Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > alt.comp.hardware > #20633 > unrolled thread

Computer turns on and then just dies

Started by"s|b" <me@privacy.invalid>
First post2024-11-21 16:19 +0100
Last post2024-12-07 15:27 +0100
Articles 20 on this page of 88 — 5 participants

Back to article view | Back to alt.comp.hardware


Contents

  Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-11-21 16:19 +0100
    Re: Computer turns on and then just dies Marco Moock <mm+solani@dorfdsl.de> - 2024-11-21 16:28 +0100
      Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-11-21 17:07 +0100
        Re: Computer turns on and then just dies "Mr. Man-wai Chang" <toylet.toylet@gmail.com> - 2024-11-22 00:18 +0800
          Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-11-21 17:57 +0100
            Re: Computer turns on and then just dies "Mr. Man-wai Chang" <toylet.toylet@gmail.com> - 2024-11-22 11:42 +0800
              Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-11-22 19:47 +0100
                Re: Computer turns on and then just dies Paul <nospam@needed.invalid> - 2024-11-22 20:34 -0500
            Re: Computer turns on and then just dies "Mr. Man-wai Chang" <toylet.toylet@gmail.com> - 2024-11-22 11:47 +0800
              Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-11-22 19:48 +0100
                Re: Computer turns on and then just dies "Mr. Man-wai Chang" <toylet.toylet@gmail.com> - 2024-11-23 12:12 +0800
            Re: Computer turns on and then just dies VanguardLH <V@nguard.LH> - 2024-11-23 15:18 -0600
    Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-11-21 16:35 +0100
      Re: Computer turns on and then just dies VanguardLH <V@nguard.LH> - 2024-11-21 12:43 -0600
        Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-11-22 19:56 +0100
          Re: Computer turns on and then just dies VanguardLH <V@nguard.LH> - 2024-11-22 17:32 -0600
    Re: Computer turns on and then just dies VanguardLH <V@nguard.LH> - 2024-11-21 12:22 -0600
      Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-11-22 20:01 +0100
        Re: Computer turns on and then just dies VanguardLH <V@nguard.LH> - 2024-11-22 18:03 -0600
    Re: Computer turns on and then just dies Paul <nospam@needed.invalid> - 2024-11-21 15:58 -0500
      Re: Computer turns on and then just dies VanguardLH <V@nguard.LH> - 2024-11-21 23:28 -0600
        Re: Computer turns on and then just dies Paul <nospam@needed.invalid> - 2024-11-22 07:24 -0500
          Re: Computer turns on and then just dies VanguardLH <V@nguard.LH> - 2024-11-22 10:20 -0600
      Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-11-23 12:09 +0100
        Re: Computer turns on and then just dies Paul <nospam@needed.invalid> - 2024-11-23 07:07 -0500
          Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-11-24 17:12 +0100
            Re: Computer turns on and then just dies Paul <nospam@needed.invalid> - 2024-11-24 15:21 -0500
              Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-11-25 14:58 +0100
                Re: Computer turns on and then just dies Paul <nospam@needed.invalid> - 2024-11-25 17:33 -0500
                  Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-11-26 18:44 +0100
                    Re: Computer turns on and then just dies VanguardLH <V@nguard.LH> - 2024-11-26 16:31 -0600
                      Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-11-27 15:38 +0100
                    Re: Computer turns on and then just dies Paul <nospam@needed.invalid> - 2024-11-26 21:04 -0500
                      Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-11-27 16:01 +0100
                        Re: Computer turns on and then just dies Paul <nospam@needed.invalid> - 2024-11-27 10:19 -0500
        Re: Computer turns on and then just dies VanguardLH <V@nguard.LH> - 2024-11-23 14:28 -0600
          Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-11-24 17:10 +0100
            Re: Computer turns on and then just dies Paul <nospam@needed.invalid> - 2024-11-24 15:27 -0500
              Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-11-25 15:08 +0100
          Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-11-25 14:38 +0100
            Re: Computer turns on and then just dies VanguardLH <V@nguard.LH> - 2024-11-26 16:37 -0600
              Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-11-27 16:05 +0100
    Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-11-24 16:51 +0100
      Re: Computer turns on and then just dies Paul <nospam@needed.invalid> - 2024-11-24 16:03 -0500
        Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-11-25 15:03 +0100
          Re: Computer turns on and then just dies Paul <nospam@needed.invalid> - 2024-11-25 18:09 -0500
            Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-11-30 12:28 +0100
      Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-11-27 15:45 +0100
    Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-11-28 17:17 +0100
      Re: Computer turns on and then just dies VanguardLH <V@nguard.LH> - 2024-11-28 11:07 -0600
        Re: Computer turns on and then just dies Paul <nospam@needed.invalid> - 2024-11-28 19:30 -0500
          Re: Computer turns on and then just dies VanguardLH <V@nguard.LH> - 2024-11-28 19:34 -0600
            Re: Computer turns on and then just dies Paul <nospam@needed.invalid> - 2024-11-29 05:21 -0500
              Re: Computer turns on and then just dies VanguardLH <V@nguard.LH> - 2024-11-29 11:35 -0600
          Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-11-29 13:49 +0100
            Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-11-29 17:11 +0100
              Re: Computer turns on and then just dies Paul <nospam@needed.invalid> - 2024-11-29 15:01 -0500
                Re: Computer turns on and then just dies VanguardLH <V@nguard.LH> - 2024-11-29 21:46 -0600
                Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-11-30 12:21 +0100
                  Re: Computer turns on and then just dies VanguardLH <V@nguard.LH> - 2024-11-30 09:15 -0600
                    Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-11-30 17:16 +0100
                      Re: Computer turns on and then just dies Paul <nospam@needed.invalid> - 2024-11-30 17:40 -0500
                        Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-12-02 11:43 +0100
                          Re: Computer turns on and then just dies VanguardLH <V@nguard.LH> - 2024-12-02 14:15 -0600
                            Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-12-02 21:19 +0100
                      Re: Computer turns on and then just dies VanguardLH <V@nguard.LH> - 2024-12-01 08:26 -0600
                        Re: Computer turns on and then just dies Paul <nospam@needed.invalid> - 2024-12-01 13:23 -0500
                        Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-12-02 11:40 +0100
                          Re: Computer turns on and then just dies VanguardLH <V@nguard.LH> - 2024-12-02 14:06 -0600
                            Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-12-02 21:30 +0100
                    Re: Computer turns on and then just dies Paul <nospam@needed.invalid> - 2024-11-30 11:31 -0500
                      Re: Computer turns on and then just dies VanguardLH <V@nguard.LH> - 2024-11-30 12:22 -0600
        Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-11-29 12:46 +0100
    Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-12-04 17:41 +0100
      Re: Computer turns on and then just dies VanguardLH <V@nguard.LH> - 2024-12-04 14:14 -0600
        Re: Computer turns on and then just dies Paul <nospam@needed.invalid> - 2024-12-04 18:28 -0500
          Re: Computer turns on and then just dies VanguardLH <V@nguard.LH> - 2024-12-05 09:30 -0600
            Re: Computer turns on and then just dies Paul <nospam@needed.invalid> - 2024-12-05 11:11 -0500
        Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-12-06 17:43 +0100
          Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-12-06 17:45 +0100
            Re: Computer turns on and then just dies VanguardLH <V@nguard.LH> - 2024-12-06 16:46 -0600
              Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-12-07 14:11 +0100
                Re: Computer turns on and then just dies VanguardLH <V@nguard.LH> - 2024-12-07 09:46 -0600
                  Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-12-07 17:02 +0100
                    Re: Computer turns on and then just dies VanguardLH <V@nguard.LH> - 2024-12-07 11:40 -0600
                      Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-12-09 15:54 +0100
          Re: Computer turns on and then just dies Paul <nospam@needed.invalid> - 2024-12-06 20:43 -0500
            Re: Computer turns on and then just dies "s|b" <me@privacy.invalid> - 2024-12-07 15:27 +0100

Page 2 of 5 — ← Prev page 1 [2] 3 4 5  Next page →


#20644

FromVanguardLH <V@nguard.LH>
Date2024-11-21 23:28 -0600
Message-ID<bzszpe56owzm.dlg@v.nguard.lh>
In reply to#20641
Paul <nospam@needed.invalid> wrote:

> ...
> 
> 2) Use a clamp-on ammeter, to monitor the +5VSB wire on the main cable.

