Motherboard killing my Hard Drives

  • Thread starter Thread starter Searcher7
  • Start date Start date
Ting Hsu said:
My suggestion on raid is to use the built in Windows 7 or Linux, OS
based raid, whenever possible.

Since the OS handles the raid, it means that if your machine crashes,
you can just pull the drive and put it into another box, and it will
be read like any other drive, with very little work. As long as you
stick with the same OS, or even an upgraded version of that OS, you
are good to go.

This is a lot better situation than a hardware raid, where you almost
always need compatible hardware to read the raid drives. And sometimes
compatible hardware is tough to find, as hardware companies are
upgrading their raid hardware on a yearly basis.

Seems to me that RAID defeats the whole point of RAID!

A member of another group was occasionally offline, when he eventually
re-appeared he explained that his RAID controller had died and the computer
shop no longer stocked that exact card - so it took him up to weeks to get
his PC up and running again!

AFAIK - he's abandoned the use of RAID altogether.
 
Seems to me that RAID defeats the whole point of RAID!

A member of another group was occasionally offline, when he eventually
re-appeared he explained that his RAID controller had died and the
computer shop no longer stocked that exact card - so it took him up to
weeks to get his PC up and running again!

AFAIK - he's abandoned the use of RAID altogether.

Raid hardware failure was one reason the IP department of the company I
worked at had a brand new, but tested, spare raid card and software
boxed up and locked inside the server cabinet. Servers they could get
in an emergency from other locations in the company, like development
labs and such but they wanted a matching raid card on hand to rebuild
things with if the worst happened.

True, they had off-site backup's of the actual data but that was
offline, not online as far as the users were concerned.
 
Ian said:
Seems to me that RAID defeats the whole point of RAID!

A member of another group was occasionally offline, when he eventually
re-appeared he explained that his RAID controller had died and the
computer shop no longer stocked that exact card - so it took him up to
weeks to get his PC up and running again!

AFAIK - he's abandoned the use of RAID altogether.

Generally, what you need, is the same *brand* of RAID card.
It doesn't have to be the same model of controller.

For example, on Promise cards, a newer Promise card will
recognize the metadata on a disk, put there by an
older Promise card. So your array will come up, on
a newer card.

Tomshardware had an article years ago, where they tested that,
and found a reasonable level of same-brand compatibility.
(One Intel RAID, will recognize another Intel RAID array.
A Promise will recognize a Promise. A Promise will not
recognize an Intel. That sort of thing.)

If the brand of controller is obscure, or the controller
company went out of business five years ago, then yes,
buying and keeping an identical card, helps. Companies with
bottomless pits of money for IT, would buy and keep a
replacement RAID card, for each RAID type in service.

An example of a situation to be watched, would be if you
own an NVidia motherboard, and are using NVRAID. Since
NVidia doesn't make desktop chipsets any more, a user
having a failure of their NVRAID array, could have a
mess to clean up. (Look on Ebay for another motherboard.)

You should be able to manually de-interleave a RAID array,
so the data should be recoverable. It's not like the
data is lost. It's merely inconvenient to restore it.
On a RAID 5, you know there are going to be blocks of
XOR data, for protection, so there's a pattern to what
is on the disks. And the appropriate data recovery software
($$$), can figure it out. The reason the software is ($$$),
is because the software company writing such software,
can charge what the market will bear.

If any form of encryption was involved, then, you'd be
screwed. Better have good backups in that case.

Paul
 
PXE boot code typically runs, if a disk cannot be found.

On a good BIOS, you can go into the hardware setup, and
disable the boot ROM for the NIC, and stop those messages
from appearing.

The "disk boot failure", means that none of the devices
in the boot order, were located.

    Paul

I have no choice but to use this system until I can get another.

The drive is still holding, but the PC will sometimes freeze or crash
and restart. I then get this 9 out of 10 times:
http://i290.photobucket.com/albums/ll257/Statenislander/Computer/AfterCrashA_zps2d57c187.jpg

Though I got this once:
http://i290.photobucket.com/albums/ll257/Statenislander/Computer/AfterCrashB_zps44114a74.jpg

I attempted to test the power supply, but for some really strange
reason I can't get it to turn on when it is disconnected from the
motherboard and drives. I shorted the green "Power-on" pin to the
ground next to it by sticking some 18g stranded wire into both, but it
still wouldn't turn on.

