Velocity Reviews > Major Computer Problem

Major Computer Problem

westom
Guest
Posts: n/a

 03-15-2009
On Mar 15, 2:09*am, Scribner <(E-Mail Removed)> wrote:
> Poking inside a computer, especially the power wires, when the
> computer is on sends cold chills down my spine. *However, maybe this
>
> Vcore1: 1.28V * * Vcore2: 1.58V * +3.3V: 1.84V *+5V: *5.00V
> *+12V: *10.50V *-12V: -0.10V * * * -5: -0.23V * * * +5V: 4.89V
> *Vbat: *3.28V

I gather this is from the motherboard monitor. That measurement
device is not sufficiently accurate until calibrated with a
multimeter. However, assuming those numbers are correct inside your
machine, your machine should not even work.

3.3 at 1.84 volts mean no operation. +12 volts at 10.5 volts is a
complete failure. -12 volts is missing, required, but typically not
needed to boot and run a computer. -5 volts typically has no function
in any typical computer and is often missing on most power supplies.
+5 is barely sufficient. Bat at 3.28 means the battery is fresh -
probably new - no problem.

If the system is a reason for failure, I would start with devices
using 3.3 volts. That eliminates any powered peripherals such as disk
drive, keyboard, CD-Rom, etc.

So, is the load excessive due to a problem with hardware or a
problem with the power supply? If I remember correctly, you spend
much time looking for a power supply that met specs. Well, your
motherboard requires 28 amps on 3.3 volts. Then that current
increases with a plug-in video controller and other peripherals. IOW
the power supply may be undersized. Assuming your voltages numbers
are correct, then a low 3.3 volts (and other voltages) may report a
power supply with insufficient current on 3.3 volts.

If "Poking inside a computer ... sends cold chills down my spine."
was a valid worry, then many are being electrocuted by jump starting a
car. Car has even higher voltages - and is an electrical threat to no
one. Anything dangerous inside a computer is heavily protected by
steel (with signed saying do not open) so that you could never be at
danger. And then the meter in VDC setting means even less threats.
If fearing what is inside a powered on computer, then never cross the
street. That is far more dangerous.

How does the power supply turn on? Does the 120 volts go through
the power switch? No. Of course not. That might put you at risk.
Even the power switch has volts so low as to not even threaten an
infant. The level of safety in an ATX computer is that large.

Better is to first confirm those numbers with a multimeter.
Measurements that would also calibrate the motherboard monitor. If
the meter confirms those readings, repeat measurements with the video
controller removed. Those voltages should increase if the power
supply is insufficient. And if the power supply really does output
and motherboard demands what is claimed, then you have confirmed the
power supply is woefully too small; especially on the 3.3 volts.

Moving on. Let's assume the 3.3 volts is too small. So you get a
power supply with more current (greater than 28 amps) on the 3.3 V.
It probably needs more current on the +12 volts (also reinstall video
controller). Then use the multimeter again to confirm that new supply
is more than sufficient when computer is booted to maximum load. If
yes (and again, post those numbers here for further analysis or to
good.

Why perform that last measurement? Because even a defective or
undersized power supply can still boot a computer. Measurements on a
new supply are an only fact to say that supply is sufficient for the
load. Many foolishly believe that if a computer boots, then the power
supply is OK. Nobody can know that without those multimeter readings.

Not detailed here (yet) is what to do if removing a video controller
causes no measurement number changes.

Finally a caution. With computer powered off, measure the purple
wire voltage (your previous post did not list that voltage). Notice
that the purple wire (also called +5VSB) always provides power when
computer is on or off. That is why you remove or install nothing
without first disconnecting the power cord. Physically remove that AC
power cord before changing anything inside the machine - to protect
some transistors on some boards. It is a simple rule and the only
internal danger. Some motherboards glow an LED just to warn you that
the power cord is still connected.

Leythos
Guest
Posts: n/a

 03-15-2009
In article <(E-Mail Removed)>,
http://www.velocityreviews.com/forums/(E-Mail Removed) says...
> Vcore1: 1.28V Vcore2: 1.58V +3.3V: 1.84V +5V: 5.00V
> +12V: 10.50V -12V: -0.10V -5: -0.23V +5V: 4.89V
> Vbat: 3.28V
>

Disconnect parts in the computer and on the motherboard until the
voltages come back to the normal levels.

