PSU Guide

[MohawkAngel]

One thing I can add afterward is that the psu don't have PFC but like cheaper CoolerMaster. eXtreme Power 500w does not have. But at least the fan is almost dead silent and rpm goes up as demand rise. So for a small all onboard htpc its enough.

 

[cdawall]

From what I was reading with corsair they are using cwt for the low end stuff same as thermaltake...not exactly bad bud not exactly seasonic either.

 

[westom]

EDIT: This test was made by sucking the 19.6 amps on a single red wires from a molex plug.

Maximum current of that Molex contact is only 6 amps. That would explain the heat and low readings.

 

[MohawkAngel]

Maximum current of that Molex contact is only 6 amps. That would explain the heat and low readings.

The heat yes but its not low reading the maximum is supposed to be 14amps on the +5v line and I pushed it to 19.6amps !!!

 

[westom]

The heat yes but its not low reading the maximum is supposed to be 14amps on the +5v line and I pushed it to 19.6amps !!!

Of course. But what was the current when the voltage remained well in spec? Does not matter how much current it can output if the voltage during that current is less than 4.87 VDC (on the meter).

A relevant term is "foldback current limiting".

Normal and necessary for any acceptable supply is to output more current than the spec number. Why so many supplies that are insufficient still boot computers. Also why a defective supply can also boot a computer. And why that one who thinks a supply is good only because the computer boots is only deceiving himself. You were using the meter. What informed computer techs always use before knowing.

But again, the point. That molex connector (and therefore associated wire) is only for 6 amps which explains tremendous heat.

 

[MohawkAngel]

I use multimeter since im 13 yo and im now 30. Its my toy I love using it as soon as I can lol Of course I dont use the same that i used at 13 but anyway i'm loving it I not tested the voltage while testing current because i had only one tester and at same time if i had another one i could have test the voltage drop at another molex that is still at room temps and the voltage at the overheating plug to see how much voltage the resistance due to heat made me loss.

 

[westom]

I use multimeter since im 13 yo and im now 30. Its my toy I love using it as soon as I can lol

Actual operation of each output is best defined by a V-I (voltage verses current) curve. In the attached figure, a black line represents an ideal power supply. Voltage stays constant until current reaches maximum. Then current remains constant for a falling voltage. The green line is a more realistic curve when a power supply is also limited by power. And finally, the red line is how we really want a power supply to perform.

To understand V-I characteristics of any supply is best accomplished with two multimeters so that one can read voltage while another reads current. Click on this image to better view it.

 

[MohawkAngel]

Actual operation of each output is best defined by a V-I (voltage verses current) curve. In the attached figure, a black line represents an ideal power supply. Voltage stays constant until current reaches maximum. Then current remains constant for a falling voltage. The green line is a more realistic curve when a power supply is also limited by power. And finally, the red line is how we really want a power supply to perform.

To understand V-I characteristics of any supply is best accomplished with two multimeters so that one can read voltage while another reads current. Click on this image to better view it.

It is OK I told you already so come on with explanations. I told ya I understand what it means and how it works. I'm used to work in electric and electronic and if it become more difficult I just ask to my brother who's an engineer at Research and Development department for Honeywell Electronics. I have everything I need as advice when I need to but if you want to start a How-To thread feel free to od it it's always helpful for people who don't have the knowledge that I have.

 

[MohawkAngel]

So Panchoman are you gonna write a complaint to the UL regulation agency to tell them that cooler master are selling psu's without UL code?

 

[cdawall]

So Panchoman are you gonna write a complaint to the UL regulation agency to tell them that cooler master are selling psu's without UL code?

they dont have to have a ul code to be sold it is up to you to purchase a ul listed powersupply

 

[MohawkAngel]

Strange because I tought it was the norm and law to have the UL code for fire safety. Here in Quebec if something electrical don't have the UL code it's removed from shelves and sent back to supplier.

 

[westom]

It is OK I told you already so come on with explanations. I told ya I understand what it means and how it works. .

Why are you being emotional? My posts agree with and reinforce what you have posted. And provide facts because everything I post is for everyone - a number far larger than just you.

I don't care if you are world's premier expert on all electronics. I am still going to post information because you are only one of maybe a thousand readers.

Your emotions puzzle me. Suggest you do not grasp the relevance of what was posted. Or think everything here is posted only for you. Anyone who would rather be an informed consumer should view that previously posted V-I graph.

UL is not required for anything. UL is an information source similar to what Consumer Reports and Energy Star also do. No power supply needs a UL (or CSA) listing. Because safety of any computer (and meeting all other standards) is 100% on the assembler. UL says nothing about a power supply even working properly.

Only an assembler is responsible for what is inside a supply that he selects. Explains why inferior supplies are so easily dumped in this market. Supply knowledge also means one knows those V-I curves.

