PSU voltage figures

A

Aaron S

Recently I have been attaching several extra devices to my PC and I am
worrying that the PSU is beginning to get overloaded.

Is my PSU getting into trouble based on this data?

---------------- START DATA ------------
When the PC booted up and all the hardware devices were attached then a
screen in the BIOS showed that the nominal 12v voltage was 2.0 % down at
11.77 and the nominal 5v was 2.2% down at 4.89v.

When I booted into XP and using MBM I saw the figure for 12v went up to
4.93v (1.4% down) and 12v went down to 11.81 (1.6% down).

Then I saw som eother figures in MBM:

+3.3V was 3.31v

-12V was between -5.90V and -10.71V with an average of -7.74V. This seems
like a huge difference.

-5.0V was showing in MBM as 0.00V. Maybe is not used or does not exist.

---------------- END DATA ------------

So what do you reckon about my PSU?
 
W

w_tom

If MBM is the onboard voltage monitor, then you still do not
know what those voltages are. It is a monitor build with low
accuracy components. Once calibrated, then the monitor can
detect voltage changes. But you must still get the 3.5 digit
multimeter to take voltage measurements and to calibrate that
voltage monitor.

Typically the -12 volts has no real load. So what is
loading down the -12 volts. Or is that voltage monitor that
inaccurate? Insufficient information until reading are
obtained from a meter.
 
K

kony

Recently I have been attaching several extra devices to my PC and I am
worrying that the PSU is beginning to get overloaded.

Is my PSU getting into trouble based on this data?

---------------- START DATA ------------
When the PC booted up and all the hardware devices were attached then a
screen in the BIOS showed that the nominal 12v voltage was 2.0 % down at
11.77 and the nominal 5v was 2.2% down at 4.89v.

When I booted into XP and using MBM I saw the figure for 12v went up to
4.93v (1.4% down) and 12v went down to 11.81 (1.6% down).

Then I saw som eother figures in MBM:

+3.3V was 3.31v

-12V was between -5.90V and -10.71V with an average of -7.74V. This seems
like a huge difference.

-5.0V was showing in MBM as 0.00V. Maybe is not used or does not exist.

---------------- END DATA ------------

So what do you reckon about my PSU?


That looks typical of a normally working PSU, there is
nothing there to be concerned about. -5V and -12V may vary
wildly since there is no load on them, it is not a problem.
That BOTH the 5V & 12V are similarly low, is actually
(usually) a good sign, opposed to one being too high and the
other too low.

However, voltage readings can't be taken this way and
considered accurate, because motherboard trace resistance
lowers the voltage by varying amounts on different boards.
In fact, extreme overclockers will often note (if they
bother to check) that their voltage levels are low on the
motherboard, but voltage at connection to motherboard is
fine... Power supply is only supposed to output at correct
voltage at it's connectors, not higher such that motherboard
reading is compensated for.

You need to use a multimeter to take voltage readings on the
backside of the ATX connector (where the wires enter it)
while system is running, to get valid output voltage
readings.
 
A

Aaron S

kony said:
That looks typical of a normally working PSU, there is
nothing there to be concerned about. -5V and -12V may vary
wildly since there is no load on them, it is not a problem.
That BOTH the 5V & 12V are similarly low, is actually
(usually) a good sign, opposed to one being too high and the
other too low.

Kony, can I ask you or your colleagues here a related question. I have got
*five* IDE hard drives attached to this PC (plus a CD burner). There is a
general shortage of mounting spaces for all the HDDs but I have found a
workable solution which seems ok.

My PSU has *three* peripheral power connectors (each with a set of wires
leading to a 4 pin Molex connector). If it is relevant, this is the actual
PSU though there doesn't seem a lot of info there. http://snipurl.com/954t

On account of the locations of the HDDs I have had to "daisy chain" three
(and maybe soon four) of the HDDs off *one* 4 pin Molex power connector. Is
this sort of arrangement ok? Maybe the wires in each 4 pin Molex connector
goes back to the same output terminals on the PSU? If you see what I mean.