For typical clamp meters, those rely on alternating current to fluctate
the magnetic field around the conductor; i.e., the common clamp meter
just measures AC.  The 5V standby line is DC, not AC.  With the
20/24-pin connector from PSU pushed onto the mobo (to account for any
load), use a voltmeter to measure DC voltage from the +5VSB line (pin 9,
purple) to a ground pin (black).

I can't think of any AC inside the computer excluding the input circuit
in the PSU.  Power into the PSU is AC.  All lines out of the PSU are DC.

DC also induces a magnetic field around the conductor, but not a
fluctuating magnetic field.  An AC clamp meter using a transformer wound
through the closing fingers of the clamp won't work on a DC wire.  For
DC, a Hall effect sensor can measure the magnetic field: the magnetic
field induces a current in the sensor.  However, those are inaccurate.
When measuring hundreds of amperes passing through a conductor, you
don't care if the actual current is different by several amps.  When
measuring 20A, or especially milliamps, the reading in the meter is so
inaccurate that it amounts to using a continuity tester to see there is
a connection versus how many ohms there are between the 2 points of
contact.  Both AC clamp-on meters and DC Hall effect meters measure by
conversion, not direct readings.  Magnetic field intensity decreases
exponentially by distance, and the huge space inside the fingers of the
clamp means the reading changes if you reposition the wire from the
center of the gap to closer to the wire, and it's never accurate,
anyway.

You and I may have AC and AC/DC clamp meters, but I suspect the OP does
not.  A good AC clamp meter at Home Depot costs $60 to $120.  A good DC
clamp meter greatly hikes the price to $500, and up.

> I try to keep a spare PSU in the house, but it's tough. 

I tried the same using a modular PSU, so I only need to connect enough
cables to the PSU as were needed for the particular build.  But
eventually it got used in the next build.  A $150+ PSU makes for an
expensive parts drawer.  For an emergency, I'll pay for overnight or
2-day delivery rather than buy a PSU that may never get used.  However,
I have other computers to use, so I go with free delivery in 5 days.

[toc] | [prev] | [next] | [standalone]


#20645

FromPaul <nospam@needed.invalid>
Date2024-11-22 07:24 -0500
Message-ID<vhpt5k$160i0$1@dont-email.me>
In reply to#20644
On Fri, 11/22/2024 12:28 AM, VanguardLH wrote:
> Paul <nospam@needed.invalid> wrote:
> 
>> ...
>>
>> 2) Use a clamp-on ammeter, to monitor the +5VSB wire on the main cable.
> 
> For typical clamp meters, those rely on alternating current to fluctate
> the magnetic field around the conductor; i.e., the common clamp meter
> just measures AC.  The 5V standby line is DC, not AC.  With the
> 20/24-pin connector from PSU pushed onto the mobo (to account for any
> load), use a voltmeter to measure DC voltage from the +5VSB line (pin 9,
> purple) to a ground pin (black).
> 
> I can't think of any AC inside the computer excluding the input circuit
> in the PSU.  Power into the PSU is AC.  All lines out of the PSU are DC.
> 
> DC also induces a magnetic field around the conductor, but not a
> fluctuating magnetic field.  An AC clamp meter using a transformer wound
> through the closing fingers of the clamp won't work on a DC wire.  For
> DC, a Hall effect sensor can measure the magnetic field: the magnetic
> field induces a current in the sensor.  However, those are inaccurate.
> When measuring hundreds of amperes passing through a conductor, you
> don't care if the actual current is different by several amps.  When
> measuring 20A, or especially milliamps, the reading in the meter is so
> inaccurate that it amounts to using a continuity tester to see there is
> a connection versus how many ohms there are between the 2 points of
> contact.  Both AC clamp-on meters and DC Hall effect meters measure by
> conversion, not direct readings.  Magnetic field intensity decreases
> exponentially by distance, and the huge space inside the fingers of the
> clamp means the reading changes if you reposition the wire from the
> center of the gap to closer to the wire, and it's never accurate,
> anyway.
> 
> You and I may have AC and AC/DC clamp meters, but I suspect the OP does
> not.  A good AC clamp meter at Home Depot costs $60 to $120.  A good DC
> clamp meter greatly hikes the price to $500, and up.
> 
>> I try to keep a spare PSU in the house, but it's tough. 
> 
> I tried the same using a modular PSU, so I only need to connect enough
> cables to the PSU as were needed for the particular build.  But
> eventually it got used in the next build.  A $150+ PSU makes for an
> expensive parts drawer.  For an emergency, I'll pay for overnight or
> 2-day delivery rather than buy a PSU that may never get used.  However,
> I have other computers to use, so I go with free delivery in 5 days.
> 

I have an AC/DC clamp meter with Hall Probe technology that
is every bit as good as your Harbor Freight multimeter, with
the exception it needs no electrical connection. Slide the
jaws around and make measurement.

Mine will measure the automotive starter current, which on a defective
start can be 150 amps. It will measure 20mA in the last digit, but
that digit is pretty noisy so the measurement at that level would
be one looking for activity and not a precise value. That's to give
some idea what range of measurements I've made with it. The meter measures
400 amps max, but because it's a Hall probe, nothing gets warm.

That meter, is the most expensive meter in the house, at around $300
or so. It does not get used often, but it's a handy thing to have.
Back when PCs were unsleeved, you could do a complete characterization
with the clamp-on meter. Check your Intel processor for consumption
and so on.

They make slightly more sensitive Hall Probes than that, but then the
current the sensor draws is 350mA and not a good match for a battery
operated hand held meter.

You could make the measurement with a Harbor Freight meter too.
But it would mean buying a 24 pin extension cable for ATX, cutting
the +5VSB wire in two, and inserting the meter in-circuit. Which is
why I don't suggest doing it that way. It interferes with circuit
operation, and if the wire opens, the PC goes off.

