efficiency. It's not at full-load as I thought, but more typically at 50-75% of rated power.

http://www.silentpcreview.com/article28-page4.html
There's a graph towards the bottom that illustrates the point I was trying to make: when the Enermax PSU (red line) is electrically loaded to less than half it's rated capacity of 600W, the efficiency is less than the peak. Obviously neither of the graphed PSUs meet the 80Plus specification for efficiency at 20% of rated load. And this point I was making is not as important with an 80Plus PSU, as the article mentions.

Regards
blue_z is offline  
post #11 of 16 Old 12-26-2008, 09:26 PM
Member
 
ps2cho's Avatar
 
Join Date: Dec 2008
Posts: 35
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
The efficiency rating is based on load.

If your HTPC uses lets say 62.5w according to the wall....then "technically" using an 80% efficiency PSU, you are really using 50w. Make sense?
ps2cho is offline  
post #12 of 16 Old 12-26-2008, 09:57 PM
Senior Member
 
hceuterpe's Avatar
 
Join Date: Aug 2007
Posts: 491
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
First, I don't know many people running monster graphics cards in their HTPC. It's stupid, if you're sitting in front of your TV get a PS3 or Xbox 360.

That being said, the biggest power consumption in consumer PCs are in fact graphics cards, especially if you go gluttonous and you run the SLI and/or X2 equivalent. Then you can see pulling 600W. Since you'd be dumb to make a gaming HTPC no big power loads.

Typical PC on full LOAD? Maybe 150W, maybe. Typical HTPC usage? 110W to 120W otherwise. So a little more than a light bulb. Trust me, I own lots of computers I actually run a kill-awatt unit to check, often.

When they talk efficiency, it's how many y watts the PSU takes to generate x watts. (AC-DC power conversion has a loss, but can be done efficiently, your PC components only use DC power). So if you have a PC that needs 120W at 80% efficiency, it's sucking 150W from the wall. Your PSU isn't constantly sucking it's max. Do you have any idea how big a PSU would need to be to dissipate 300W or so watts of heat??? Also, some of the biggest loss of efficiency is not having perfectly clean AC power, which is actually hard and expensive to achieve. A lower efficient PSU "throws" away or filters the bad, the higher efficiency ones correct the bad. Which is why my reasoning later makes sense...

PSUs are all switching power supplies, they only take how much they need.

So in this example you are using 30W wasted, assuming you are doing sleep (assume this is negligible). So, 4 hours a day if you're a total potato, 30 days a month, you're saving about 3.6KWH of power, or about $.50 a month.

That's comparing an an 80% to 100% (impossible), so 20% diff. For about 15%, it is..about $.40 savings a month.

The biggest savings is if you implement sleep intelligently.

Also it also means don't splurge on the PSU (ie spend an extra $30 just for efficiency). The biggest difference mostly is the higher efficiency PSUs are likely to be active PFC and the lower to be passive PFC. The main functional distinguishing factor here is the Passive versions have the 110/220V red toggle switch, while the active PFC versions don't need one. They seem to cost well over $30 for similar wattage.

Also cooling is not a biggy. I actually had a PSU run for months and not noticing the fan died on it. The DC voltages were fine and the PSU wasn't hot. Though running the PSU at full capacity with about 600W would have been a problem. It also was a passive PFC.

K-Spaz: your temps went lower because you probably had more consistent and accuratevoltage. An adequate PSU that can maintain it's voltages generates heat in the PSU itself and blows it straight out. So a more efficient and still adequate PSU wouldn't have made a diff. Also cheap PSUs tend to greatly over estimate their capacities. And they cheat by measuring in amperage. As a PSU (and any power source actually) gets taxed, the voltage drops.
hceuterpe is offline  
post #13 of 16 Old 12-26-2008, 10:08 PM
Senior Member
 
NotSoCoolJ's Avatar
 
Join Date: Jul 2007
Posts: 402
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Efficiency aside, the most important part of your computer is the power supply! I dare say that you won't find a decent PSU with lower than 80% anyway. Get yourself a Corsair PSU and remove all doubt. Bad PSUs tend to take more expensive components with them when they decide to give up the ghost. Don't skimp on the PSU. It's just not worth it.
NotSoCoolJ is offline  
post #14 of 16 Old 12-29-2008, 08:59 AM - Thread Starter
AVS Special Member
 
bjmarchini's Avatar
 
Join Date: Sep 2007
Posts: 3,977
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by hceuterpe View Post

First, I don't know many people running monster graphics cards in their HTPC. It's stupid, if you're sitting in front of your TV get a PS3 or Xbox 360.

That being said, the biggest power consumption in consumer PCs are in fact graphics cards, especially if you go gluttonous and you run the SLI and/or X2 equivalent. Then you can see pulling 600W. Since you'd be dumb to make a gaming HTPC no big power loads.

Typical PC on full LOAD? Maybe 150W, maybe. Typical HTPC usage? 110W to 120W otherwise. So a little more than a light bulb. Trust me, I own lots of computers I actually run a kill-awatt unit to check, often.

When they talk efficiency, it's how many y watts the PSU takes to generate x watts. (AC-DC power conversion has a loss, but can be done efficiently, your PC components only use DC power). So if you have a PC that needs 120W at 80% efficiency, it's sucking 150W from the wall. Your PSU isn't constantly sucking it's max. Do you have any idea how big a PSU would need to be to dissipate 300W or so watts of heat??? Also, some of the biggest loss of efficiency is not having perfectly clean AC power, which is actually hard and expensive to achieve. A lower efficient PSU "throws" away or filters the bad, the higher efficiency ones correct the bad. Which is why my reasoning later makes sense...

