Problems with 12V and 5V lines on a PC ATX supply

On Thu, 20 Feb 2020 03:43:58 -0000, default <default@defaulter.net> wrote:

On Thu, 20 Feb 2020 00:32:05 -0000, "Commander Kinsey"
CFKinsey@military.org.jp> wrote:

On Wed, 19 Feb 2020 23:31:01 -0000, Jasen Betts <jasen@xnet.co.nz> wrote:

On 2020-02-19, Commander Kinsey <CFKinsey@military.org.jp> wrote:
Why do (cheap? expensive ones may be better) PC ATX power supplies need current drawn from the 5V line to make the 12V line work correctly?

I have a PC with 3 graphics cards running scientific applications. I acquired three old graphics cards that take about 300W each, and have loads of cheap (CIT) PSUs that are rated at 650W on the 12V line, which is what those cards use. So I run each card off its own supply. But the 12V line at no load, or even at 300W, is only giving out 10 to 10.5V. If I attach a small dummy load of an amp or so to the 5V line, the 12V line suddenly becomes 12V.

Why are the two lines related in any way?

because all the output voltages come from taps on the same transformer
and the voltage regulation is applied to the input to that transformer
and the voltage regulation only watches the 5V line.

Ok, but why does current need to be taken from 5V to make the voltage monitor work?

It is designed to be in a computer, and there's always some load on
the 5 V line. It probably didn't seem terribly important to worry
about a high voltage condition where none should ever exist.

What load would that be? In my newest computer for example (which has a decent supply which doesn't need the 5V load), it uses an SSD which draws fuck all power. The CPU and graphics card and memory all operate from 12V with their own regulator modules. I can't think of anything drawing much 5V, and the dodgy supplies I have need approaching 2 amps! I tried 1 amp and that didn't completely stabilize the 12V line.

Some power supplies sit and oscillate if they don't have a load on the
5 volt line...

I notice my desktop has 5v present on the USB connector even when it
is turned off, turned on, or just in standby. I suspect it may have a
small independent supply to run the USB connectors for power, and
perhaps that also supplies the CMOS memory so the clock and settings
don't drain the battery.
 
On Thu, 20 Feb 2020 20:07:47 -0000, "Commander Kinsey"
<CFKinsey@military.org.jp> wrote:

On Thu, 20 Feb 2020 03:43:58 -0000, default <default@defaulter.net> wrote:

On Thu, 20 Feb 2020 00:32:05 -0000, "Commander Kinsey"
CFKinsey@military.org.jp> wrote:

On Wed, 19 Feb 2020 23:31:01 -0000, Jasen Betts <jasen@xnet.co.nz> wrote:

On 2020-02-19, Commander Kinsey <CFKinsey@military.org.jp> wrote:
Why do (cheap? expensive ones may be better) PC ATX power supplies need current drawn from the 5V line to make the 12V line work correctly?

I have a PC with 3 graphics cards running scientific applications. I acquired three old graphics cards that take about 300W each, and have loads of cheap (CIT) PSUs that are rated at 650W on the 12V line, which is what those cards use. So I run each card off its own supply. But the 12V line at no load, or even at 300W, is only giving out 10 to 10.5V. If I attach a small dummy load of an amp or so to the 5V line, the 12V line suddenly becomes 12V.

Why are the two lines related in any way?

because all the output voltages come from taps on the same transformer
and the voltage regulation is applied to the input to that transformer
and the voltage regulation only watches the 5V line.

Ok, but why does current need to be taken from 5V to make the voltage monitor work?

It is designed to be in a computer, and there's always some load on
the 5 V line. It probably didn't seem terribly important to worry
about a high voltage condition where none should ever exist.

What load would that be? In my newest computer for example (which has a decent supply which doesn't need the 5V load), it uses an SSD which draws fuck all power. The CPU and graphics card and memory all operate from 12V with their own regulator modules. I can't think of anything drawing much 5V, and the dodgy supplies I have need approaching 2 amps! I tried 1 amp and that didn't completely stabilize the 12V line.

Some power supplies sit and oscillate if they don't have a load on the
5 volt line...

I notice my desktop has 5v present on the USB connector even when it
is turned off, turned on, or just in standby. I suspect it may have a
small independent supply to run the USB connectors for power, and
perhaps that also supplies the CMOS memory so the clock and settings
don't drain the battery.
 