   Paul

[toc] | [prev] | [next] | [standalone]


#20646

FromVanguardLH <V@nguard.LH>
Date2024-11-22 10:20 -0600
Message-ID<8o1tit2fncq$.dlg@v.nguard.lh>
In reply to#20645
Paul <nospam@needed.invalid> wrote:

> On Fri, 11/22/2024 12:28 AM, VanguardLH wrote:
>> Paul <nospam@needed.invalid> wrote:
>> 
>>> ...
>>>
>>> 2) Use a clamp-on ammeter, to monitor the +5VSB wire on the main cable.
>> 
>> For typical clamp meters, those rely on alternating current to fluctate
>> the magnetic field around the conductor; i.e., the common clamp meter
>> just measures AC.  The 5V standby line is DC, not AC.  With the
>> 20/24-pin connector from PSU pushed onto the mobo (to account for any
>> load), use a voltmeter to measure DC voltage from the +5VSB line (pin 9,
>> purple) to a ground pin (black).
>> 
>> I can't think of any AC inside the computer excluding the input circuit
>> in the PSU.  Power into the PSU is AC.  All lines out of the PSU are DC.
>> 
>> DC also induces a magnetic field around the conductor, but not a
>> fluctuating magnetic field.  An AC clamp meter using a transformer wound
>> through the closing fingers of the clamp won't work on a DC wire.  For
>> DC, a Hall effect sensor can measure the magnetic field: the magnetic
>> field induces a current in the sensor.  However, those are inaccurate.
>> When measuring hundreds of amperes passing through a conductor, you
>> don't care if the actual current is different by several amps.  When
>> measuring 20A, or especially milliamps, the reading in the meter is so
>> inaccurate that it amounts to using a continuity tester to see there is
>> a connection versus how many ohms there are between the 2 points of
>> contact.  Both AC clamp-on meters and DC Hall effect meters measure by
>> conversion, not direct readings.  Magnetic field intensity decreases
>> exponentially by distance, and the huge space inside the fingers of the
>> clamp means the reading changes if you reposition the wire from the
>> center of the gap to closer to the wire, and it's never accurate,
>> anyway.
>> 
>> You and I may have AC and AC/DC clamp meters, but I suspect the OP does
>> not.  A good AC clamp meter at Home Depot costs $60 to $120.  A good DC
>> clamp meter greatly hikes the price to $500, and up.
>> 
>>> I try to keep a spare PSU in the house, but it's tough. 
>> 
>> I tried the same using a modular PSU, so I only need to connect enough
>> cables to the PSU as were needed for the particular build.  But
>> eventually it got used in the next build.  A $150+ PSU makes for an
>> expensive parts drawer.  For an emergency, I'll pay for overnight or
>> 2-day delivery rather than buy a PSU that may never get used.  However,
>> I have other computers to use, so I go with free delivery in 5 days.
>> 
> 
> I have an AC/DC clamp meter with Hall Probe technology that
> is every bit as good as your Harbor Freight multimeter, 

I don't buy much at Harbor Freight.  Mostly just focus on Pittsburgh
brand.  For multimeters, never bought one there.  Instead of Ames, I'd
rather spend the money on Klein at Home Depot, and the other brands at
Harbor aren't work even putting in my car's toolbox.

> Mine will measure 

Brand?  Model?  Price?

> That meter, is the most expensive meter in the house, at around $300
> or so. 

So, I wasn't far off in guessing $500, or more, for a clamp meter that
includes a Hall sensor.  Guess I'm looking for a couple extra features
as baseline that up the price.

You and I like good tools, and collect them like women collect shoes
(sorry, couldn't help using the sexist analogy).  I have oscilloscopes,
too, but I don't expect typical computer users to have many tools other
than a few hand tools they acquired for fixing doors and tables.

> You could make the measurement with a Harbor Freight meter too.
> But it would mean buying a 24 pin extension cable for ATX, cutting
> the +5VSB wire in two, and inserting the meter in-circuit. Which is
> why I don't suggest doing it that way. It interferes with circuit
> operation, and if the wire opens, the PC goes off.

That's why I figured the OP wouldn't be measuring the amp draw on 5VSB.
Really all he needs to check is there is 5VDC on the 5VSB line.  Even
the lowest grade PSUs should be able to provide enough amps for that.

[toc] | [prev] | [next] | [standalone]


#20655

From"s|b" <me@privacy.invalid>
Date2024-11-23 12:09 +0100
Message-ID<lqdrg5FnbraU1@mid.individual.net>
In reply to#20641
On Thu, 21 Nov 2024 15:58:30 -0500, Paul wrote:

> "flip the switch... standby power led on ASUS F1A75... stays on for a bit and then dies"
> 
> The power supply consists of two halves.

It's a bit too technical for me, but I appreciate the response. I'd
expect the PSU to either work or not. Is the behaviour that I'm
experiencing possible with a broken PSU?

I'm going to replace the thermal paste of the CPU first and see what
happens.

-- 
s|b

[toc] | [prev] | [next] | [standalone]


#20656

FromPaul <nospam@needed.invalid>
Date2024-11-23 07:07 -0500
Message-ID<vhsgi8$1mso2$1@dont-email.me>
In reply to#20655
On Sat, 11/23/2024 6:09 AM, s|b wrote:
> On Thu, 21 Nov 2024 15:58:30 -0500, Paul wrote:
> 
>> "flip the switch... standby power led on ASUS F1A75... stays on for a bit and then dies"
>>
>> The power supply consists of two halves.
> 
> It's a bit too technical for me, but I appreciate the response. I'd
> expect the PSU to either work or not. Is the behaviour that I'm
> experiencing possible with a broken PSU?
> 
> I'm going to replace the thermal paste of the CPU first and see what
> happens.
> 

Just change the PSU and retest :-)
I've done that a few times, as a first step.

*******

Never hurts to change thermal paste. Especially the hobby non-permanent kind.

Some of the paste materials are good enough to be permanently used.
The screened thick materials provided initially on heatsinks
can be quite good.

Hobby pastes vary in endurance. One (which I won't promote by
naming it), reacts with the metal in the CPU and the metal in the
heatsink. That's an example of a bad choice, as some people have
experienced (no warranty support when that damage is detected!).

Some of the others, are well designed pastes, with the right
kind of viscosity, but if you "abuse" the PC, by dropping or
kicking it, that affects the paste integrity (air ingress), and
re-pasting those is a good idea. Like, just handling the heavy
heatsink assembly, applying too much "torque" to it while working
on a PC, can affect it by a few degrees of performance. It's because
the huge heatsinks weigh so much, that this kind of thing happens.
A light heatsink has sufficient normal force to remain in contact.

   Paul

[toc] | [prev] | [next] | [standalone]


#20661

From"s|b" <me@privacy.invalid>
Date2024-11-24 17:12 +0100
Message-ID<lqh1jaF8n4tU2@mid.individual.net>
In reply to#20656
On Sat, 23 Nov 2024 07:07:35 -0500, Paul wrote:

> Just change the PSU and retest :-)
> I've done that a few times, as a first step.

Yes, I just read VanguardLH's advice, but it's too late; i already
removed the paste from the CPU and heatsink.

> Never hurts to change thermal paste. Especially the hobby non-permanent kind.

Is Arctic MX-4 any good?

-- 
s|b

[toc] | [prev] | [next] | [standalone]


#20662

FromPaul <nospam@needed.invalid>
Date2024-11-24 15:21 -0500
Message-ID<vi01sp$2ckbh$1@dont-email.me>
In reply to#20661
On Sun, 11/24/2024 11:12 AM, s|b wrote:
> On Sat, 23 Nov 2024 07:07:35 -0500, Paul wrote:
> 
>> Just change the PSU and retest :-)
>> I've done that a few times, as a first step.
> 
> Yes, I just read VanguardLH's advice, but it's too late; i already
> removed the paste from the CPU and heatsink.
> 
>> Never hurts to change thermal paste. Especially the hobby non-permanent kind.
> 
> Is Arctic MX-4 any good?
> 

You can see they have a lot of similarities. In terms of slump
and texture. A lot of these look like the AS-5 I use right now.

https://www.tweaktown.com/image.php?image=https://static.tweaktown.com/content/3/3/3383_04_full.jpg

No, you don't generally spread them with a credit card any more.
Some of the products are "crumbly" and the reputation is they
are hard to apply, which is probably quite true. The others
are more gel-like, and when crushed by the load of the heatsink,
they spread quite nicely.

You can tell this, by doing a calibration fitting. Say the instructions
tell you to install "five dots".

        x       x

            X

        x       x

You can apply your heatsink for a test, press down, then remove
the heatsink and look at how well it spread out. This also functions
as your "tinting cycle", as you'll be wiping that off and cleaning
most all of it off, and the grooves in the metal will have some
of the new paste in them. I don't leave a lot of material during
the cleaning, in order to not apply too much later.