Darren Harris
Staten Island, New York.
 
Searcher7 said:
I have no choice but to use this system until I can get another.

The drive is still holding, but the PC will sometimes freeze or crash
and restart. I then get this 9 out of 10 times:
http://i290.photobucket.com/albums/ll257/Statenislander/Computer/AfterCrashA_zps2d57c187.jpg

Though I got this once:
http://i290.photobucket.com/albums/ll257/Statenislander/Computer/AfterCrashB_zps44114a74.jpg

I attempted to test the power supply, but for some really strange
reason I can't get it to turn on when it is disconnected from the
motherboard and drives. I shorted the green "Power-on" pin to the
ground next to it by sticking some 18g stranded wire into both, but it
still wouldn't turn on.

Darren Harris
Staten Island, New York.

Verify your pinout. Just in case the colors on the harness are
wrong.

(20 pin modern - page 30 - pin 14 is PS_ON#, COM on either side)
http://www.formfactors.org/developer/specs/atx/ATX12V_1_3dg.pdf

(24 pin modern - page 37 - pin 16 is PS_ON#...)
http://www.formfactors.org/developer/specs/ATX12V_PSDG_2_2_public_br2.pdf

If a power supply "latches off" on a fault, you would need to flip
the switch on the back to the off position, wait 30 seconds,
then flip it on again. Then try holding PS_ON# to COM, for as long
as you want the PSU to run.

If the power supply is latching off, it may be "weak" and need replacement.

If I was present in the room, I'd have my multimeter out by now,
set to 20V full scale, and would be doing a lap of the main connector
pins while the computer is running. Just to see if the voltages are
within 5% of nominal. Then, I'd repeat with Prime95 running,
to add some electrical load.

There are several ways to kill a hard drive.

1) The hard drive has devices across +5 to GND and +12 to GND,
to absorb transients (such as disconnecting a drive "hot").
If the power supply goes to 7V and 15V perhaps, those
components burn. Too much PSU voltage, will damage a
drive.

2) The SATA and IDE, use relatively low voltages for operation.
Sending a voltage down the line, outside the allowed range,
burns the I/O.

So those are ways the disk could get ruined. Using my
multimeter, I'm effectively checking for (1). With (2),
the motherboard I/O are likely to be ruined, so no
further connected hard drives would ever be detected.
And yet, if you connect new drives, they are present
for a period of time. Thus, I suspect (1). The
reason (2) is less common, is the entire motherboard
might be ruined by the root cause of such a thing,
rather than just blowing out the disk I/O. If it was
(2), maybe the motherboard would simply have stopped
POSTing entirely.

Test with another power supply, if you have any doubts.
Check Newegg reviews, to find units with good reports.

Paul
 
I attempted to test the power supply, but for some really strange
reason I can't get it to turn on when it is disconnected from the
motherboard and drives. I shorted the green "Power-on" pin to the
ground next to it by sticking some 18g stranded wire into both, but it
still wouldn't turn on.

A power supply, simply approached, should be effectively connected to
the MB when turned on, with both components being also in excellent
working order. Beyond which even 5 years is OK, 10 being an excellent
run for your money. (Know someone with a 12 year old computer I
built. Once when I tried and warn him, he got pissed ... 'Are you
TELLING me my computer is going to break now?') I know you're not
exactly in a rush to drop more money, but when considering either MB
or PS, I try and buy the best I can afford. Usually well within
budget considerations, discounting for sales, rebates, and such, even
if they've been a bit sparse of late with the PC economy the way it is
facing a mobile front. And, yes, I do firmly believe I lost one of my
three golden 200G Seagates due an ailing ASUS MB. Tremendously robust
HDs, although I dicked around with that ASUS too long and got the
horns for it to show. Bad power supplies and bad motherboards, when
they're bad they can be really bad. Want to hear the saddest song
ever sung -- a HD may cost more than to replace both with viable
quality sales items.

