How does one test efficiency rating on a power supply? As a consumer just being curious?

[Frick]

Don't listen to the luddites that say everything should be left to professionals; a professional is just a guy who read a book and passed a test, you can read the book for the same knowledge.
The most dangerous guy I ever met was a freshly graduated Electrical Engineer; he destroyed most of the lab within a week, lol.
They don't teach anything practical in School these days, just theory.

There are boneheads everywhere, that's for sure. But ... well, that's what schools and tests and certificates are for: to sort out the people that shouldn't do stuff and the people that officially can. And also it's a liability thing. Personally I try to tell people to hire professionals at all times because a) depending on the work it can be illegal to not be certified (well "illegal" is probably not exactly the correct word in this context), and b) I have no idea what people can do and what they think they can do and as you point out there are so many whackos who think it's an excellent idea to use nails for breakers, and finally c) insurance. I'd be able to do a good smaller electrical installation (because one of my best friends and former boss educates electricians and troubleshoot weird electrical problems in mostly industries no one else can figure out and he had me proofread his books) but it'd be "illegal" and the insurance wouldn't cover it. That actually is the point: to create an official system of liability. Also good teachers have experience and may be able to instill some of that into the students. Some people are blessed with insane technical talents (I know at least one of those), but for most people it's all about skills learned.

Tangential, but I find it an interersting topic. My father (freestyle electrician, plumber, carpenter, farmer, hunter, semi-hippie and all around fixer) has always said "Those rules are for people who don't know what they're doing" and I can see his point, but I also really, really disagree with him.

But I digress.

 

[Space Lynx]

On-topic, thanks for your input everyone, it has been very helpful. I currently have a Titianium PSU on order, going to be interesting to try to compare it to my Bronze rated Corsair which is like 7 years old.

 

[EarthDog]

I currently have a Titianium PSU on order, going to be interesting to try to compare it to my Bronze rated Corsair which is like 7 years old.

What are you trying to compare it to? You don't have the tools to do so, really...

 

[Space Lynx]

What are you trying to compare it to? You don't have the tools to do so, really...

wattage draw mainly. I want to achieve 50% load on both ( have a lot of old parts laying around, and hard drives, so will add and subtract drives until I reach 50% for each drive, and the wattage wall reader will also help me calculate cost of electricity. I want to see how much titianium actually will save over bronze at 50% load with my electricity bill, it will take some calculations, I could guess work it, but I want real world scenario, plus I been wanting a titanium drive for a long time now since it basically will last me a life time anyway, might as well spend a little extra on one.

 

[Bill_Bright]

I said AT, I meant AT.

Fair enough. That is why I asked for clarification, after all. But how do 25 year old, obsolete technology AT power supplies that won't support electronically, or physically any of today's motherboards, or physically mount in any of today's ATX cases (due to the AT's wiring harness) help the OP?

Don't listen to the luddites...

That's sounding like better advice all along!

 

[EarthDog]

wattage draw mainly. I want to achieve 50% load on both ( have a lot of old parts laying around, and hard drives, so will add and subtract drives until I reach 50% for each drive, and the wattage wall reader will also help me calculate cost of electricity. I want to see how much titianium actually will save over bronze at 50% load with my electricity bill, it will take some calculations, I could guess work it, but I want real world scenario, plus I been wanting a titanium drive for a long time now since it basically will last me a life time anyway, might as well spend a little extra on one.

all you need to do is look at the 80 plus chart dude....

~9% is going to be the difference between bronze and titanium at 50% load. 85% vs 94%. Obviously it will be a bit different as the spec is a minimum, but you'll get the idea.

It's all there for you without spending that time and effort to confirm spec.

Unless you F@H or have hella expensive electricity, you'll never make up the cost titanium over a gold over the life of the psu.

 

[Space Lynx]

all you need to do is look at the 80 plus chart dude....

~9% is going to be the difference between bronze and titanium at 50% load. 85% vs 94%. Obviously it will be a bit different as the spec is a minimum, but you'll get the idea.

It's all there for you without spending that time and effort to confirm spec.

Unless you F@H or have hella expensive electricity, you'll never make up the cost titanium over a gold over the life of the psu.

Aye, I suppose there is no point, not sure what I was thinking. Just bored as crap I guess.