Then, when you do the real installation, you will know whether the
dots should have been bigger or not

        X       X

           ***

        X       X

You don't want too much shooting out of the gap,
as then it's a mess to clean up later if it gets
into the wrong places on the motherboard.

I have a tube of AS-5 here, and it's suitable for
applying dots (according to what the manufacturer
recommends for each processor type), then just seat
the heatsink in it and tighten it up. Some processors
only use a central dot. Some processors are convex,
some processors are flat on top, which is why the
dot pattern recommendation varies.

I like screw fittings and permanently attached back plates.
Some of the solutions that "come in the box" with motherboards,
they fall off when unscrewed, then you have to remove the
motherboard from the computer, apply paste, assemble,
and insert the reassembled mobo+HS into the case. This is
a major pain. And something they don't tell you, is you
can SCRAPE SMALL CAPS off the board when reinserting the
board into the computer case.

It is for this reason, that I recommend third party HS
assemblies which allow removal and re-seating of the
heatsink, without having to pull the motherboard out
of the case. As long as the manufacturers use 1x2 cellphone
capacitors on the motherboard surface, you have to be a lot
more careful to not damage the surface mount components.
Your motherboard is an older generation and might have
0603 or 0402 caps and the older components are a bit
bigger and harder to damage without a bit more effort.

*******

The MX-4 has the ability to spread easily. The temperature
different generally is only a couple degrees C between
competitors. But when a compound spreads (these are not
phase change compounds like the screened ones that come
with the heatsink are), you may have to redo it five years
from now. The temperature rise will tell you it is time to change it.

https://www.tweaktown.com/reviews/3383/arctic_cooling_mx_4_thermal_compound/index.html

The problem with getting, say, a new heatsink for a build,
is you ruin the screened material while having trouble
fitting the screws. I've had trouble tightening the screws
on any of the last four (third-party) heatsinks, and none of
them had the screw threads "catch" on the first try. I was
a lather of sweat, lots of cursing, before the damn screws
were finally fitted. And that might have been the third time
I applied paste. This is why, if a screened material is placed
on the brand new heatsink, you hardly get to keep that compound
and have to use your tube of MX-4 instead. Fitting these things
is a nuisance.

   Paul

[toc] | [prev] | [next] | [standalone]


#20666

From"s|b" <me@privacy.invalid>
Date2024-11-25 14:58 +0100
Message-ID<lqje4oFkfg9U1@mid.individual.net>
In reply to#20662
On Sun, 24 Nov 2024 15:21:47 -0500, Paul wrote:

> No, you don't generally spread them with a credit card any more.

Some versions of the MX-4 come with a spatula. This guy seems to be
doing it right: <https://youtu.be/eyNTXDoLyEU?feature=shared>

> You don't want too much shooting out of the gap,
> as then it's a mess to clean up later if it gets
> into the wrong places on the motherboard.

The guy in the YT spilled on a few places, but he seems to be alright.
In comparison to others he uses very little paste, seems even less than
the 'pea' a lot of people advise.
 
> I have a tube of AS-5 here, and it's suitable for
> applying dots (according to what the manufacturer
> recommends for each processor type), then just seat
> the heatsink in it and tighten it up. Some processors
> only use a central dot. Some processors are convex,
> some processors are flat on top, which is why the
> dot pattern recommendation varies.

Good to know.
 
> It is for this reason, that I recommend third party HS
> assemblies which allow removal and re-seating of the
> heatsink, without having to pull the motherboard out
> of the case.

I can remove the heatsink without a problem.

> As long as the manufacturers use 1x2 cellphone
> capacitors on the motherboard surface, you have to be a lot
> more careful to not damage the surface mount components.
> Your motherboard is an older generation and might have
> 0603 or 0402 caps and the older components are a bit
> bigger and harder to damage without a bit more effort.

That's a good thing, right? (-:

> The MX-4 has the ability to spread easily. The temperature
> different generally is only a couple degrees C between
> competitors. But when a compound spreads (these are not
> phase change compounds like the screened ones that come
> with the heatsink are), you may have to redo it five years
> from now. The temperature rise will tell you it is time to change it.

I'm making a note of that. I bought the PC in 2012, so there's a good
chance it's the CPU and not the PSU that causes the shutdown?
 
> https://www.tweaktown.com/reviews/3383/arctic_cooling_mx_4_thermal_compound/index.html
> 
> The problem with getting, say, a new heatsink for a build,
> is you ruin the screened material while having trouble
> fitting the screws. I've had trouble tightening the screws
> on any of the last four (third-party) heatsinks, and none of
> them had the screw threads "catch" on the first try. I was
> a lather of sweat, lots of cursing, before the damn screws
> were finally fitted. And that might have been the third time
> I applied paste. This is why, if a screened material is placed
> on the brand new heatsink, you hardly get to keep that compound
> and have to use your tube of MX-4 instead. Fitting these things
> is a nuisance.

The heatsink I have works with two clamps (?) which have to be attached
left and right and then you pull a lever to tighten it to the CPU.

-- 
s|b

[toc] | [prev] | [next] | [standalone]


#20669

FromPaul <nospam@needed.invalid>
Date2024-11-25 17:33 -0500
Message-ID<vi2tv2$321ei$1@dont-email.me>
In reply to#20666
On Mon, 11/25/2024 8:58 AM, s|b wrote:

> I'm making a note of that. I bought the PC in 2012, so there's a good
> chance it's the CPU and not the PSU that causes the shutdown?
>  

Unlikely.

If a heatsink falls off the CPU, the temperature shoots up to 200C,
the thermal protection does not engage immediately, that
could damage a CPU.

But really, with the heatsink fastened to the top, there is
thermal inertia (mass) and the temperature does not shoot up
quite as rapidly.

It's more likely a capacitor is leaking near the CPU socket,
if the board won't start.

But your case, your initial symptoms were an overloaded +5VSB.
I suggested measuring the current, as one way to ascertain
if there is a specific problem with the combination of
components. If something is shorted on the motherboard,
then the problem (box shuts off), may still be present
with your spare supply installed. You would put in the
spare supply. The LED would be lit for a minute or two,
then go off on its own. That could be an overload of +5VSB
for example.

Modern motherboards (yours is modern enough), have THERMTRIP.
If the CPU goes over 100C to 120C or so, THERMTRIP causes
PSON# to be deasserted and the PC fans stop spinning, and
3.3V/5V/12V are then powered off. You can have a shutdown
due to a thermal issue. The Asus LED remains lit when that happens
(a THERMTRIP).

But the thing is, you told us that the Asus LED, which
monitors +5VSB, that went OFF. That's an indication of
a power supply problem, not a THERMTRIP. THERMTRIP will
not turn that LED off.

The LED then, is key to determining whether you have
a "NEW" or "DIFFERENT" problem than the first one reported.

1) Swap out power supply. Put in known-good supply.
2) Record symptoms again. Does PC shut off as before ?
   Does the Asus LED stay lit after the erroneous shutdown ?

Report back.

   Paul

[toc] | [prev] | [next] | [standalone]


#20671

From"s|b" <me@privacy.invalid>
Date2024-11-26 18:44 +0100
Message-ID<lqmfomF50lhU1@mid.individual.net>
In reply to#20669
On Mon, 25 Nov 2024 17:33:05 -0500, Paul wrote:


> > I'm making a note of that. I bought the PC in 2012, so there's a good
> > chance it's the CPU and not the PSU that causes the shutdown?

> Unlikely.