I must have fed that ASUS at least 3 power supplies. Should have
known a damn sight better than to keep on doing it. But, like some
people, I too was getting by. . .
 
Searcher7 wrote: > On Apr 25, 5:01 am, Paul <[email protected]> wrote: >>tumppiw wrote: >>> On 24.4.2013 1:07, (e-mail address removed)2.com wrote: >>>> Over the last several days I lost the ability to boot from three hard >>>> drives. When one went down I plugged in another only to have it follow >>>>suit. >>>> The following is the screen I get when I power on the PC. >>>> http://i290.photobucket.com/albums/ll257/Statenislander/Computer/Driv... >>>> Obviously there is an issue with the motherboard because one can only >>>> give coincidence so much credit. >>>> Each time it's happened was on "power up" after a previously >>>> successful "power off". So it wasn't as if the PC crashed beforehand >>>> or exhibited any unusual behavior to signifythat the next boot up >>>> attempt would fail. >>>> Can anyone tell me what the possible problems is and a solution so >>>> this system doesn't do the same to the fourth drive I just plugged in? >>>> Thanks. >>>> Darren Harris >>>> Staten Island, New York. >>> For some strange reason, you're tryingto boot from the network, not >>> from internal disks.. >>> I don't have asecond computer, so I can't check how one disabled that >>> function in the BIOS >> PXE boot code typically runs, if a disk cannot be found. >> >> Ona good BIOS, you can go into the hardware setup, and >> disable the boot ROM for the NIC, and stop those messages >> from appearing. >> >> The "disk boot failure", means that none of the devices >> in the boot order, were located. >> >> Paul > > I have no choice but to use this system until I can get another. > > The drive is still holding, but the PC will sometimes freeze or crash > and restart. I then get this 9 out of 10 times: > http://i290.photobucket.com/albums/ll257/Statenislander/Computer/AfterCrashA_zps2d57c187.jpg > > Though I got this once: > http://i290.photobucket.com/albums/ll257/Statenislander/Computer/AfterCrashB_zps44114a74.jpg > > I attempted to test the power supply, but for some really strange > reason I can't get it toturn on when it is disconnected from the > motherboard and drives. I shorted the green "Power-on" pin to the > ground next to it by sticking some 18gstranded wire into both, but it > still wouldn't turn on. > > Darren Harris > Staten Island, New York. Verify your pinout. Just in case the colors onthe harness are wrong. (20 pin modern - page 30 - pin 14 is PS_ON#, COM oneither side) http://www.formfactors.org/developer/specs/atx/ATX12V_1_3dg.pdf (24 pin modern - page 37 - pin 16 is PS_ON#...) http://www.formfactors.org/developer/specs/ATX12V_PSDG_2_2_public_br2.pdf If a power supply "latches off" on a fault, you would need to flip the switch on the back to the offposition, wait 30 seconds, then flip it on again. Then try holding PS_ON# to COM, for as long as you want the PSU to run. If the power supply is latching off, it may be "weak" and need replacement. If I was present in the room, I'd have my multimeter out by now, set to 20V full scale, and would be doing a lap of the main connector pins while the computer is running. Just to see if the voltages are within 5% of nominal. Then, I'd repeat with Prime95 running, to add some electrical load. There are several ways to kill a hard drive. 1) The hard drive has devices across +5 to GND and +12 to GND, to absorb transients (such as disconnecting a drive "hot"). If the power supply goes to 7V and 15V perhaps, those components burn. Too much PSU voltage, will damage a drive. 2) The SATA and IDE, use relatively low voltages for operation. Sending a voltage down the line, outside the allowed range, burns the I/O. So those are ways the disk could get ruined. Using my multimeter, I'm effectively checking for (1). With (2), the motherboard I/O are likely to be ruined, so no further connected hard drives would ever be detected. And yet, if you connect new drives, they are present for a period of time. Thus, I suspect (1). The reason (2) is less common, is the entire motherboard might be ruined by the root cause of such a thing, rather than just blowing out the disk I/O. If it was (2), maybe the motherboard would simply have stopped POSTing entirely. Test with another power supply, if you have any doubts. Check Newegg reviews, to find units with good reports. Paul

I know the pinouts are correct. I rechecked several times with the ITX motherboard pinouts in the manual and also online.