I would start with disconnecting EVERYTHING that is not fixed to the
motherboard, you need the CPU and Video card - see if there are any
changes.

If you have a spare video card I would use it also, for testing.

It would appear you either have a bad PSU or a major motherboard
problem.

--
- Igitur qui desiderat pacem, praeparet bellum.
- Calling an illegal alien an "undocumented worker" is like calling a
(E-Mail Removed) (remove 999 for proper email address)

Leythos
Guest
Posts: n/a

 03-15-2009
In article <(E-Mail Removed)>,
(E-Mail Removed) says...
> Vcore1: 1.28V Vcore2: 1.58V +3.3V: 1.84V +5V: 5.00V
> +12V: 10.50V -12V: -0.10V -5: -0.23V +5V: 4.89V
> Vbat: 3.28V
>

Disconnect parts in the computer and on the motherboard until the
voltages come back to the normal levels.

I would start with disconnecting EVERYTHING that is not fixed to the
motherboard, you need the CPU and Video card - see if there are any
changes.

If you have a spare video card I would use it also, for testing.

It would appear you either have a bad PSU or a major motherboard
problem.

--
- Igitur qui desiderat pacem, praeparet bellum.
- Calling an illegal alien an "undocumented worker" is like calling a
(E-Mail Removed) (remove 999 for proper email address)

Desk Rabbit
Guest
Posts: n/a

 03-15-2009
westom wrote:
> On Mar 13, 8:44 am, "Mike Easter" <(E-Mail Removed)> wrote:
>> So; I definitely would *NOT* say that UPS has only one purpose.

>
> If any extra features were necessary, then that UPS provides more
> than just a battery backup function. But those functions such as line
> conditioning are performed by the computer's power supply. Why are
> 200 volts square waves with a 270 volt spike not a problem?
> Computer's power supply contains line conditioning. The UPS does
> nothing because those other functions are made irrelevant by a
> computer's power supplies.

Let's have some facts for that backed with some technical specs and
documentation.

> Spend \$500 on a UPS that only does what the computer already makes
> irrelevant? That \$500 UPS simply does the same thing that a \$100 UPS
> does - supply battery backup power so that data is not lost.
>
> When a computer grade UPS switches to and from battery backup, power
> is temporarily lost. No problem. A computer's supply must even
> output power without any voltage drop when incoming AC is temporarily
> lost. Just another example of line conditioning already performed
> inside a computer's power supply.

Come on, come clean, you are just making this up as you go along aren't you?

Desk Rabbit
Guest
Posts: n/a

 03-15-2009
Scribner wrote:
> On Fri, 13 Mar 2009 20:27:01 -0700 (PDT), westom <(E-Mail Removed)>
> wrote:
>
>> On Mar 13, 10:42 am, Scribner <(E-Mail Removed)> wrote:
>>> I'm thinking that when I was trying to copy the VHS to DVD I
>>> overheated the CPU and it blew the power supply. I sincerely doubt if
>>> my power outages had anything to do with the power supply blowing. But
>>> the dead CPU fan probably did.