)-:
 
> If a heatsink falls off the CPU, the temperature shoots up to 200C,
> the thermal protection does not engage immediately, that
> could damage a CPU.
> 
> But really, with the heatsink fastened to the top, there is
> thermal inertia (mass) and the temperature does not shoot up
> quite as rapidly.

That's disappointing, sort of. But I do get to play with the paste.
 
> It's more likely a capacitor is leaking near the CPU socket,
> if the board won't start.

Well then it's removing the hard drives and dump the tower at a recycle
point! If I'll bring it to a computer repair store they will charge me
an arm and a leg.
 
> But your case, your initial symptoms were an overloaded +5VSB.
> I suggested measuring the current, as one way to ascertain
> if there is a specific problem with the combination of
> components. If something is shorted on the motherboard,
> then the problem (box shuts off), may still be present
> with your spare supply installed. You would put in the
> spare supply. The LED would be lit for a minute or two,
> then go off on its own. That could be an overload of +5VSB
> for example.

I'm learning new things here. It's my brother who knows electricity; I'm
100% sure he's got the equipment to measure.

I checked the PSU that I hopefully can use:

AOpen
Model no: Z400-08FC (400W peak)
 
> Modern motherboards (yours is modern enough), have THERMTRIP.
> If the CPU goes over 100C to 120C or so, THERMTRIP causes
> PSON# to be deasserted and the PC fans stop spinning, and
> 3.3V/5V/12V are then powered off. You can have a shutdown
> due to a thermal issue. The Asus LED remains lit when that happens
> (a THERMTRIP).
> 
> But the thing is, you told us that the Asus LED, which
> monitors +5VSB, that went OFF. That's an indication of
> a power supply problem, not a THERMTRIP. THERMTRIP will
> not turn that LED off.

OK, tnx. Does this also fit in the fact that I sometimes get to the BIOS
(however briefly it is) ?
 
> The LED then, is key to determining whether you have
> a "NEW" or "DIFFERENT" problem than the first one reported.
> 
> 1) Swap out power supply. Put in known-good supply.
> 2) Record symptoms again. Does PC shut off as before ?
>    Does the Asus LED stay lit after the erroneous shutdown ?
> 
> Report back.

Will do when you confirm I can use the other PSU. It's ATX, so I'm
guessing yes?

-- 
s|b

[toc] | [prev] | [next] | [standalone]


#20672

FromVanguardLH <V@nguard.LH>
Date2024-11-26 16:31 -0600
Message-ID<1hjl5k1atqw3s.dlg@v.nguard.lh>
In reply to#20671
s|b <me@privacy.invalid> wrote:

> Well then it's removing the hard drives and dump the tower at a recycle
> point! If I'll bring it to a computer repair store they will charge me
> an arm and a leg.

If you're removing the components from the case leaving just a metal
shell for the case, that can go in your recycle bin along with all the
cans for food (after rinsing them).  Just remove any screws since those
are hardened, and not recyclable.  If there are plastic covers, like
over the front, remove those to dump in the trash.

The PCBs have lead for the solder.  Those, and other components with
hazardous materials can be dropped off at the local hazard waste site.
They charge me $10 per load no matter how much I bring in (so I amass a
bunch of stuff before going) except for special items, like tires and
big heavy bulky stuff.  Don't know if you have any hazardous waste
collection locations.  Often they are run by the municipality or
counties.

[toc] | [prev] | [next] | [standalone]


#20675

From"s|b" <me@privacy.invalid>
Date2024-11-27 15:38 +0100
Message-ID<lqop6hFg8m9U1@mid.individual.net>
In reply to#20672
On Tue, 26 Nov 2024 16:31:52 -0600, VanguardLH wrote:

> If you're removing the components from the case leaving just a metal
> shell for the case, that can go in your recycle bin along with all the
> cans for food (after rinsing them).  Just remove any screws since those
> are hardened, and not recyclable.  If there are plastic covers, like
> over the front, remove those to dump in the trash.

In Belgium we have Recupel. I can either bring it back to the place that
sold it to me or I can bring it to a 'recyclagepark' and they're
responsible for the recycling. I'll do the latter.

-- 
s|b

[toc] | [prev] | [next] | [standalone]


#20674

FromPaul <nospam@needed.invalid>
Date2024-11-26 21:04 -0500
Message-ID<vi5umi$3n0hl$1@dont-email.me>
In reply to#20671
On Tue, 11/26/2024 12:44 PM, s|b wrote:
> On Mon, 25 Nov 2024 17:33:05 -0500, Paul wrote:
> 
> 
>>> I'm making a note of that. I bought the PC in 2012, so there's a good
>>> chance it's the CPU and not the PSU that causes the shutdown?
> 
>> Unlikely.
> 
> )-:
>  
>> If a heatsink falls off the CPU, the temperature shoots up to 200C,
>> the thermal protection does not engage immediately, that
>> could damage a CPU.
>>
>> But really, with the heatsink fastened to the top, there is
>> thermal inertia (mass) and the temperature does not shoot up
>> quite as rapidly.
> 
> That's disappointing, sort of. But I do get to play with the paste.
>  
>> It's more likely a capacitor is leaking near the CPU socket,
>> if the board won't start.
> 
> Well then it's removing the hard drives and dump the tower at a recycle
> point! If I'll bring it to a computer repair store they will charge me
> an arm and a leg.
>  
>> But your case, your initial symptoms were an overloaded +5VSB.
>> I suggested measuring the current, as one way to ascertain
>> if there is a specific problem with the combination of
>> components. If something is shorted on the motherboard,
>> then the problem (box shuts off), may still be present
>> with your spare supply installed. You would put in the
>> spare supply. The LED would be lit for a minute or two,
>> then go off on its own. That could be an overload of +5VSB
>> for example.
> 
> I'm learning new things here. It's my brother who knows electricity; I'm
> 100% sure he's got the equipment to measure.
> 
> I checked the PSU that I hopefully can use:
> 
> AOpen
> Model no: Z400-08FC (400W peak)
>  
>> Modern motherboards (yours is modern enough), have THERMTRIP.
>> If the CPU goes over 100C to 120C or so, THERMTRIP causes
>> PSON# to be deasserted and the PC fans stop spinning, and
>> 3.3V/5V/12V are then powered off. You can have a shutdown
>> due to a thermal issue. The Asus LED remains lit when that happens
>> (a THERMTRIP).
>>
>> But the thing is, you told us that the Asus LED, which
>> monitors +5VSB, that went OFF. That's an indication of
>> a power supply problem, not a THERMTRIP. THERMTRIP will
>> not turn that LED off.
> 
> OK, tnx. Does this also fit in the fact that I sometimes get to the BIOS
> (however briefly it is) ?
>  
>> The LED then, is key to determining whether you have
>> a "NEW" or "DIFFERENT" problem than the first one reported.
>>
>> 1) Swap out power supply. Put in known-good supply.
>> 2) Record symptoms again. Does PC shut off as before ?
>>    Does the Asus LED stay lit after the erroneous shutdown ?
>>
>> Report back.
> 
> Will do when you confirm I can use the other PSU. It's ATX, so I'm
> guessing yes?
> 

AOpen  Z400-08FC (400W peak)   [Numbers are different than the 380D supply]

    5V  3.3V  12V1  12V2  -12V  +5VSB    [ no  PCIe  1 x 6-pin ? ]

   20A  20A   10A    13A   0.3A   2.5A
   \       /  \        /
     130W      23A common
   combined     source

12V2 is CPU
12V1 is peripherals, PCIe slots

"ASUS F1A75" it is probably the 3870K entry below.