I'm not sure what "latches off" means, but there is no switch on the back of the power supply. (This is a mini-ITX PC which I've really only been using since January).

And I don't see a way to test the power supply with it connected to the motherboard. (And again, I cannot buy any new hardware).

I'll take a look at hooking up one of my regular size ATX power supplies tothis system for a while to see if it experiences the same symptoms. But since the ITX motherboard has a 20 pin main connector, I'm not sure I can do that.

Thanks.

Darren Harris
Staten Island, New York.
 
I'll take a look at hooking up
one of my regular size ATX power supplies to this system for a while
to see if it experiences the same symptoms. But since the ITX
motherboard has a 20 pin main connector, I'm not sure I can do that.

-
Need an adaptor often included with the PS packaging. Used to have
one like that, boutique high-end PS to last forever. My
bulging-capacitor, ASUS MB promptly ate it, though. With so much for
forethought, I retorted by feeding it my HD, too. Yummy. Also a
denotation to precepts of higher intelligence. Fully exercising a
system's right to conclude squatting down to dim itself out before
rebooting when sticking in a CD.
 
n Island, New York. >>> For some strange reason, you're trying to boot from the network, not >>> from internal disks.. >>> I don't have a second computer, so I can't check how one disabled that >>> function in the BIOS >> PXE boot code typically runs, if a disk cannot be found. >> >> On a good BIOS, you can go into the hardware setup, and >> disable the boot ROM for the NIC, and stop those messages >> from appearing. >> >> The "disk boot failure", means that none of the devices >> in the boot order, were located. >> >> Paul > > I have no choice but to use this system until I can get another. > > The drive is still holding, but the PC will sometimes freeze or crash > and restart. I then get this 9 out of 10 times: > http://i290.photobucket.com/albums/ll257/Statenislander/Computer/AfterCrashA_zps2d57c187.jpg > > Though I got this once: > http://i290.photobucket.com/albums/ll257/Statenislander/Computer/AfterCrashB_zps44114a74.jpg > > I attempted to test the power supply, but for
some really strange > reason I can't get it to turn on when it is disconnected from the > motherboard and drives. I shorted the green "Power-on" pin to the > ground next to it by sticking some 18g stranded wire into both, but it > still wouldn't turn on. > > Darren Harris > Staten Island, New York. Verify your pinout. Just in case the colors on the harness are wrong. (20 pin modern - page 30 - pin 14 is PS_ON#, COM on either side) http://www.formfactors.org/developer/specs/atx/ATX12V_1_3dg.pdf (24 pin modern - page 37 - pin 16 is PS_ON#...) http://www.formfactors.org/developer/specs/ATX12V_PSDG_2_2_public_br2.pdf If a power supply "latches off" on a fault, you would need to flip the switch on the back to the off position, wait 30 seconds, then flip it on again. Then try holding PS_ON# to COM, for as long as you want the PSU to run. If the power supply is latching off, it may be "weak" and need replacement. If I was present in the room, I'd have my multimeter out by now, set
to 20V full scale, and would be doing a lap of the main connector pins while the computer is running. Just to see if the voltages are within 5% of nominal. Then, I'd repeat with Prime95 running, to add some electrical load. There are several ways to kill a hard drive. 1) The hard drive has devices across +5 to GND and +12 to GND, to absorb transients (such as disconnecting a drive "hot"). If the power supply goes to 7V and 15V perhaps, those components burn. Too much PSU voltage, will damage a drive. 2) The SATA and IDE, use relatively low voltages for operation. Sending a voltage down the line, outside the allowed range, burns the I/O. So those are ways the disk could get ruined. Using my multimeter, I'm effectively checking for (1). With (2), the motherboard I/O are likely to be ruined, so no further connected hard drives would ever be detected. And yet, if you connect new drives, they are present for a period of time. Thus, I suspect (1). The reason (2) is less common, is
the entire motherboard might be ruined by the root cause of such a thing, rather than just blowing out the disk I/O. If it was (2), maybe the motherboard would simply have stopped POSTing entirely. Test with another power supply, if you have any doubts. Check Newegg reviews, to find units with good reports. Paul
I know the pinouts are correct. I rechecked several times with the ITX motherboard pinouts in the manual and also online.