On Thu, 20 Feb 2020 21:06:22 -0000, default <default@defaulter.net> wrote:

On Thu, 20 Feb 2020 20:07:47 -0000, "Commander Kinsey"
CFKinsey@military.org.jp> wrote:

On Thu, 20 Feb 2020 03:43:58 -0000, default <default@defaulter.net> wrote:

On Thu, 20 Feb 2020 00:32:05 -0000, "Commander Kinsey"
CFKinsey@military.org.jp> wrote:

On Wed, 19 Feb 2020 23:31:01 -0000, Jasen Betts <jasen@xnet.co.nz> wrote:

On 2020-02-19, Commander Kinsey <CFKinsey@military.org.jp> wrote:
Why do (cheap? expensive ones may be better) PC ATX power supplies need current drawn from the 5V line to make the 12V line work correctly?

I have a PC with 3 graphics cards running scientific applications. I acquired three old graphics cards that take about 300W each, and have loads of cheap (CIT) PSUs that are rated at 650W on the 12V line, which is what those cards use. So I run each card off its own supply. But the 12V line at no load, or even at 300W, is only giving out 10 to 10.5V. If I attach a small dummy load of an amp or so to the 5V line, the 12V line suddenly becomes 12V.

Why are the two lines related in any way?

because all the output voltages come from taps on the same transformer
and the voltage regulation is applied to the input to that transformer
and the voltage regulation only watches the 5V line.

Ok, but why does current need to be taken from 5V to make the voltage monitor work?

It is designed to be in a computer, and there's always some load on
the 5 V line. It probably didn't seem terribly important to worry
about a high voltage condition where none should ever exist.

What load would that be? In my newest computer for example (which has a decent supply which doesn't need the 5V load), it uses an SSD which draws fuck all power. The CPU and graphics card and memory all operate from 12V with their own regulator modules. I can't think of anything drawing much 5V, and the dodgy supplies I have need approaching 2 amps! I tried 1 amp and that didn't completely stabilize the 12V line.

Some power supplies sit and oscillate if they don't have a load on the
5 volt line...

I notice my desktop has 5v present on the USB connector even when it
is turned off, turned on, or just in standby. I suspect it may have a
small independent supply to run the USB connectors for power, and
perhaps that also supplies the CMOS memory so the clock and settings
don't drain the battery.
 
On Thu, 20 Feb 2020 20:07:47 -0000, "Commander Kinsey"
<CFKinsey@military.org.jp> wrote:

On Thu, 20 Feb 2020 03:43:58 -0000, default <default@defaulter.net> wrote:

On Thu, 20 Feb 2020 00:32:05 -0000, "Commander Kinsey"
CFKinsey@military.org.jp> wrote:

On Wed, 19 Feb 2020 23:31:01 -0000, Jasen Betts <jasen@xnet.co.nz> wrote:

On 2020-02-19, Commander Kinsey <CFKinsey@military.org.jp> wrote:
Why do (cheap? expensive ones may be better) PC ATX power supplies need current drawn from the 5V line to make the 12V line work correctly?

I have a PC with 3 graphics cards running scientific applications. I acquired three old graphics cards that take about 300W each, and have loads of cheap (CIT) PSUs that are rated at 650W on the 12V line, which is what those cards use. So I run each card off its own supply. But the 12V line at no load, or even at 300W, is only giving out 10 to 10.5V. If I attach a small dummy load of an amp or so to the 5V line, the 12V line suddenly becomes 12V.

Why are the two lines related in any way?

because all the output voltages come from taps on the same transformer
and the voltage regulation is applied to the input to that transformer
and the voltage regulation only watches the 5V line.

Ok, but why does current need to be taken from 5V to make the voltage monitor work?

It is designed to be in a computer, and there's always some load on
the 5 V line. It probably didn't seem terribly important to worry
about a high voltage condition where none should ever exist.

What load would that be? In my newest computer for example (which has a decent supply which doesn't need the 5V load), it uses an SSD which draws fuck all power. The CPU and graphics card and memory all operate from 12V with their own regulator modules. I can't think of anything drawing much 5V, and the dodgy supplies I have need approaching 2 amps! I tried 1 amp and that didn't completely stabilize the 12V line.