>> Nothing in a computer can blow a power supply. True even long
>> before PCs existed. Furthermore, overheating a CPU does no supply
>> harm. Meanwhile, do you also know that the CPU has its own separate
>> supply?
>>
>> Don't recall which CPU you had. But had it been an Intel,
>> overheating would cause no damage - even to the CPU.
>>
>> With a multimeter - there is no way around best advise from someone
>> who has done this for generations - then you would have seen that
>> power supply and other power system components were OK in only two
>> minutes. Then the third minute was spent looking for the actual
>> problem.
>>
>> Is a power supply sufficiently sized? Again a meter would have
>> reported that immediately. An undersized power supply can still boot
>> and run a computer. But that same undersized supply is immediately
>> known, without doubt, if using the meter.
>>
>> Video cards cannot "pull so much power that the blow the ps".
>> Obvious as it was long before PCs existed. But hearsay is alive and
>> well.
>>
>> Many supplies sold to computer assemblers are missing essential
>> functions to sell only on dollars and watts. Therefore numeric specs
>> get 'forgotten'. No specs mean the 1% who actually know this stuff
>> cannot 'blow the whistle'.
>>
>> What are those essential functions? All supply outputs can be
>> shorted and the supply is not harmed. Intel specs even define that
>> test by listing the minimum testing wire diameter. Overpower
>> protection means nothing in a computer can harm that supply - draw too
>> much power. Overvoltage protection means a supply can never harm
>> computer components. All functions and others were standard even
>> before and in the original PC.
>>
>> Some power supply manufacturers may 'forget' to install required
>> functions. Then their supply might harm computer components. Or too
>> much load might damage the supply. Failures found in some (but not
>> all) supplies provided without numeric specs. Suspect the worst if a
>> supply sells for \$25 or \$40 retail.
>>
>> Your symptoms in that first post are also explained by another
>> function of a power supply system. Apparently the safety lockout
>> function was triggered. You then assumed that power supply failed.
>> Replacing the power supply restart that safety lockout. Did you
>> confuse safety lockout with a defective power supply? However we
>> would know this immediately and without doubt had the multimeter been
>> used.
>>
>> Moving on to include information in afternoon posts: You can
>> keep replacing power supplies - or first learn what exists. That
>> means two minutes, a computer under maximum load, and the multimeter.
>> If the meter defines a power supply definitively good, then move on to
>> other suspects - and don't even look back.
>>
>> First are the four voltages when under maximum load from any one
>> orange, purple, red, and yellow wires. Then are the signal lines that
>> the supply and supply controller talk on - gray and green wires. Those
>> numbers necessary to confirm the supply, it controller, etc. .. so
>> that we can move on to other suspects.
>>
>> For example, if a supply controller has a defective driver, than all
>> power supplies will perform strangely.

>
>
> Poking inside a computer, especially the power wires, when the
> computer is on sends cold chills down my spine. However, maybe this
>
> Vcore1: 1.28V Vcore2: 1.58V +3.3V: 1.84V +5V: 5.00V
> +12V: 10.50V -12V: -0.10V -5: -0.23V +5V: 4.89V
> Vbat: 3.28V
>
> My motherboard requires the following from a power supply:
> +3.3V Must be "28A" or more.
> +5V Must be "28A" or more.
> +12V Must be "18A" or more.
> The rest are "Any."
>
> My Thermaltake 600W specs are as follows:
> +3.3V 30A
> +5V 28A
> +12V1 18A
> +12V2 18A
>
> And this power supply lists for over \$100.00.

Ignore westom's verbal diarrhoea. Your -12 +3.3 and -5 voltages look all
wrong. Were these measured with a multimeter? Your computer should not
even be working with those levels. Are you absolutely 100% certain you
measured at the right points?

measurement device is not sufficiently accurate until calibrated with a
multimeter." This is utter *******s and I've never seen one yet that you
could calibrate. Sure they are not accurate but they are good enough for
the purpose.

Lets look at some facts. We can be fairly certain that your brand new
PSU is working within reasonable limits because your computer is

Leythos
Guest
Posts: n/a

 03-15-2009
In article <gpit77\$7i0\$(E-Mail Removed)>, (E-Mail Removed)
says...
> Ignore westom's verbal diarrhoea. Your -12 +3.3 and -5 voltages look all
> wrong. Were these measured with a multimeter? Your computer should not
> even be working with those levels. Are you absolutely 100% certain you
> measured at the right points?
>
> measurement device is not sufficiently accurate until calibrated with a
> multimeter." This is utter *******s and I've never seen one yet that you
> could calibrate. Sure they are not accurate but they are good enough for
> the purpose.
>
> Lets look at some facts. We can be fairly certain that your brand new
> PSU is working within reasonable limits because your computer is

A PC could allow entry into the POST without everything working, and I
suspect he got the values from the BIOS monitoring section.

Having worked with more than 10K motherboards in my years, I've also
never seen one that allows a user or tech to adjust the readings, they
are done by the vendor and are not adjustable.

devices being connected - it could also be them.

Basic diagnostic processes indicate removing all devices/attachments to
see what the POST shows and start adding one at a time to determine the
problem device - I've not seen where the OP has done that yet.

--
- Igitur qui desiderat pacem, praeparet bellum.
- Calling an illegal alien an "undocumented worker" is like calling a
(E-Mail Removed) (remove 999 for proper email address)

Buffalo
Guest
Posts: n/a

 03-15-2009

Scribner wrote:
> Poking inside a computer, especially the power wires, when the
> computer is on sends cold chills down my spine. However, maybe this
>
> Vcore1: 1.28V Vcore2: 1.58V +3.3V: 1.84V +5V: 5.00V
> +12V: 10.50V -12V: -0.10V -5: -0.23V +5V: 4.89V
> Vbat: 3.28V
>
> My motherboard requires the following from a power supply:
> +3.3V Must be "28A" or more.
> +5V Must be "28A" or more.
> +12V Must be "18A" or more.
> The rest are "Any."
>
> My Thermaltake 600W specs are as follows:
> +3.3V 30A
> +5V 28A
> +12V1 18A
> +12V2 18A
>
> And this power supply lists for over \$100.00.