I'm not sure what "latches off" means, but there is no switch on the back of the power supply. (This is a mini-ITX PC which I've really only been using since January).

And I don't see a way to test the power supply with it connected to the motherboard. (And again, I cannot buy any new hardware).

I'll take a look at hooking up one of my regular size ATX power supplies to this system for a while to see if it experiences the same symptoms. But since the ITX motherboard has a 20 pin main connector, I'm not sure I can do that.

Thanks.

Darren Harris
Staten Island, New York.

There is information on connectors here.

http://www.playtool.com/pages/psuconnectors/connectors.html

You can plug a 24 pin supply, into a 20 pin motherboard, like this.
Pin 1 aligns with pin 1, and there will be four pins hanging off one
end. The plastic latch on the connector, will align with its mate
on the motherboard. This will not work, if there happens to be an
electrolytic capacitor on the right-hand end of the white connector
(which sometimes happens).

http://www.playtool.com/pages/psuconnectors/24in20.jpg

You can also plug a 20 into a 24. Again, pin 1 to pin 1,
latches mating. The 24 pin ends up with four pins unconnected,
but they are redundant pins. The motherboard still works.
Only if a couple 6600 video cards were installed, would
there be a problem with too much current flowing in the
single yellow wire. Your mini-ITX would not allow that
anyway (not enough expansion slots).

http://www.playtool.com/pages/psuconnectors/20in24.jpg

When I need to measure voltages on the connector (taking
last picture as a reference), if I look down into the
back of the power supply connector (black), I will see
exposed metal for each crimped pin. Those side-angle
photos don't show enough of the back of the connector,
to make that apparent. I shove the multimeter probe, into
the back of the PSU connector, to take measurements. I
sometimes use a needle probe (sewing needle with wire
wrapped around the outside to hold it in place), for
recessed measurements. So it is possible to get a
measurement off the thing.

*******

I purchased an extension cable at the computer store, like this.

http://www.playtool.com/pages/psuconnectors/20to24adapter.jpg

Those come in various combinations. 20 to 20. 24 to 20. The
20 to 20 (cost me $8.00), could be used to solve the
problem of a capacitor getting in the way of plugging a
24 PSU to a 20 motherboard. The nice cables, will be constructed
with proper wire colors, to hint at proper connection. But the
one I bought, they didn't bother with proper color coding. And
$8 was a bit much for such a thing. It should have been
cheaper than that.

You can also build your own cables. My one good electronics
store in town, used to stock all the parts. That's how I got
the connector bits, to build my own load box for PSU testing.

*******

A "latching" power supply, is one in which the voltage
regulator detects a current flow fault (short circuit),
and the regulator shuts off and "remembers" that a
fault has occurred. It does this, to prevent fire or
damage, and is appropriate for high power devices. It
takes electricity to do the "remembering", and completely
turning off the flow of power (unplug from wall if need be),
will reset the detection feature.

There were some other power devices, where "put put" mode
is used. The regulating device, retries about once a
second, and shuts off immediately if overcurrent is
detected again. Circuits that retry, are good for
situations where you know there may be a heavy load
at startup, which might trigger overcurrent, but you
are confident it will resolve itself. A circuit like
that, may detect just one overcurrent event at startup,
retry, and come all the way up. So that is the other
philosophy of design. That was popular with things
around the year 2000. If there is a real short
circuit, the physical location where the fault is
may get warm or hot, from the retries happening
once a second. With the power circuit that latches
off, things remain ice cold (safe).