Or alternatively, does one set of 4 pin Molex wires go back to a different
winding on the transformer than another set? If so, then I don't want to
overload one part of the transformer or PSU circuitry by attaching too many
devices to just one power connector.
 
C

Conor

Recently I have been attaching several extra devices to my PC and I am
worrying that the PSU is beginning to get overloaded.

Is my PSU getting into trouble based on this data?

---------------- START DATA ------------
When the PC booted up and all the hardware devices were attached then a
screen in the BIOS showed that the nominal 12v voltage was 2.0 % down at
11.77 and the nominal 5v was 2.2% down at 4.89v.

When I booted into XP and using MBM I saw the figure for 12v went up to
4.93v (1.4% down) and 12v went down to 11.81 (1.6% down).

Then I saw som eother figures in MBM:

+3.3V was 3.31v

-12V was between -5.90V and -10.71V with an average of -7.74V. This seems
like a huge difference.

-5.0V was showing in MBM as 0.00V. Maybe is not used or does not exist.

---------------- END DATA ------------

So what do you reckon about my PSU?
Unless you've verified those details with a multimeter then they're not
reliable. As a system gets warmer, internal resistance increases thus
lowering internal voltages.
 
K

kony

Kony, can I ask you or your colleagues here a related question. I have got
*five* IDE hard drives attached to this PC (plus a CD burner). There is a
general shortage of mounting spaces for all the HDDs but I have found a
workable solution which seems ok.

My PSU has *three* peripheral power connectors (each with a set of wires
leading to a 4 pin Molex connector). If it is relevant, this is the actual
PSU though there doesn't seem a lot of info there. http://snipurl.com/954t

On account of the locations of the HDDs I have had to "daisy chain" three
(and maybe soon four) of the HDDs off *one* 4 pin Molex power connector. Is
this sort of arrangement ok? Maybe the wires in each 4 pin Molex connector
goes back to the same output terminals on the PSU? If you see what I mean.

Or alternatively, does one set of 4 pin Molex wires go back to a different
winding on the transformer than another set? If so, then I don't want to
overload one part of the transformer or PSU circuitry by attaching too many
devices to just one power connector.

All wires of same color join at same place on the circuit
board, not a different transformer winding or anything like
that. The issue is generally how much current the wire
itself can sustain, and 4 HDDs is better avoided but
possible. In any event, measurements can be taken with a
multimeter at the end, to see if voltage has dropped
significantly.

Otherwise, this is not any more of a load or overload than
if you had the drives all connected differently, the total
load remains (almost, very close to) the same.

Ideally with so many drives, and presumably a lot of data on
them, you might consider a higher wattage power supply.
Even though voltages look ok, data is often worth more than
the cost of PSU replacement. You didn't mention the rest of
the system, how old or power hungry it is, but generally a
box with that many drives (and considering you may've
upgraded other parts as well) is better paired with a 350W
PSU if not higher... particularly with an Athlon or P4 box,
instead of something less demanding like a P3.
 
W

w_tom

Often not discussed, but if we review the manufacturer's
specs for Molex connector pins, well those exapansion cords
are assembled by violating the manufacturer specs. For
example, if the manufacturer says wire must be 18-22 AWG, then
that means even 23 AWG wire if not acceptable for reliable
operation.

Two 22 AWG wires may be pushed into a crimp connector
designed only for one wire. Normally not a problem even
thought it outrightly violates the manufacturer spec.
However, on at least one occassion, I found this resulted is
peripheral failure. The wire power wire was intermittent.

Best to verify that voltage at each peripheral maintains
same voltage on all power supply connectors. Another reason
why multimeters are an important tool. A low voltage would
imply a defective crimp inside the Molex connector. I solved
that connector problem by soldering those crimped wires into
that crimp connection.
 
F

Franc Zabkar

Recently I have been attaching several extra devices to my PC and I am
worrying that the PSU is beginning to get overloaded.

Is my PSU getting into trouble based on this data?

---------------- START DATA ------------
When the PC booted up and all the hardware devices were attached then a
screen in the BIOS showed that the nominal 12v voltage was 2.0 % down at
11.77 and the nominal 5v was 2.2% down at 4.89v.