A guess would be, it can't be over about 100W or 12V @ 10A to allow some
efficiency room. The Z400 has 13A available for the 10A load, which is
pretty full.

https://www.asus.com/me-en/supportonly/f1a75-v%20pro/helpdesk_cpu/         <=== check the table, your CPU
                                                                                is in that table
   AMD A8-3870K (AD3870WNZ43GX, rev.B0, 3.0GHz, QC, L2:4M, HD6550D, 100W)

The 12V1 has 10A for things like a video card (if present). If
you took the video output of your APU (CPU has GPU inside), then
the biggest drain after that would be HDD and ODD, and 10A is
enough for that (12V @ 1.5A ODD, 12V @ 2A HDD).

The AOpen is enough for a basic PC.

CPU: AMD-A8-3870 BLACK EDITION WITH RADEON HD 6550D
Tower/PSU: ANTEC-NSK4482B-EC 380 WATT
Graphic card: AMD-AP38G160U2K  <=============================== I can find no info for this.
SSD: INTEL-SSDSC2CW120A310 120GB 520 SERIES 34NM SATA III2 .
HDD: WESTERN DIGITAL-WD5000AAKX 500GB 16MB SATA600 7200RPM
CD/DVD (re)writer: LITE-ON-IHAS122-18 SATA 22X DVD-RW

I'm not getting a match on your video card.
Both search engines aren't matching on "AP38G160U2K"

The graphics function inside the 100W APU is HD6550D.

The thing is, your Peripheral supply 12V1 is only 10 amps,
and that's not going to be enough for much of a standalone
video card in an expansion slot.

The motherboard has four video connectors on the plate.
It has DVI-D, VGA, HDMI, DisplayPort. These run off the APU (CPU+GPU) chip
and so to make the PC work, you don't even need a separate video card.

   https://www.legitreviews.com/images/reviews/1721/asus-f1a75-v-pro-layout-6.jpg

Normally, if you had purchased a separate video card, it would
go in the long, dark-blue slot on the left of this picture.

   https://www.legitreviews.com/images/reviews/1721/asus-f1a75-v-pro-layout-5.jpg

  ( https://www.legitreviews.com/asus-f1a75-v-pro-motherboard-review_1721/3 )

The thing is, the AOpen supply is unlikely to have a PCI Express 2x3 power
connector on it. And a small video card might need a power source like that.
That's why knowing some details about what you are actually
using for video is important.

Next to the long, dark-blue4 slot on the left of the picture,
is the round CMOS battery CR2032.

Using your multimeter, you can connect the black lead to the chassis metal
(the VGA screw would do) and the red lead to the top of the CMOS cell, and
that should read +3.0 volts on the 20V scale of the meter, if the cell was new.
If the PC has been sitting in a closet for more than 3 years, then
the CMOS battery will be flat, and you will need to purchase a
replacement CR2032 to put in the round socket. I have replaced a lot
of these here, because the PCs sitting in the junk room, wear that
battery down quickly. Only a few PCs will not start, unless that
battery has at least 2.0V on it. The battery is a bit higher than
3.0V when new.

The motherboard will not "remember" the BIOS settings you enter, unless
the battery is functioning. The clock for example, will show the wrong time.

Using your debug technique of "swap it out", you do not need to
replace the CR2032 right away. However, if you fit the Z400 supply,
and the Asus LED lights up, but pressing the power button does not
work, then replacing the CR2032 would be the next step.

*******

summary: Please examine the contents of the PC and confirm how your
         video works. Is there an add-in card in one of the long slots ?
         Or are you using a video connector in the group-of-four connectors
         in the Legitreviews picture ? It you are using a separate video card,
         there may not be enough power for it, or a connector to feed the
         card power.

   Paul

[toc] | [prev] | [next] | [standalone]


#20677

From"s|b" <me@privacy.invalid>
Date2024-11-27 16:01 +0100
Message-ID<lqoqhtFgf4kU1@mid.individual.net>
In reply to#20674
On Tue, 26 Nov 2024 21:04:00 -0500, Paul wrote:

> Graphic card: AMD-AP38G160U2K  <=============================== I can find no info for this.

That's my fault; these are the 2 DDR3 modules. This is the graphic card:

ASUS-EAH5450 SILENT/DI/ 512MB GDDR PCI EXPRESS

> I'm not getting a match on your video card.
> Both search engines aren't matching on "AP38G160U2K"

<https://www.startpage.com/do/dsearch?q=AP38G1608U2K&cat=web&language=english>

<https://tweakers.net/pricewatch/302780/amd-performance-ap38g1608u2k/specificaties/>
 
> The graphics function inside the 100W APU is HD6550D.

I know. I should have used that one and the shop should have mentioned
it to me, but I got a PCIe ienstead.
 
> The thing is, your Peripheral supply 12V1 is only 10 amps,
> and that's not going to be enough for much of a standalone
> video card in an expansion slot.

So better  not try then.

> The motherboard has four video connectors on the plate.
> It has DVI-D, VGA, HDMI, DisplayPort. These run off the APU (CPU+GPU) chip
> and so to make the PC work, you don't even need a separate video card.
> 
>    https://www.legitreviews.com/images/reviews/1721/asus-f1a75-v-pro-layout-6.jpg

See above. I got the card that I didn't really need. I don't play games
on PC or anything.

> Normally, if you had purchased a separate video card, it would
> go in the long, dark-blue slot on the left of this picture.
> 
>    https://www.legitreviews.com/images/reviews/1721/asus-f1a75-v-pro-layout-5.jpg

That's where it is.

> The thing is, the AOpen supply is unlikely to have a PCI Express 2x3 power
> connector on it. And a small video card might need a power source like that.
> That's why knowing some details about what you are actually
> using for video is important.

My fault, I'm sorry. This shoul d be it:

ASUS-EAH5450 SILENT/DI/ 512MB GDDR PCI EXPRESS
 
> Next to the long, dark-blue4 slot on the left of the picture,
> is the round CMOS battery CR2032.

Yeah, I replaced that already and then did a BIOS reset by replacing the
jumper (and then back).
 
> The motherboard will not "remember" the BIOS settings you enter, unless
> the battery is functioning. The clock for example, will show the wrong time.

These things I'm familiar with. Long ago I did a course 'computer
hardware'. Nothing fancy, just the basics, enough to build your own PC.
I can do that, although I did it only twice. It's just when things
(hardware) start breaking down I lack the knowledge.
 
> Using your debug technique of "swap it out", you do not need to
> replace the CR2032 right away. However, if you fit the Z400 supply,
> and the Asus LED lights up, but pressing the power button does not
> work, then replacing the CR2032 would be the next step.

I checked ealier: the PSU has an on/off button at the back. If I switch
it on the LED lights up, but only for a short time. After it switches
off nothing happens when I push the button of the tower.

> summary: Please examine the contents of the PC and confirm how your
>          video works. Is there an add-in card in one of the long slots ?
>          Or are you using a video connector in the group-of-four connectors
>          in the Legitreviews picture ? It you are using a separate video card,
>          there may not be enough power for it, or a connector to feed the
>          card power.