*******

The pico supplies, use a DC-DC converter on a
circuit board, to convert voltage from a wall adapter,
into multiple DC voltages for a miniITX board. This could
be the style of power source you have for your mini-ITX.
These are available, up to about 120 watts. The one
pictured here is rated for 80 watts max.

http://static.mini-itx.com/store/images/1876-picopsu80.jpg

You power the input of a pico supply, with a wall adapter
like this one. This is rated 110 watts for example, enough
to run a pico 80. You are expected to read the voltage
rating of the two devices (input range), to make sure you're
mating "like with like".

http://www.mini-itx.com/store/images/psu.jpg

There are some pico supplies, intended for automotive
usage, with wide input range, and cutoff to prevent
total draining of the car battery.

Plenty of fun toys for mini-ITX users, assuming
the mini-ITX actually does something useful for them...

Paul
 
You can plug a 24 pin supply, into a 20 pin motherboard, like this.
Pin 1 aligns with pin 1, and there will be four pins hanging off one
end. The plastic latch on the connector, will align with its mate
on the motherboard. This will not work, if there happens to be an
electrolytic capacitor on the right-hand end of the white connector
(which sometimes happens).

With the PSUs I have the 4 pin bit unclips from the 20 pin bit so that
it converts from 24 to 20 pin if needed and latches back again.
 
Rodney Pont wrote: > On Fri, 07 Jun 2013 14:19:57 -0400, Paul wrote: > >>You can plug a 24 pin supply, into a 20 pin motherboard, like this. >> Pin1 aligns with pin 1, and there will be four pins hanging off one >> end. The plastic latch on the connector, will align with its mate >> on the motherboard. This will not work, if there happens to be an >> electrolytic capacitor on the right-hand end of the white connector >> (which sometimes happens). > > With the PSUs I have the 4 pin bit unclips from the 20 pin bit so that > it converts from 24 to 20 pin if needed and latches back again. > True. I forgot to write that part up :-) The detachable kind. Mine uses a plastic hinge, rather than the metal hooks shown here. http://www.playtool.com/pages/psuconnectors/all20plus4.jpg Paul

I actually use an adaptor between the power supply and Mini-ITX motherboardbecause the cable would not reach.

http://i290.photobucket.com/albums/ll257/Statenislander/Computer/PowerConnector.jpg

The adaptor is a 24 pin so 4 positions are not used.

Anyway I'm getting that screen again: http://i290.photobucket.com/albums/ll257/Statenislander/Computer/AfterCrashA_zps2d57c187.jpg

And that is what I get no matter what har drive I plug in.

I did figure out how to test the voltages from the power supply. I used thepointing test probes and got the following voltages by sticking them into the insertion side of the power supply connector as it was on connected to the motherboard.

2.377--3.398--GND--5.08--GND--5.07--GND--4.62--GND---12.26
3.390--11.88--GND--0.222-GND--GND---GND 5.08--5.10

The hard drive connector was good also.

I'm guessing the motherboard is toast, because I can't think of anything else to do.

Thanks.

Darren Harris
Staten Island, New York.
 
I actually use an adaptor between the power supply and Mini-ITX motherboard because the cable would not reach.

http://i290.photobucket.com/albums/ll257/Statenislander/Computer/PowerConnector.jpg

The adaptor is a 24 pin so 4 positions are not used.

Anyway I'm getting that screen again: http://i290.photobucket.com/albums/ll257/Statenislander/Computer/AfterCrashA_zps2d57c187.jpg

And that is what I get no matter what har drive I plug in.

I did figure out how to test the voltages from the power supply. I used the pointing test probes and got the following voltages by sticking them into the insertion side of the power supply connector as it was on connected to the motherboard.

2.377--3.398--GND--5.08--GND--5.07--GND--4.62--GND---12.26
3.390--11.88--GND--0.222-GND--GND---GND 5.08--5.10

The hard drive connector was good also.

I'm guessing the motherboard is toast, because I can't think of anything else to do.

Thanks.

Darren Harris
Staten Island, New York.

So, let's redraw the pinout and your voltage measurements.
Pin 1 is upper left. Pin 18 is blank (no pin).

