• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Is upgrading to more efficient psu really gonna save me money?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

abarth_1200

Member
Joined
Jun 3, 2009
Location
Scotland, Inverness
Hi guys,

Concerning the fact that i will have to be paying my own electric bills now I have bought a house, would I benefit from getting a new PSU? other than the fact that the one I currently own cost me only £30 and doesnt have twin rails for my next upgrade, a shiny new gtx 260
 
Depends on how efficient the psu you have now is. Even if you get a really efficient psu. It will take a long time for the energy savings to offset the cost of the new psu
 
Right, first you have to overcome the cost of the PSU, then you start "saving" money. I cant imagine, unless its ancient PSU with 70% efficiency and moving to a top of the line 85+ efficiency PSU that you would notice anything. Some quick math would help you out. Look at your power bill and the cost per kw. ;)
 
I swapped up from a PC Power & Cooling 610w Silencer to a Silverstone Zeus 1200w. The savings was astronomical as now I have enough wattage to power a small city. Just waiting on a power outage of said city to then make money on the power I supply to them :)

:cool:
 
Well, it depends on how long the part lasts...an 80+ supply is supposed to be guaranteed to be 80% efficient, but if it only lasts a year, then you will not see any savings, but you could potentially see better performance out of a newer power supply, and the more steady the output, the more stable your system will be and maybe allow a slightly higher O'C.
I like overkill when it comes to power supplies... IMO too much is not enough :cool:
 
other than the fact that the one I currently own cost me only £30

Sounds like efficiency might be the least of your concerns. What sort of PSU is it? With PSUs, 'cheap' is generally synonymous with 'tendency to die and maybe kill other stuff'.
 
Maybe it's me... but skimming through I worry we might be dodging the real question here? The right answer isn't always a short one...

Question: Will a more efficient PSU lower your energy bill? Yes.
Question: Will a more efficient PSU pay itself off quickly, or have a short ROI? Maybe - read on to find out how.
Clarification: Will an oversized PSU save me money? No - This is thinking in a skewed manner!

Alright... here's my take on a complete answer!

If you are of the mindset to save some on your utility bill when you are making a PSU purchase (say if/when your current one dies), an intelligent purchase would involve the understanding that there is an efficiency CURVE associated with any given PSU. As a *rough and conservative (see following post)* rule of thumb, running any power supply within the top or bottom third of its capacity is going to result in a significant drop in efficiency. Rest un-assured that any touted efficiency figure on the box is there because of the marketing demons, and will only indicate maximum efficiency but not the corresponding power draw.

The most informed way of making a PSU purchase for an existing system is to get one of those cord+plug gadgets you can use to measure power being used, take a measurement of your idling power and a maximum draw (adding some reasonable cushion for upgrades and murphy's law, to ensure stability). Take these figures and do your homework to compare with efficiency curves from well-documented reviews of PSU's. Find a PSU that sits your idling (or "normal") power draw near the top of the unit's curve, and still has the capacity you want to ensure stability and upgradeability.

Further reading: http://www.anandtech.com/casecoolingpsus/showdoc.aspx?i=3413&p=3

And now to finally answer the question of how to figure out a PSU's ROI. To rephrase: how soon will a new PSU start "saving" me money in the sense of paying for itself?
You will need:
- Your utility rate(s). These will include a Consumption rate (kWh) and potentially a Demand rate (kW) as well. Your utility may additionally have different rates based on the season.
- Your measurements of idle and max power draws for your computer, as described above. Be sure you didn't include your monitor or anything else not powered by your PSU when taking them.
- A general idea of how many hours of the day on average your computer is on, both in "idle" and "max"
- A calculator, unless you're a real bean-counter!