Again, tnx for taking your time. This is the graphic card (that I didn't
really need :-) :

ASUS-EAH5450 SILENT/DI/ 512MB GDDR PCI EXPRESS

-- 
s|b

[toc] | [prev] | [next] | [standalone]


#20679

FromPaul <nospam@needed.invalid>
Date2024-11-27 10:19 -0500
Message-ID<vi7daq$2ars$1@dont-email.me>
In reply to#20677
On Wed, 11/27/2024 10:01 AM, s|b wrote:
> On Tue, 26 Nov 2024 21:04:00 -0500, Paul wrote:
> 
>> Graphic card: AMD-AP38G160U2K  <=============================== I can find no info for this.
> 
> That's my fault; these are the 2 DDR3 modules. This is the graphic card:
> 
> ASUS-EAH5450 SILENT/DI/ 512MB GDDR PCI EXPRESS
> 
>> I'm not getting a match on your video card.
>> Both search engines aren't matching on "AP38G160U2K"
> 
> <https://www.startpage.com/do/dsearch?q=AP38G1608U2K&cat=web&language=english>
> 
> <https://tweakers.net/pricewatch/302780/amd-performance-ap38g1608u2k/specificaties/>
>  
>> The graphics function inside the 100W APU is HD6550D.
> 
> I know. I should have used that one and the shop should have mentioned
> it to me, but I got a PCIe ienstead.
>  
>> The thing is, your Peripheral supply 12V1 is only 10 amps,
>> and that's not going to be enough for much of a standalone
>> video card in an expansion slot.
> 
> So better  not try then.
> 
>> The motherboard has four video connectors on the plate.
>> It has DVI-D, VGA, HDMI, DisplayPort. These run off the APU (CPU+GPU) chip
>> and so to make the PC work, you don't even need a separate video card.
>>
>>    https://www.legitreviews.com/images/reviews/1721/asus-f1a75-v-pro-layout-6.jpg
> 
> See above. I got the card that I didn't really need. I don't play games
> on PC or anything.
> 
>> Normally, if you had purchased a separate video card, it would
>> go in the long, dark-blue slot on the left of this picture.
>>
>>    https://www.legitreviews.com/images/reviews/1721/asus-f1a75-v-pro-layout-5.jpg
> 
> That's where it is.
> 
>> The thing is, the AOpen supply is unlikely to have a PCI Express 2x3 power
>> connector on it. And a small video card might need a power source like that.
>> That's why knowing some details about what you are actually
>> using for video is important.
> 
> My fault, I'm sorry. This shoul d be it:
> 
> ASUS-EAH5450 SILENT/DI/ 512MB GDDR PCI EXPRESS
>  
>> Next to the long, dark-blue4 slot on the left of the picture,
>> is the round CMOS battery CR2032.
> 
> Yeah, I replaced that already and then did a BIOS reset by replacing the
> jumper (and then back).
>  
>> The motherboard will not "remember" the BIOS settings you enter, unless
>> the battery is functioning. The clock for example, will show the wrong time.
> 
> These things I'm familiar with. Long ago I did a course 'computer
> hardware'. Nothing fancy, just the basics, enough to build your own PC.
> I can do that, although I did it only twice. It's just when things
> (hardware) start breaking down I lack the knowledge.
>  
>> Using your debug technique of "swap it out", you do not need to
>> replace the CR2032 right away. However, if you fit the Z400 supply,
>> and the Asus LED lights up, but pressing the power button does not
>> work, then replacing the CR2032 would be the next step.
> 
> I checked ealier: the PSU has an on/off button at the back. If I switch
> it on the LED lights up, but only for a short time. After it switches
> off nothing happens when I push the button of the tower.
> 
>> summary: Please examine the contents of the PC and confirm how your
>>          video works. Is there an add-in card in one of the long slots ?
>>          Or are you using a video connector in the group-of-four connectors
>>          in the Legitreviews picture ? It you are using a separate video card,
>>          there may not be enough power for it, or a connector to feed the
>>          card power.
> 
> Again, tnx for taking your time. This is the graphic card (that I didn't
> really need :-) :
> 
> ASUS-EAH5450 SILENT/DI/ 512MB GDDR PCI EXPRESS
> 

The HD5450 is the same as an HD6450 (I have that one) and
it is a 13 watt maximum card. You will nave no trouble
driving that, as that's a low power card suited to media center
applications. I have that card in my Dell Optiplex, as the card
has no power connector and runs off slot power.

The CPU is 10A on a 13A output, which is fine. Because
you haven't used up all your Peripheral power, there will be
sufficient shared amperes, to not be a problem.

The HD5450 is anywhere between 3 watts (idle) and 13 watts.
You can't do much better than that, on a discrete card.
And that card of mine, works in Win10 22H2. The WDM driver
for it, is "good enough" for the job.

*******

Now, let's check my memory.

Listed as 19 watts.

https://www.techpowerup.com/gpu-specs/radeon-hd-5450.c503

Mine is 18 watts.

https://www.techpowerup.com/gpu-specs/radeon-hd-6450.c402

I don't know if Furmark will run on that, for test.

It's not a closed loop card, so if you can find a program
to drive is hard enough, it might draw 20 watts :-)

   Paul

[toc] | [prev] | [next] | [standalone]


#20657

FromVanguardLH <V@nguard.LH>
Date2024-11-23 14:28 -0600
Message-ID<1vgrm5ur0t99h$.dlg@v.nguard.lh>
In reply to#20655
s|b <me@privacy.invalid> wrote:

> On Thu, 21 Nov 2024 15:58:30 -0500, Paul wrote:
> 
>> "flip the switch... standby power led on ASUS F1A75... stays on for a bit and then dies"
>> 
>> The power supply consists of two halves.
> 
> It's a bit too technical for me, but I appreciate the response. I'd
> expect the PSU to either work or not. Is the behaviour that I'm
> experiencing possible with a broken PSU?
> 
> I'm going to replace the thermal paste of the CPU first and see what
> happens.

Make sure to use non-metallic paste.  Use ceramic.  Since you're not
overclocking, you won't achieve a one to three degree decrease in
temperature, anyway.  Not overclocking already reduced heat generation.
Also, make sure it is thermal PASTE, not thermal adhesive which is an
epoxy.  You might even go with a thermal strip aka thermal pad for less
mess.  Although not as efficient as paste, strips will work well enough,
especially in non-overclocking scenarios.  Way too many users gob on too
much paste resulting is less than an ideal mating between CPU plate and
heatsink. More is NOT better.

You'll need a piece of mylar or other thin plastic to apply the paste.
A credit card is too rough on its edges.  You want a shear coat, not an
opaque coat.  After pressing the heatsink to the CPU plate, and before
clamping down the heatsink, rotate the heatsink to ensure the paste gets
distributed across the CPU plate.  You can watch Youtube videos to get
techniques and suggestions.  However, even this guy put on way too much:
https://www.youtube.com/watch?v=m7x2sUt0mqo.  Yet his moniker claims he
is from Arctic which is the pervasive leader in after-market pastes.
About a quarter of what he applied is needed.  Of course, when using the
plastic plate to watch the spread, probably about half of the paste
ended up on the plastic plate after removal.  Here's another guy gobbing
on way too much thermal paste: https://youtu.be/a0LlMDk_ljQ?t=12.

I've seen many techniques suggested for applying the correct amount of
thermal paste.  One is to use a piece of thin mylar to spread around the
paste to coat a thin layer you can see through.  Another is to put one
drop in the middle of the CPU plate, or 5 tiny drops in the center and
at midway to the corners, to push down the heatsink to rotate and spread
the paste.  If any oozes out paste the CPU plate, you put on too much
which also means there is likely too much between CPU and heatsink.

You'll need isopropyl alcohol and a plastic scraper to remove the old
paste.  Make sure the CPU plate and heatsink are devoid of any old paste
or thermal pad, and are clean.  After cleaning, keep your fingers off.

Considering your lack of electronic test equipment, I suspect paste
might not be the way for you to go.  A thermal pad gets squeezed between
CPU plate and heatsink, won't ooze out, and liquifies (melts) under
pressure and heat.  Considering how many users apply way too much paste,
and if not overclocking, a pad is just as good.  I've never gotten into
pads, but https://www.youtube.com/watch?v=2BhKx0iQ4K8 and
https://www.youtube.com/watch?v=niAQs8dZohE look interesting.  Despite
the claim reusability of the graphic pad, I'd never reuse one.  I'd
replace it just I would replace paste.  I'm not building and rebuilding
hundreds of boxes.