 

[moproblems99]

all you need to do is look at the 80 plus chart dude....

~9% is going to be the difference between bronze and titanium at 50% load. 85% vs 94%. Obviously it will be a bit different as the spec is a minimum, but you'll get the idea.

It's all there for you without spending that time and effort to confirm spec.

Unless you F@H or have hella expensive electricity, you'll never make up the cost titanium over a gold over the life of the psu.

I have always wanted to do it for science but without the dedicated hardware but it is just too dangerous to accidentally slip and touch something you weren't supposed to. I have Fluke amp probes and multimeters so I don't have any extra expenses. I think my smallest amp probe is for 30amps which would probably work but I might have to build a multiplier for some of the other circuits. Not sure off the top of my head how much the 3.3 lines actually draw.

 

[Space Lynx]

I have always wanted to do it for science but without the dedicated hardware but it is just too dangerous to accidentally slip and touch something you weren't supposed to. I have Fluke amp probes and multimeters so I don't have any extra expenses. I think my smallest amp probe is for 30amps which would probably work but I might have to build a multiplier for some of the other circuits. Not sure off the top of my head how much the 3.3 lines actually draw.

I actually had not thought of all the dangers involved, you are correct, it is a bit dangerous even if you do wear protective gear, ground yourself etc. Sometimes accidents just happen. Heh, it's probably a good thing I don't have the equipment for it, thanks for saving my life with this post lol

 

[Deathy]

Power factor correction is more likely to make a measurable difference in your power bill.

Where do you live that you pay your bill by apparent power instead of actual power? I don't, in Germany. Unless you are just talking about huge businesses.

 

[Grog6]

You aren't going to hurt yourself, as long as you don't open the PS box.

5V and 12V won't shock you with dry skin, and if you're digging around in your computer with wet hands, it's on you, lol.

The only time I managed to shock myself with 12V was being sweaty in shorts, sitting on a cars metal doorframe, reaching up into the dash to replace a fuse.
Made my butt tingle for sure.

DC also isn't like AC; it doesn't cause fibrillation of the heart, it more like the paddles they hit you with to start it back up.

I've gotten a 3kV 10mA PMT power supply from hand to hand, and it won't kill you, but it will body slam your ass.
I got the same power supply across one hand, when it was damaged, and was over 19999V; I still have two small scars from that one. If the display is flashing on a TC952, unplug it, don't reach for it.

Everyone in our group got one of those 1kV to 3kV hits eventually; it teaches the "one hand rule" very effectively.
(If working with anything over 12V, put one hand in your back pocket. This reduces problems to almost zero. Standing with your crotch pressed up to a metal table, while reaching for the HV wire is a different problem. I saw that, I knew better, lol.)

Where do you live that you pay your bill by apparent power instead of actual power? I don't, in Germany. Unless you are just talking about huge businesses.

It's not billed like that residentially except with some newer power meters; they get pissed if you let power factor go too low; too much current for the lines.
My power bill now has a power factor number; I can tell when I have the old servers all on.
I'm paying from $0.12-0.16/kwH for power, depending.
Actually, looking at this bill, I dan see when I've been doing data runs, and it's about $25 a month extra to run 4x vintage 1991 IBM EISA servers with SCSI raid arrays, lol.

The business example I mentioned above was causing problems.
We got a $2000 a month fine, and had been for years, until they added one more furnace which burnt some switchgear.
We were drawing over 3MW. Large induction furnaces running 24/7, with ~2 hours downtime every few days.
That's the only time I ever had to deal with 480 three phase; I prefer benchtop power supplies.
(Remember ELI the ICE man (A mnemonic for phase in AC power)? In capacitive circuits current leads voltage, poor power factor is based on how far they differ from optimum.)

 

[Bill_Bright]

but it is just too dangerous to accidentally slip and touch something you weren't supposed to.

If you don't expose the mains (wall outlet) contacts/conductors, and you don't stick probes inside the PSU itself where those deadly voltages exist,(and there's no need to be inside there either), you will only be dealing with low DC voltages,12VDC at most. It would actually be a challenge to accidentally cross that 12V to something it could destroy.