Savings in $/month is: (average power draw worked out between your idle/max measurements and how often you figure you're in either state, *Watts*) x (average hours per day your computer is on, *hrs/day*) x (1/1000, *kW/W) x (30, *days/month*) x (weighted average of your consumption rate(s) through the year, *$/kWh*) x (difference between the new and old efficiencies - use the same weighting you used for 'average power draw' to find efficiencies on the corresponding curves, *unitless*)

Divide that figure into the cost of the new unit and that's how many months it'll take to pay for itself - easy! ;) Note where I said $ you can of course use any currency. Also note that the demand rates on your utility bill won't play into this if you keep using the same computer.
 
Last edited:
Great read. However looking at jonnyguru reviews on most name brand quality PSUs after about 140W+ are in the 80%+ efficiency range. If that "1/3" statement came from Anand, they debunked their own statement with their example PSU. You can see by the graph linked its already in the 80%+ range by ~140W. A mere 15% output. That grows to 85% at 200W, less than 25%. Most any quad core with a decent GT200/HD4850 and above will idle in that ballpark.

A quick glance at the efficiency of all the PSU's (aside from besttec LOL) on the main PSU review page at jonnyguru shows that even at 140W+ idle loads (about normal for a midrange gaming system) shows 80% efficiency+ on the lowest tests across all PSUs..
 
Last edited:
Great read. However looking at jonnyguru reviews on most name brand quality PSUs after about 140W+ are in the 80%+ efficiency range. If that "1/3" statement came from Anand, they debunked their own statement with their example PSU. You can see by the graph linked its already in the 80%+ range by ~140W. A mere 15% output. That grows to 85% at 200W, less than 25%. Most any quad core with a decent GT200/HD4850 and above will idle in that ballpark.

A quick glance at the efficiency of all the PSU's (aside from besttec LOL) on the main PSU review page at jonnyguru shows that even at 140W+ idle loads (about normal for a midrange gaming system) shows 80% efficiency+ on the lowest tests across all PSUs..

the "1/3 rule" is a rough rule of thumb based on my experience, not from anywhere else ;). I didn't myself realize the curves in the computer PSU world are currently consistently rising at the 140W mark regardless of capacity (it's been a while since I went PSU shopping!). In my defense, if my strategy of staying in the "middle third" is followed, you should still be sitting pretty, if a bit conservative =).

I've edited my little tirade above to reference your post for more info =).
 
Last edited:
i dont think your worrying about the wrong things....that psu is probably about to blow lol you need a better psu asap....

and what are your system specs?
 
Some extra info... I was asked by a friend who read my post above about the "cord+plug gadgets" i mentioned.

Here is an example of one - lately they've been popping up in hardware and big-box stores around my area:
http://www.amazon.com/P3-International-P4400-Electricity-Monitor/dp/B00009MDBU

They seem to cost around the $20 range. I'm pretty confident the average homeowner who has the interest to use one will immediately have this thing pay for itself as you realize how unexpectedly wasteful certain appliances can be... it will also allow you to make all kinds of appliance purchases in the future with an understanding of what significance to put into energy efficiency premiums... i.e. you'll be able to more easily recognize that the dryer model that costs $100 more to buy in the store might actually cost you about $50 less per month in your utility bill, which quickly pays for the premium!

It's not insignificant stuff ;)
 
those killawatt units are awesome i have one of them the only thing i dont like about them is that it blocks out both power plugs on the wall...
 
Some extra info... I was asked by a friend who read my post above about the "cord+plug gadgets" i mentioned.

Here is an example of one - lately they've been popping up in hardware and big-box stores around my area:
http://www.amazon.com/P3-International-P4400-Electricity-Monitor/dp/B00009MDBU

They seem to cost around the $20 range. I'm pretty confident the average homeowner who has the interest to use one will immediately have this thing pay for itself as you realize how unexpectedly wasteful certain appliances can be... it will also allow you to make all kinds of appliance purchases in the future with an understanding of what significance to put into energy efficiency premiums... i.e. you'll be able to more easily recognize that the dryer model that costs $100 more to buy in the store might actually cost you about $50 less per month in your utility bill, which quickly pays for the premium!