When I booted into XP and using MBM I saw the figure for 12v went up to
4.93v (1.4% down) and 12v went down to 11.81 (1.6% down).

Then I saw som eother figures in MBM:

+3.3V was 3.31v

-12V was between -5.90V and -10.71V with an average of -7.74V. This seems
like a huge difference.

-5.0V was showing in MBM as 0.00V. Maybe is not used or does not exist.

---------------- END DATA ------------

So what do you reckon about my PSU?

Your data are meaningless unless you can be certain that MBM is
telling you the truth. As others have said, buy yourself a DMM and
measure the actual voltages. Failing that, believe what the BIOS tells
you, rather than MBM, because the BIOS is written to match the
motherboard hardware. Incidentally, you didn't mention what the BIOS
had to say about the -5V and -12V rails.

The -5V rail is not bussed to PCI slots, so if the motherboard doesn't
use it for anything, then it may as well not exist. OTOH, the -12V
rail *is* specified for PCI use, so something is wrong there. Contrary
to what others have said, the -12V rail should not fluctuate anywhere
near as much as you have observed, not even under no-load conditions.
I doubt that your PSU is faulty, however. Rather, I suspect that MBM
does not understand how your motherboard is laid out.


- Franc Zabkar
 
S

Steve Wolfe

On account of the locations of the HDDs I have had to "daisy chain" three
(and maybe soon four) of the HDDs off *one* 4 pin Molex power connector. Is
this sort of arrangement ok? Maybe the wires in each 4 pin Molex connector
goes back to the same output terminals on the PSU? If you see what I
mean.

That's just fine. There is a limit as to how much current you can push
down those wires (they're not very large), but you're not going to come
anywhere near that limit with just three or four hard drives. I wouldn't
worry until I had at least 8 drives daisy-chained.
Or alternatively, does one set of 4 pin Molex wires go back to a different
winding on the transformer than another set? If so, then I don't want to
overload one part of the transformer or PSU circuitry by attaching too many
devices to just one power connector.

They all go back to the same spot. The only consideration (and this is
why the GF6800U cards need connectors on two different sets of wires) is
that because PSU wires are typically pretty thin wire, you're not going to
be able to get all of the 12V amperage that the PSU can supply down one set
of wires without a good bit of voltage drop.

steve
 
K

kony

Try this power calcultor an add in what u got

Even though your intentions may have been good, this is not
a binary newsgroup, posting such attachments is not
appropriate. Instead, please host the file on any server
available to you, and link to it, thanks.
 
A

Aaron S

w_tom said:
Often not discussed, but if we review the manufacturer's
specs for Molex connector pins, well those exapansion cords
are assembled by violating the manufacturer specs. For
example, if the manufacturer says wire must be 18-22 AWG, then
that means even 23 AWG wire if not acceptable for reliable
operation.

Two 22 AWG wires may be pushed into a crimp connector
designed only for one wire. Normally not a problem even
thought it outrightly violates the manufacturer spec.
However, on at least one occassion, I found this resulted is
peripheral failure. The wire power wire was intermittent.

What you write is interesting. Reading about specifications for the Molex
connector makes me ask two questions

(1) Do you know what current the Molex 4-pin connector is rated for?

(2) Approx what current capacity does 18 to 22 AWG wire have? We don't use
AWG to refer to our wires here in the UK.
 
F

Franc Zabkar

What you write is interesting. Reading about specifications for the Molex
connector makes me ask two questions

(1) Do you know what current the Molex 4-pin connector is rated for?

About 9A or 12A per pin.

http://www.jameco.com/webapp/wcs/st...toreId=10001&catalogId=10001&productId=121585
http://www.jameco.com/webapp/wcs/st...toreId=10001&catalogId=10001&productId=128349
(2) Approx what current capacity does 18 to 22 AWG wire have? We don't use
AWG to refer to our wires here in the UK.

http://www.dse.com.au/cgi-bin/dse.s...4a273fc0a87f9c06bb/Export/catalogs/DTS0000048


- Franc Zabkar
 
W

w_tom

I am not going to dig through the Molex data book. However
I recall those Molex pins were rated only for something like 5
amps per pin.