Name Name
---- ----
2.377 3.3V X X 3.3V 3.390
3.398 3.3V X X -12V -11.88
GND COM X X COM GND
5.08 +5V X X PS_ON# 0.222 (logic low, is OK)
GND COM X X COM GND
5.07 +5V X X COM GND
4.62 PWR_OK X X <-- used to be a -5V pin here (no pin now)
??? +5VSB X X +5V 5.08
12.26 +12V X X +5V 5.10

Page 30 from one of these copies of the spec. The primary site
(first link) doesn't seem to be working right.

http://www.formfactors.org/developer/specs/atx/ATX12V_1_3dg.pdf

http://web.archive.org/web/20090529...tors.org/developer/specs/atx/ATX12V_1_3dg.pdf

*******

The 2.377 reading is wrong. Check why that is happening.
Maybe you meant 3.377 ?

Note that, power supplies use "remote feedback" on one of
the 3.3V pins. On the end of the main power cable, on the
supply, you may see two wires entering a single pin. The
thinner wire, caries a "copy" of the voltage level, back
to the supply. That is how the supply measures the 3.3V more
carefully, to make sure it is adjusted right.

When you add an adapter cable on the end of a supply, it
does not properly extend the sense wire back to the supply.
And what that means, is any 3.3V drop in the extension cable,
is not compensated for by the supply. This isn't normally
a big deal, as long as the extension cable uses a decent
heavy gauge wire.

You have two 3.3V pins, where the readings are great. Only
one of the pins is way off. Take another reading. Only
if that 2.377 reading is actual, do you have a problem.
Normally, all the 3.3V pins are joined to a common copper
plane on the motherboard, so the voltages can't differ all
that much, unless a pin is snapped off.

The 4.62V on PWK_OK, is a logic level coming out of an IC.
So it doesn't have to be exactly 5V in practice. Similarly,
the PS_ON# is 0.222V, and that is a valid logic low going from
the motherboard back to the PSU. Since the PSU is running,
that looks good. In some cases, if you had around 1.2V on
PS_ON#, it is possible for an ATX power supply to be "half-on",
in which case the outputs will deliver less than the normal
flow of current (supply is "weak"). And a bad level on
PS_ON# can do that. This is because the PSU does not use
a conventional logic IC to check the level on PS_ON#,
and it's likely to be some kind of transistor circuit with
poor logic characteristics. You need good clean levels on
PS_ON#, because the power supply won't be using a Schmidt
Trigger on its end.

http://en.wikipedia.org/wiki/Schmidt_trigger

So the problem might not be power. As long as that 2.377
is a typing error :-)