The whole idea of taking a supply designed for a specific purpose,
using it for a different purpose is "dodgy," to use your words. Your
expectations may not be realistic.

If I'm designing something to be used in an industrial application I
can make it damn near foolproof and bullet proof, because the
application justifies it. If it's a military contract cost is
secondary to reliability and ruggedness. If I'm designing for a mass
market commercial application cost is very important...

Engineering is all about compromise. There's a lot of different ways
to do things. Sure the PS could be made better, but if that means you
price yourself out of the market what did you achieve?

You should also consider that ATX isn't much of a standard in the
sense that it wasn't carved in stone and handed down from the
mountain, never to deviate. There have been many iterations of the
basic ATX since it was introduced. Requirements change, obsolete
parts get supplanted with newer ones, etc..

The only thing you can count on is change. (and humans are involved -
lower your expectations or face disappointment)
Some power supplies sit and oscillate if they don't have a load on the
5 volt line...

I notice my desktop has 5v present on the USB connector even when it
is turned off, turned on, or just in standby. I suspect it may have a
small independent supply to run the USB connectors for power, and
perhaps that also supplies the CMOS memory so the clock and settings
don't drain the battery.
 
On Thu, 20 Feb 2020 01:22:00 -0000, Jasen Betts <jasen@xnet.co.nz> wrote:

On 2020-02-20, Commander Kinsey <CFKinsey@military.org.jp> wrote:
On Wed, 19 Feb 2020 23:31:01 -0000, Jasen Betts <jasen@xnet.co.nz> wrote:

On 2020-02-19, Commander Kinsey <CFKinsey@military.org.jp> wrote:
Why do (cheap? expensive ones may be better) PC ATX power supplies need current drawn from the 5V line to make the 12V line work correctly?

I have a PC with 3 graphics cards running scientific applications. I acquired three old graphics cards that take about 300W each, and have loads of cheap (CIT) PSUs that are rated at 650W on the 12V line, which is what those cards use. So I run each card off its own supply. But the 12V line at no load, or even at 300W, is only giving out 10 to 10.5V. If I attach a small dummy load of an amp or so to the 5V line, the 12V line suddenly becomes 12V.

Why are the two lines related in any way?

because all the output voltages come from taps on the same transformer
and the voltage regulation is applied to the input to that transformer
and the voltage regulation only watches the 5V line.

Ok, but why does current need to be taken from 5V to make the voltage monitor work?

because if there's enough voltage on the 5V no effort is made to
supply the 12V

Sounds like a horrid design making way too many assumptions.
 
On Thu, 20 Feb 2020 01:29:02 -0000, Jasen Betts <jasen@xnet.co.nz> wrote:

On 2020-02-20, Commander Kinsey <CFKinsey@military.org.jp> wrote:
On Wed, 19 Feb 2020 23:46:45 -0000, Jasen Betts <jasen@xnet.co.nz> wrote:

On 2020-02-19, Commander Kinsey <CFKinsey@military.org.jp> wrote:
On Wed, 19 Feb 2020 18:54:56 -0000, Phil Hobbs <pcdhSpamMeSenseless@electrooptical.net> wrote:

On 2020-02-19 10:43, Commander Kinsey wrote:
Why do (cheap? expensive ones may be better) PC ATX power supplies need
current drawn from the 5V line to make the 12V line work correctly?

I have a PC with 3 graphics cards running scientific applications. I
acquired three old graphics cards that take about 300W each, and have
loads of cheap (CIT) PSUs that are rated at 650W on the 12V line, which
is what those cards use. So I run each card off its own supply. But
the 12V line at no load, or even at 300W, is only giving out 10 to
10.5V. If I attach a small dummy load of an amp or so to the 5V line,
the 12V line suddenly becomes 12V.

Why are the two lines related in any way?

Sorry for the crosspost, I'm not sure which of these groups are active.

A lot of cheap supplies regulate only one output, and rely on
cross-regulation via the transformer to control the others. If the
regulated output isn't loaded, it rises out of spec and so do the others.

Strangely, with no 5V load, I get 5.2V and 10.5V. A small rise and a large drop.

I can't understand why the following happens: No 5V load, 12V is out. Small (2A) 5V load, 12V is ok. Yet if I draw 30A from 5V, the 12V is still ok? How can zero load upset it, but 2A or 30A (big difference) both be ok?