Franc is a well respected source of information. However I
recall having to use two pins in a Molex connector because my
current draw was something less than 10 amps - too much for a
single connector pin. 10 amps would also be too much for the
22 AWG wire connected to that pin. 22 AWG would be about .65
mm. If I remember, 10 amps on a 22 AWG wire would leave the
wire at something approaching 100 degree C.

These numbers should also be useful to put the original post
question into perspective.
 
F

Franc Zabkar

However I
recall having to use two pins in a Molex connector because my
current draw was something less than 10 amps - too much for a
single connector pin.

My old Molex catalogue states that the 0.062" (1.58mm) connector
series had a max current rating of 5A per pin. The larger 0.093"
(2.36mm) series are rated from 7.5 to 12A per pin depending on the
number of circuits.

The 8981 series 4-pin disc drive headers were rated at 10A max per
pin.

See http://www.molex.com/pdm_docs/ps/b9clpatf_ps.pdf
10 amps would also be too much for the
22 AWG wire connected to that pin. 22 AWG would be about .65
mm. If I remember, 10 amps on a 22 AWG wire would leave the
wire at something approaching 100 degree C.

The URL below specifies a nominal current rating of ~3A for 22 AWG
wire. The above Molex connector spec allows 6A for this gauge. said:
These numbers should also be useful to put the original post
question into perspective.


- Franc Zabkar
 
W

w_tom

Those numbers are more in line with what I remember.
However I never encountered 'larger' pins. Is this something
new?

Power supplies (not PC type) use standard configurations and
standard connectors. Even location of connector on power
supply is standardized. Problem is that power supplies now
output many times more power for the same configuration. But
that configuration must use same connector pins. IOW that
would explain why Molex now has a larger pin for the same
connector.
 
V

VWWall

w_tom said:
Those numbers are more in line with what I remember.
However I never encountered 'larger' pins. Is this something
new?

All of the 4-pin connectors on ATX power supplies use larger pins.
This may be what Molex calls: 2.13mm or 0.084" The female sockets will
accept a 0.093 solid pin. The male Molex pins are compressible and
fit the 0.093 sockets snugly.
Power supplies (not PC type) use standard configurations and
standard connectors. Even location of connector on power
supply is standardized. Problem is that power supplies now
output many times more power for the same configuration. But
that configuration must use same connector pins. IOW that
would explain why Molex now has a larger pin for the same
connector.

10 amps on a 22 AWG wire 18" long produces a watts loss of ~0.25 W. over
the entire length of the wire. It's hard to see how that would produce
100 degrees C. Try it yourself. 6 A is a more reasonable number.
Virg Wall
 
F

Franc Zabkar

V

VWWall

Franc said:
According to ...

http://www.dse.com.au/cgi-bin/dse.filereader?4151fcd200eed316273fc0a87f9c070b+EN/catalogs/DTS0000048

... 21 AWG copper wire has a max resistance of 0.05 ohms per metre. At
10A this equates to a dissipation of 5W/m. That's about 2.3W in total
for an 18" cable.

#22 wire is ~16 ohm/1000' or 0.024 for 10' At 10 A, this is 2.4 W. I
should have used my slide rule and checked for decimal position! It
still seems low for a 70 degree C temp rise. Thus 3 A limit would seem
prudent.

Virg Wall
 
S

Steve Wolfe

According to ...
http://www.dse.com.au/cgi-bin/dse.filereader?4151fcd200eed316273fc0a87f9c070b+EN/catalogs/DTS0000048

... 21 AWG copper wire has a max resistance of 0.05 ohms per metre. At
10A this equates to a dissipation of 5W/m. That's about 2.3W in total
for an 18" cable.

That resistance is measured at negligible current draw. As current goes
up, so will the resistance. Pretty soon (especially on such small wire),
you hit a point where even a small increase in current starts to cause LARGE
resistances - which mean significant voltage drops across the cable.

With 21-gauge aluminum wire at 10 amps, you're going to have a voltage
drop of around 0.85 volts just over a 2-foot length of cable. That means
you're dissipating 8.5 watts of power over the length of the cable, and it
means that you're really only getting 11.15 volts at the far end.

steve
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Top