PSUs are all switching power supplies, they only take how much they need.

So in this example you are using 30W wasted, assuming you are doing sleep (assume this is negligible). So, 4 hours a day if you're a total potato, 30 days a month, you're saving about 3.6KWH of power, or about $.50 a month.

That's comparing an an 80% to 100% (impossible), so 20% diff. For about 15%, it is..about $.40 savings a month.

The biggest savings is if you implement sleep intelligently.

Also it also means don't splurge on the PSU (ie spend an extra $30 just for efficiency). The biggest difference mostly is the higher efficiency PSUs are likely to be active PFC and the lower to be passive PFC. The main functional distinguishing factor here is the Passive versions have the 110/220V red toggle switch, while the active PFC versions don't need one. They seem to cost well over $30 for similar wattage.

Also cooling is not a biggy. I actually had a PSU run for months and not noticing the fan died on it. The DC voltages were fine and the PSU wasn't hot. Though running the PSU at full capacity with about 600W would have been a problem. It also was a passive PFC.

K-Spaz: your temps went lower because you probably had more consistent and accuratevoltage. An adequate PSU that can maintain it's voltages generates heat in the PSU itself and blows it straight out. So a more efficient and still adequate PSU wouldn't have made a diff. Also cheap PSUs tend to greatly over estimate their capacities. And they cheat by measuring in amperage. As a PSU (and any power source actually) gets taxed, the voltage drops.

That is what I am looking at. For the most part, the higher efficiency models do cost more. It doesn't mean that it is a better PSU. It just means that it is more efficient. Ask your self this? If you could buy PSU A that has an 80% efficiency for $80 rated at lets say 500 watts, a 500 watt PSU rated at 70% for $45, and an 70% efficiency rated at 700 watts for $80, wouldn't the 700 watt one be the most reliable? Granted, most of the time you are going to draw about 120-200 watts. It is like running a v8 on the highway versus an underpowered 4 cylinder. The 8 will be less efficient but won't be worked as hard.

As you stated, I think the energy saving really isn't that much at all. If I am only drawing 120 watts for 18 hours per day, a savings of 20% really isn't going to equate to much.

I have 8 PCs in the house. The HTPC, bedroom desktop (storage system) are usually on 24/7. I have it set up to turn the hard disks and such off after a certain period of idle. The 2 other bedroom PCs I have set up to hibernate when not in use for 3 hours. 3 are laptops. 1 is a mame arcade which is on maybe 10 hours per month.

I am rebuilding the bedroom desktop. It currently is a massively upgraded Dell 4700 (4gigs, 3.4 P4 HT, HD2600XT, 4 gbs / 1.5 TBs). I am pulling the drives and vid card and re building from scratch on this one. Probably go with an AMD set up. It is my only remaining non custom PC aside from the laptops of course. I use it for gaming and it also acts as a media server for the HTPC as well the rest of the house. The HTPC is more of a client/direct player at this point. Heat output is really not a concern as it is a tower case and not a heat oven HTPC case like my ATC-620. May go with a Cooler Master Case again. I have used those 2x now. Once with my Sons and my HTPC. I really like their build quality.

And as far as the gaming point of picking up a 360 or PS3, you must not be a PC gamer. Most PC gamers that I know do not like the console feel that many games have. I am a diehard UT2004 fan and hate UT3 because it is too much of a "console" game. I am not saying console games aren't fun. I play madden on my Wii as well as alot of other games. But for the strategy and first person shooters, I prefer a PC.

HD DVD: 45 SD DVD: 350 BR: 120
PCs: 12, Mame Arcade:1, HD HTPC: 1, WHS Server: 1, HD A3: 2, HD-A30: 1

bjmarchini is offline  
post #15 of 16 Old 12-29-2008, 10:09 AM
Senior Member
 
JoeWPgh's Avatar
 
Join Date: Oct 2007
Posts: 254
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by bjmarchini View Post

...As you stated, I think the energy saving really isn't that much at all. If I am only drawing 120 watts for 18 hours per day, a savings of 20% really isn't going to equate to much.

I have 8 PCs in the house. The HTPC, bedroom desktop (storage system) are usually on 24/7. I have it set up to turn the hard disks and such off after a certain period of idle. The 2 other bedroom PCs I have set up to hibernate when not in use for 3 hours. 3 are laptops. 1 is a mame arcade which is on maybe 10 hours per month.
...

Just a couple quick thoughts: By what you say of your typical usage, I'd pay more attention to the efficiency curves than the actual rating.
Though efficiency savings may be small, once you get several machines going, it does add up.
Does this mean I would replace all the PSUs? Not in a million years. But as I need a PSU, I do pay for efficiency, because I also have several machines running.
JoeWPgh is offline  
post #16 of 16 Old 12-29-2008, 01:04 PM - Thread Starter
AVS Special Member
 
bjmarchini's Avatar
 
Join Date: Sep 2007
Posts: 3,977
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by JoeWPgh View Post

Just a couple quick thoughts: By what you say of your typical usage, I'd pay more attention to the efficiency curves than the actual rating.
Though efficiency savings may be small, once you get several machines going, it does add up.
Does this mean I would replace all the PSUs? Not in a million years. But as I need a PSU, I do pay for efficiency, because I also have several machines running.

That is a good thought and a practical approach. Perhaps, it is best to just keep an eye on the sales and pick one up when available. I have found from experience that it is best to have an extra power supply hanging around anyway.

HD DVD: 45 SD DVD: 350 BR: 120
PCs: 12, Mame Arcade:1, HD HTPC: 1, WHS Server: 1, HD A3: 2, HD-A30: 1

bjmarchini is offline  
Reply Home Theater Computers

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off