The difference between nothing and 2A is a factor of infinity
The difference between 2A and 20A is only a factor of 10

Actually the fisrt is more like the difference betweem 2mA and 2A
because the internal feedback takes avout 2mA to run the LM431 and the
optocoupler. So going to 2A loads the 5V output by 1000 times more, better
than infinity. but not by much.

If it's going to be so shit, they could have added a dummy load inside the PSU to make it work properly.

They could, but it was already working propperly with their existing
external loads.

They assume all computers take a couple of amps at 5V. With modern machines that may not be true. Almost everything takes 12V now.
 
On Thu, 20 Feb 2020 21:33:42 -0000, default <default@defaulter.net> wrote:

On Thu, 20 Feb 2020 20:07:47 -0000, "Commander Kinsey"
CFKinsey@military.org.jp> wrote:

On Thu, 20 Feb 2020 03:43:58 -0000, default <default@defaulter.net> wrote:

On Thu, 20 Feb 2020 00:32:05 -0000, "Commander Kinsey"
CFKinsey@military.org.jp> wrote:

On Wed, 19 Feb 2020 23:31:01 -0000, Jasen Betts <jasen@xnet.co.nz> wrote:

On 2020-02-19, Commander Kinsey <CFKinsey@military.org.jp> wrote:
Why do (cheap? expensive ones may be better) PC ATX power supplies need current drawn from the 5V line to make the 12V line work correctly?

I have a PC with 3 graphics cards running scientific applications. I acquired three old graphics cards that take about 300W each, and have loads of cheap (CIT) PSUs that are rated at 650W on the 12V line, which is what those cards use. So I run each card off its own supply. But the 12V line at no load, or even at 300W, is only giving out 10 to 10.5V. If I attach a small dummy load of an amp or so to the 5V line, the 12V line suddenly becomes 12V.

Why are the two lines related in any way?

because all the output voltages come from taps on the same transformer
and the voltage regulation is applied to the input to that transformer
and the voltage regulation only watches the 5V line.

Ok, but why does current need to be taken from 5V to make the voltage monitor work?

It is designed to be in a computer, and there's always some load on
the 5 V line. It probably didn't seem terribly important to worry
about a high voltage condition where none should ever exist.

What load would that be? In my newest computer for example (which has a decent supply which doesn't need the 5V load), it uses an SSD which draws fuck all power. The CPU and graphics card and memory all operate from 12V with their own regulator modules. I can't think of anything drawing much 5V, and the dodgy supplies I have need approaching 2 amps! I tried 1 amp and that didn't completely stabilize the 12V line.

The whole idea of taking a supply designed for a specific purpose,
using it for a different purpose is "dodgy," to use your words. Your
expectations may not be realistic.

I didn't. I'm powering computer parts. Just a few less than it was designed for. For something to be upset because it's doing less work is crazy.

If I'm designing something to be used in an industrial application I
can make it damn near foolproof and bullet proof, because the
application justifies it. If it's a military contract cost is
secondary to reliability and ruggedness. If I'm designing for a mass
market commercial application cost is very important...

Engineering is all about compromise. There's a lot of different ways
to do things. Sure the PS could be made better, but if that means you
price yourself out of the market what did you achieve?

Corsair make supplies that don't need a 5V current. Everything is rock solid at the correct voltage no matter what. That's why they sell so many, because their units just work.

You should also consider that ATX isn't much of a standard in the
sense that it wasn't carved in stone and handed down from the
mountain, never to deviate. There have been many iterations of the
basic ATX since it was introduced. Requirements change, obsolete
parts get supplanted with newer ones, etc..

The only thing you can count on is change. (and humans are involved -
lower your expectations or face disappointment)

It states on the label the maximum current I can draw. It does not state any minimum. Shortfalls should be clearly advertised.

Some power supplies sit and oscillate if they don't have a load on the
5 volt line...

I notice my desktop has 5v present on the USB connector even when it
is turned off, turned on, or just in standby. I suspect it may have a
small independent supply to run the USB connectors for power, and
perhaps that also supplies the CMOS memory so the clock and settings
don't drain the battery.
 

Welcome to EDABoard.com

Sponsor

Back
Top