A more reasonable response would be to agree - to encourage others to also become informed.

 

[cdawall]

Strange because I tought it was the norm and law to have the UL code for fire safety. Here in Quebec if something electrical don't have the UL code it's removed from shelves and sent back to supplier.

all of the cheapy PSU's are none UL listed shit jonny guru has tested a couple dozen non-UL PSU's

 

[MohawkAngel]

all of the cheapy PSU's are none UL listed shit jonny guru has tested a couple dozen non-UL PSU's

I'll have to take a look then because like I told ya seeing them around here is not the norm. If you have a link or two to the guru site ill go take a look faster.

 

[erixx]

What do you guys think of those PEAKS? (Nearly 15V!!!!)

My PSU is a Corsair HX750.

As you can see, the pic is from a stress test. And after the voltage peak, the processor hung for a sec. I was browsing TPU and the broser also frooze for a sec.

Mamma mia, never thought I would see this! Is this very bad?

Thx!

 

[cdawall]

http://img.techpowerup.org/110130/peak.png

What do you guys think of those PEAKS? (Nearly 15V!!!!)

My PSU is a Corsair HX750.

As you can see, the pic is from a stress test. And after the voltage peak, the processor hung for a sec. I was browsing TPU and the broser also frooze for a sec.

Mamma mia, never thought I would see this! Is this very bad?

Thx!

The is voltage reg on the corsairs that would shut the psu off if it was @15v the dumb monitor program is wrong. Get a meter out and check it yourself and it will show exactly that

 

[erixx]

Thanks CDAWALL, but I have never seen that peak before and I doubt that I will see it again, ever, it is totally random. (I use Aida/Everest since years and never came across such a peak.) But I am curious about the reason and the meaning of it.

 

[westom]

What do you guys think of those PEAKS? (Nearly 15V!!!!)

Important facts are missing. Including what hardware made those measurements, horizontal and vertical resolutions, and the horizontal time scale.

Speculating only from what is provided, I would start with a defective or problematic ground for that measuring hardware. A suspicion based in what voltages vary and by how much. If those voltage spikes existed, functions inside a power supply would have cut off power to each voltage (3.3, 5, and 12). Then other functions would have locked out the CPU and other motherboard functions. Obviously those safety features never triggered.

Numbers make no sense. So questions start by learning why measurement hardware has a fault.

Appreciate how much information was in those numbers. Most would not realize how much information was contained.

 

[westom]

I use Aida/Everest ...

That is software? Software never measures voltages. Software only reads data provided by hardware. Noted in the previous post, address how hardware measures. Not how software translated those numbers into a human readable form.

 

[erixx]

Hardware: the mobo only
Resolution, vert and horizontal: no idea
Time scale: no idea, but the peak was '1 unit' of update time= 3 sec max

 

[cdawall]

Well if they were peaking around 15v you would be seeing fried parts

 

[westom]

Hardware: the mobo only

OK. Now let's discuss how computers work. That motherboard contains a large plate of copper end to end across the board. Voltage across that 'plane' can vary significantly. Many assume voltage well be the same zero volts everywhere on that ground plane. It is not. That is also why ceramic bypass capacitors are scattered across the motherboard. And why more reliable computers connect the motherboard to the chassis at only one point.

And to calibrate hardware on that motherboard. The motherboard monitor is just that. A monitor. Apparently it detected some difference on the motherboard (up to less than 3 volts) ground plane that probably should not exist. And that only occurs when certain semiconductors suddenly demand power.

Microsecond pulses could be continous. But captured rarely by a motherboard monitor only being read by software on a 3 second resolution. An example of not only seeing something - also know the whats and whys.

As another so accurately posted, get a multimeter to make useful measurements. And welcome to the #1 reason why we fix things. To learn.

 

[westom]

Well if they were peaking around 15v you would be seeing fried parts

15 volts on 12 volt parts will not harm them. 7 volts also does not harm 5 volt parts. But safety functions required in a supply (and sometimes missing on supplies selling on price) must keep voltages lower.

 

[Swamp Monster]

15 volts on 12 volt parts will not harm them. 7 volts also does not harm 5 volt parts. But safety functions required in a supply (and sometimes missing on supplies selling on price) must keep voltages lower.

+1 My PSU will cut output when 15,6V is on 12V; 7V is on 5V and 4.5V is on 3V rail. But under normal operation it regulates voltage +3/-3% on all rails. It should be similar on all other PSU's, varying only percentage and voltage a little. I know that on some PSU's they have 5% regulation.

 

[erixx]

I have a hard time to understand, but appreciate the information! Stil don't know 'what has happened' and if I have to do something. (I don't have know-how of multimeters, sorry)