With just a little care (and long PSU cables), virtually all measurements can be done outside the computer case, away from sensitive components and contacts on the motherboard. In fact, with a proper test rig, all testing can be done away from the motherboard completely. So actually, the bigger risk is probably accidentally digging a trench across several motherboard circuit traces should you accidentally slip with the sharp meter probe. And should you find yourself working near the motherboard, ESD may be a greater risk too, if the necessary precautions are not observed.

Still, as noted, accidents do happen. And since most PSU makers are [more or less] honest, and they know there are well equipped experts willing to test, evaluate, verify stated specifications and publish those results, I think it safe to trust if a maker is willing to pay Plug Load Solutions to test and certify their supplies (and pay they must) as 80 PLUS certified, then they are going to make sure efficiency values do indeed meet or exceed those certification minimums - or risk W1zzard's wrath.

 

[moproblems99]

If you don't expose the mains (wall outlet) contacts/conductors, and you don't stick probes inside the PSU itself where those deadly voltages exist,(and there's no need to be inside there either), you will only be dealing with low DC voltages,12VDC at most. It would actually be a challenge to accidentally cross that 12V to something it could destroy.

I didn't mean dangerous to me. A little bit of 120V never hurt nobody.

I meant dangerous to the PC. Working inside a cramped environment with large (sometimes) pokey probes or clamp on meters can mean one slip of the hand sends a tool into something it wasn't meant for. Since you have to have the machine on, running and loaded to test, it is 'dangerous' to the machine to be back probing connections, etc.

 

[Grog6]

I would make a test fixture, not use the pc for load testing; measuring the current introduces unknowns, unless you have multiple hall effect probes.

If you short or connect the 12V to the 5V, you'll be shopping for a new pc.

For measuring voltages, I carefully solder wire wrap wire to pads on the motherboard while it's off, and run them to a connector. It goes to a matching connector with dual banana plugs, so they can't short easily.

That cuts down on accidents.

And wire wrap wire is small enough, a computer power supply will make it disappear if you short it somehow; while not affecting the voltage measurement with a modern 10M ohm impedance meter.

 

[FinneousPJ]

wattage draw mainly. I want to achieve 50% load on both ( have a lot of old parts laying around, and hard drives, so will add and subtract drives until I reach 50% for each drive, and the wattage wall reader will also help me calculate cost of electricity. I want to see how much titianium actually will save over bronze at 50% load with my electricity bill, it will take some calculations, I could guess work it, but I want real world scenario, plus I been wanting a titanium drive for a long time now since it basically will last me a life time anyway, might as well spend a little extra on one.

It seems like you still don't understand. If you don't already know the efficiency, you cannot know when the unit is at 50% load without special equipment. The reading at the wall is

reading = load / efficiency.

If you're just going to assume you know the efficiency in order to calculate the load, what's the point of trying to measure efficiency.

 

[Space Lynx]

It seems like you still don't understand. If you don't already know the efficiency, you cannot know when the unit is at 50% load without special equipment. The reading at the wall is

reading = load / efficiency.

If you're just going to assume you know the efficiency in order to calculate the load, what's the point of trying to measure efficiency.

I already admitted I messed up when i was responding to EarthDog. your post was unnecessary.

 

[FinneousPJ]

I already admitted I messed up when i was responding to EarthDog. your post was unnecessary.

He didn't make the same point as I did.

 

[Frick]

(If working with anything over 12V, put one hand in your back pocket.

And if working on stuff closer to the ground, crouch instead of sitting on a chair. I've no idea if that is how the body works, but that was what dad taught me. (the muscles in the legs strech out when shocked sufficiently so you'd be thrown backwards instead of being frozen in place if sitting on something)

 

[Bill_Bright]

And if working on stuff closer to the ground, crouch instead of sitting on a chair.

That one I have never heard of. But if the chair is made of metal, that could make sense. But then we always had metal bench stools to sit on - however, we also always had rubber floor matting too.

 

[sneekypeet]

This thread is not about your ignore lists. Last time I am mass cleaning this thread without handing out points for driving this thread off topic.

 

[crmaris]

actually, this is my old set-up. Need to change this picture at some point

To measure efficiency accurately you need a good power meter (those go over 1k) and a steady and variable load. This is where we, reviewers, come in. We do that to save you from all the trouble and the expenses