What are you using to get those voltages? Something is wrong. If you are
using MotherBoardMonitor, you most likely are using the wrong inputs under
the Voltage Configuration box in the Voltage section.
Everest Home Edition seems to do a pretty good job of determining the
correct voltages. Just expand the Computer header and select Sensor. I
believe it is still a free program.
Of course, the multimeter will usually be the most accurate.
Just out of curiousity, what is the make and model of your MB?
Buffalo

chuckcar
Guest
Posts: n/a

 03-15-2009
Desk Rabbit <(E-Mail Removed)> wrote in
news:gpisbu\$ud3\$(E-Mail Removed):

>> I *suggested* it as a solution. The problem is that the computer and
>> the converter draw more power than the computer's power supply can
>> provide, That's *why* they've been blowing so a UPS wouldn't fix the
>> problem. You'd still have the power coming from one source - the
>> computer's power supply. You need another power source for the VCR to
>> DVD device, hence a hub for it.

>
> You didn't mention a hub at all. In fact your exact words were "Also do
> *not* connect your VCR to DVD device to your computer's power supply,
> that's what blew it. Use a separate power supply plugged into the
> wall."
>

As in the usb hub. Which I *did* mention in the first thread.

> The OP did not plug hs VCR/DVD into his power supply. He plugged his USB
> powered VCR/DVD into a usb slot.

So the power from the computer's USB ports don't come from the computers
power supply???

> Seems perfectly reasonable to me to
> plug a USB powered device into a USB port. If though you feel the urge
> to plug a USB device into a mains wall socket, I'll be the first to
> provide details instructions for you.
>

Now you're just trolling.

--
(setq (chuck nil) car(chuck) )

Pennywise@DerryMaine.Gov
Guest
Posts: n/a

 03-16-2009
bud-- <(E-Mail Removed)> wrote:

>> A UPS also offers hardware protection, with Zener Diode (depending
>> upon the circuit), If the surge is too high the diode will direct it
>> to ground

>
>The vast majority of devices that actually do the surge protection are MOVs.

Your right MOV's are used, I couldn't think of it's name and seriously
google would only give me Zener Diode's with my key search words.

>> And yes you can use a volt/ohm meter to verify the diode is still
>> working.

>I know of no easy practical way to test MOVs.

I wasn't aware of that one, I thought there was resistance in one
direction if working correctly.

I had never tried to test a MOV myself, as unsoldering mine to test
them wasn't practicable (I was just after the batteries).
--

(Click on the dog house)
http://amanita-design.net/samorost-2/

Desk Rabbit
Guest
Posts: n/a

 03-16-2009
chuckcar wrote:
> Desk Rabbit <(E-Mail Removed)> wrote in
> news:gpisbu\$ud3\$(E-Mail Removed):
>
>>> I *suggested* it as a solution. The problem is that the computer and
>>> the converter draw more power than the computer's power supply can
>>> provide, That's *why* they've been blowing so a UPS wouldn't fix the
>>> problem. You'd still have the power coming from one source - the
>>> computer's power supply. You need another power source for the VCR to
>>> DVD device, hence a hub for it.

>> You didn't mention a hub at all. In fact your exact words were "Also do
>> *not* connect your VCR to DVD device to your computer's power supply,
>> that's what blew it. Use a separate power supply plugged into the
>> wall."
>>

> As in the usb hub. Which I *did* mention in the first thread.

Which you then changed to a power strip.

>
>> The OP did not plug hs VCR/DVD into his power supply. He plugged his USB
>> powered VCR/DVD into a usb slot.

>
> So the power from the computer's USB ports don't come from the computers
> power supply???

Via an on-board regulator.

>> Seems perfectly reasonable to me to
>> plug a USB powered device into a USB port. If though you feel the urge
>> to plug a USB device into a mains wall socket, I'll be the first to
>> provide details instructions for you.
>>

> Now you're just trolling.

Nope, just wishfull thinking