Personally I'd go with the PSU substitution suggestion first.  The load
a PSU can handle with proper regulation varies with age, and crappy PSUs
age faster.  Replacing thermal paste isn't a job for a newbie.  I always
lap the CPU plate and heatsink to get them as flat as possible for best
mating since metal-to-metal contact is by far the best method for
thermal transfer.  However, rare few users lap their CPU and heatsink.
I don't overclock, but do want thermal transfer to work best, and want
the least amount of paste since it also ages.  

Although the focus has been on the paste, another facet of concern is if
the clamping of the heatsink is even.  You don't want to cinch down one
side more than the other.  That will squeeze out more paste on one side,
but leave too much on the side with less pressure.  You want paste to
substitute for air, not thermal paste to be the primary transfer media.

[toc] | [prev] | [next] | [standalone]


#20660

From"s|b" <me@privacy.invalid>
Date2024-11-24 17:10 +0100
Message-ID<lqh1fkF8n4tU1@mid.individual.net>
In reply to#20657
On Sat, 23 Nov 2024 14:28:54 -0600, VanguardLH wrote:

> > I'm going to replace the thermal paste of the CPU first and see what
> > happens.
 
> Make sure to use non-metallic paste.  Use ceramic.  Since you're not
> overclocking, you won't achieve a one to three degree decrease in
> temperature, anyway.  Not overclocking already reduced heat generation.
> Also, make sure it is thermal PASTE, not thermal adhesive which is an
> epoxy.

I already ordered Arctic MX-4. What can I say? I'm an impulsive buyer.
(-:

> You'll need a piece of mylar or other thin plastic to apply the paste.
> A credit card is too rough on its edges.  You want a shear coat, not an
> opaque coat.

I'm hardly an expert, I just saw a YT trying one dot, 5 dot, cross and
then some tool to spread the paste, like what you're saying. Some guy
commented that one dot is the best and that using a tool causes air
bubbles. (?)

> I've seen many techniques suggested for applying the correct amount of
> thermal paste.  One is to use a piece of thin mylar to spread around the
> paste to coat a thin layer you can see through.  Another is to put one
> drop in the middle of the CPU plate, or 5 tiny drops in the center and
> at midway to the corners, to push down the heatsink to rotate and spread
> the paste.  If any oozes out paste the CPU plate, you put on too much
> which also means there is likely too much between CPU and heatsink.

I think I'm going to go with the one dot method while keeping in mind:
less is more. I'll practice on a napkin first.
 
> You'll need isopropyl alcohol and a plastic scraper to remove the old
> paste.  Make sure the CPU plate and heatsink are devoid of any old paste
> or thermal pad, and are clean.  After cleaning, keep your fingers off.

Done.
 
> Considering your lack of electronic test equipment, I suspect paste
> might not be the way for you to go.  A thermal pad gets squeezed between
> CPU plate and heatsink, won't ooze out, and liquifies (melts) under
> pressure and heat.  Considering how many users apply way too much paste,
> and if not overclocking, a pad is just as good.  I've never gotten into
> pads, but https://www.youtube.com/watch?v=2BhKx0iQ4K8 and
> https://www.youtube.com/watch?v=niAQs8dZohE look interesting.  Despite
> the claim reusability of the graphic pad, I'd never reuse one.  I'd
> replace it just I would replace paste.  I'm not building and rebuilding
> hundreds of boxes.

One of the few computers I assembled myself had AMD CPU's with a
heatsink that had such a thermal pad. I'm just going to with the paste.
It's not like it's a brand new computer.
 
> Personally I'd go with the PSU substitution suggestion first.

So you're conspiring with Paul eh? ;-) I already removed the paste, so
I'll have to replace that first. If that doesn't work I'll come back to
ask if the even older PSU that I've got is compatible.

>  The load
> a PSU can handle with proper regulation varies with age, and crappy PSUs
> age faster.  Replacing thermal paste isn't a job for a newbie.  I always
> lap the CPU plate and heatsink to get them as flat as possible for best
> mating since metal-to-metal contact is by far the best method for
> thermal transfer.  However, rare few users lap their CPU and heatsink.
> I don't overclock, but do want thermal transfer to work best, and want
> the least amount of paste since it also ages.  

I understand your concern, but it's not a new computer. For me it's more
for the fun of it. If I fail, then I fail, but I will still have learned
something and that's worth something (for me anyway).
 
> Although the focus has been on the paste, another facet of concern is if
> the clamping of the heatsink is even.  You don't want to cinch down one
> side more than the other.  That will squeeze out more paste on one side,
> but leave too much on the side with less pressure.  You want paste to
> substitute for air, not thermal paste to be the primary transfer media.

Thanks for the advice. I've been practicing on the clamping of the
heatsink.

-- 
s|b

[toc] | [prev] | [next] | [standalone]


#20663

FromPaul <nospam@needed.invalid>
Date2024-11-24 15:27 -0500
Message-ID<vi027t$2cm1m$1@dont-email.me>
In reply to#20660
On Sun, 11/24/2024 11:10 AM, s|b wrote:

> 
> I think I'm going to go with the one dot method while keeping in mind:
> less is more. I'll practice on a napkin first.

No, you do a test fitting, applying your proposed dot pattern, and
checking the spread of the compound when the heatsink is pressed into it.

The size of the "contact circle" as the paste spreads, tells you

1) Whether the dot was the right size.
2) Whether it should have been a five dot pattern.

You do not need to do a second calibration cycle, as you now know
how big the total quantity of compound needs to be, to fill
out to the edges.

Since you are changing paste types, you need to recalibrate
the quantity so you do a good job. The purpose of doing a
good job, is not creating a cleanup problem later.

   Paul

[toc] | [prev] | [next] | [standalone]


#20668

From"s|b" <me@privacy.invalid>
Date2024-11-25 15:08 +0100
Message-ID<lqjemfFkfg9U3@mid.individual.net>
In reply to#20663
On Sun, 24 Nov 2024 15:27:43 -0500, Paul wrote:

> No, you do a test fitting, applying your proposed dot pattern, and
> checking the spread of the compound when the heatsink is pressed into it.

And then clean up that mess again?
 
> The size of the "contact circle" as the paste spreads, tells you
> 
> 1) Whether the dot was the right size.
> 2) Whether it should have been a five dot pattern.

I think I'll try the small dot and (self made) spatula like the guy in
the YT vid. He says everybody's going to hate him for it, but he's been
doing it for years.

-- 
s|b

[toc] | [prev] | [next] | [standalone]


#20665

From"s|b" <me@privacy.invalid>
Date2024-11-25 14:38 +0100
Message-ID<lqjctrFk8ttU1@mid.individual.net>
In reply to#20657
On Sat, 23 Nov 2024 14:28:54 -0600, VanguardLH wrote:

> Here's another guy gobbing
> on way too much thermal paste: https://youtu.be/a0LlMDk_ljQ?t=12.

This guy is doing a lot better, I think:
<https://youtu.be/eyNTXDoLyEU?feature=shared>

I don't think my MX-4 comes with a spatula, but I've cut 1/3 of a
plastic card which I'll be using.

-- 
s|b

[toc] | [prev] | [next] | [standalone]


Page 2 of 5 — ← Prev page 1 [2] 3 4 5  Next page →

Back to top | Article view | alt.comp.hardware


csiph-web