It's not insignificant stuff ;)
$50 /month on one item? No chance. My electric in the summer with AC running constantly is $250. You think any one appliance is going to save me 20% on my bills? Im sorry, but that is impossible! Drop the zero off that $50 and you are likely in the ballpark. Or maybe $10 /month.

Its not insignificant certainly, but its nowhere near what you make it out to be with one appliance. I could see if you replaced ALL the appliances and get $50 /month or had appliances from 1970 or some rare case like that.

I see your overall point and it certainly stands, just not in the monetary savings range for XX item you thought initially. :)
 
OK so a new PSU first. Now i need to work out how much my rig draws and what it will draw in the future...

Yeah I havent seen any of those in any shops around here but have heard of them, I havent yet moved in so I dont have any bills to show my kWh's.

I think Ill try get a hold of a killiwatt reader and take readings of my current psu.

Its occuring to me that all I really want is a new solid PSU not long term savings lol
 
$50 /month on one item? No chance. My electric in the summer with AC running constantly is $250. You think any one appliance is going to save me 20% on my bills? Im sorry, but that is impossible! Drop the zero off that $50 and you are likely in the ballpark. Or maybe $10 /month.

Its not insignificant certainly, but its nowhere near what you make it out to be with one appliance. I could see if you replaced ALL the appliances and get $50 /month or had appliances from 1970 or some rare case like that.

I see your overall point and it certainly stands, just not in the monetary savings range for XX item you thought initially. :)

You're missing the point. And it is that significant. Let's say you conservatively trim your averaged $250 bill to $240 - $10 just like you said - with some simple practices and better awareness of your energy uses. This $20 device starts "making you money" after 2 months. Payed itself off.

Another way to look at it is - how long have you been paying your utility bills without such knowlege? multiply those months as well by $10 and you see how much you have been wasting (in my case many years). No small amount!

More drastic energy bill cuts are definitely within economical reason for most homeowners if and when you get around to making purchasing decisions for new/replacement appliances - particularly dryers, refrigerators and HVAC equipment.

Testimonials abound - in my case I have dropped my energy bills from about $180 for the summer to $120. I credit primarily my purchase and use of a programmable thermostat along with the knowledge and resulting appliance purchasing decisions that a device such as the kill a watt can help teach people.
 
No, I got it. I just didnt agree with the inflated numbersr to prove a point. You are correct though, every little bit helps. But buying one appliance like a dryer isnt going to save that much per month.

All good. Your point is correct just the example needed tweaked in my OCD like head!

But you take that $20 device and buy a $300 unit and you have years to pay it off through efficiency savings. But we are nitpicking at this point. You are absolutely right. I said that in that post as well............:)
 
No, I got it. I just didnt agree with the inflated numbersr to prove a point. You are correct though, every little bit helps. But buying one appliance like a dryer isnt going to save that much per month.

All good. Your point is correct just the example needed tweaked in my OCD like head!

But you take that $20 device and buy a $300 unit and you have years to pay it off through efficiency savings. But we are nitpicking at this point. You are absolutely right. I said that in that post as well............:)

Sorry if I came off like a porcupine =)!

Rather than dabble in imaginary numbers, I'm guessing it'd be more productive for everyone to examine the following, published by the US D.O.E. this last Janurary:
pg26_appliance_use_chart.gif


National US average utility rates of 2008 are being used to figure out costs shown, along with "standard" efficiencies for each appliance.

I will stubbornly stand by my real-life example of dropping my summer monthly energy bills by more than $50 within a year's time though!

Apologies to abarth for taking this thread on an obvious tangent - I hope you feel your question was answered!
 
Not at all chix..dont sweat it. Your thoughts are spot on, but ONE appliance will not drop a bill by $50 /month. Replacing all the appliances, sure. Not one dryer though. I also apologize for the assiting in derailing the thread and hope the question was answered. :)
 
Back