Paul
 
(e-mail address removed) wrote: > On Friday, June 7, 2013 3:59:19 PM UTC-4, Paul wrote: >> Rodney Pont wrote: > On Fri, 07 Jun 2013 14:19:57 -0400,Paul wrote: > >> You can plug a 24 pin supply, into a 20 pin motherboard, like this. >> Pin 1 aligns with pin 1, and there will be four pins hanging off one >> end. The plastic latch on the connector, will align with its mate >> on the motherboard. This will not work, if there happens to be an >> electrolytic capacitor on the right-hand end of the white connector >> (which sometimes happens). > > With the PSUs I have the 4 pin bit unclips from the 20 pin bit so that > it converts from 24 to 20 pin if needed and latchesback again. > True. I forgot to write that part up :-) The detachable kind.. Mine uses a plastic hinge, rather than the metal hooks shown here. http://www.playtool.com/pages/psuconnectors/all20plus4.jpg Paul > > I actually use an adaptor between the power supply and Mini-ITX motherboard because the cable would not reach. > > http://i290.photobucket.com/albums/ll257/Statenislander/Computer/PowerConnector.jpg > > The adaptor is a 24 pin so 4 positions are not used. > > Anyway I'm getting that screen again: http://i290.photobucket.com/albums/ll257/Statenislander/Computer/AfterCrashA_zps2d57c187.jpg > > And that is what I get no matter what har drive I plug in. > > I didfigure out how to test the voltages from the power supply. I used the pointing test probes and got the following voltages by sticking them into the insertion side of the power supply connector as it was on connected to the motherboard. > > 2.377--3.398--GND--5.08--GND--5.07--GND--4.62--GND---12.26 > 3.390--11.88--GND--0.222-GND--GND---GND 5.08--5.10 > > The hard drive connector was good also. > > I'm guessing the motherboard is toast, because I can't think of anything else to do. > > Thanks. > > Darren Harris > Staten Island, New York. So, let's redraw the pinout and your voltage measurements.. Pin 1 is upper left. Pin 18 is blank (no pin). Name Name ---- ---- 2.377 3.3V X X 3.3V 3.390 3.398 3.3V X X -12V -11.88 GND COM X X COM GND 5.08 +5VX X PS_ON# 0.222 (logic low, is OK) GND COM X X COM GND 5.07 +5V X X COM GND 4.62 PWR_OK X X <-- used to be a -5V pin here (no pin now) ??? +5VSB X X+5V 5.08 12.26 +12V X X +5V 5.10 Page 30 from one of these copies of the spec. The primary site (first link) doesn't seem to be working right. http://www.formfactors.org/developer/specs/atx/ATX12V_1_3dg.pdf http://web.archive.org/web/20090529...tors.org/developer/specs/atx/ATX12V_1_3dg.pdf ******* The 2.377 reading is wrong. Check why that is happening. Maybe you meant 3.377 ? Note that, power supplies use "remote feedback"on one of the 3.3V pins. On the end of the main power cable, on the supply, you may see two wires entering a single pin. The thinner wire, caries a "copy" of the voltage level, back to the supply. That is how the supply measures the 3.3V more carefully, to make sure it is adjusted right. When you add an adapter cable on the end of a supply, it does not properly extend thesense wire back to the supply. And what that means, is any 3.3V drop in the extension cable, is not compensated for by the supply. This isn't normally a big deal, as long as the extension cable uses a decent heavy gauge wire.. You have two 3.3V pins, where the readings are great. Only one of the pins is way off. Take another reading. Only if that 2.377 reading is actual, do you have a problem. Normally, all the 3.3V pins are joined to a common copper plane on the motherboard, so the voltages can't differ all that much, unless a pin is snapped off. The 4.62V on PWK_OK, is a logic level coming out of an IC. So it doesn't have to be exactly 5V in practice. Similarly, the PS_ON# is 0.222V, and that is a valid logic low going from the motherboard back to the PSU. Since the PSU is running, that looks good. In some cases, if you had around 1.2V on PS_ON#, it is possible for an ATX power supply to be "half-on", in which case the outputs will deliver less than the normal flow of current (supply is "weak"). And a bad level on PS_ON# can do that.. This is because the PSU does not use a conventional logic IC to check thelevel on PS_ON#, and it's likely to be some kind of transistor circuit with poor logic characteristics. You need good clean levels on PS_ON#, becausethe power supply won't be using a Schmidt Trigger on its end. http://en.wikipedia.org/wiki/Schmidt_trigger So the problem might not be power. As longas that 2.377 is a typing error :-) Paul

Yes, that should have been 3.377V.

There are no positions that have two leads in the connector.

And the adaopter I have actually has thicker guage leads than the wires comiong from the power supply.

Since this is the Mini-ITX board that has been killing my hard drives I canonly assume that it is a problem buried deep in the software/chips, as oddas that sounds. (Unless there are power spikes from the power supply that I'm not catching).

Thanks.

Darren Harris
Staten Island, New York.
 
ot reach. > > http://i290.photobucket.com/albums/ll257/Statenislander/Computer/PowerConnector.jpg > > The adaptor is a 24 pin so 4 positions are not used. > > Anyway I'm getting that screen again: http://i290.photobucket.com/albums/ll257/Statenislander/Computer/AfterCrashA_zps2d57c187.jpg > > And that is what I get no matter what har drive I plug in. > > I did figure out how to test the voltages from the power supply. I used the pointing test probes and got the following voltages by sticking them into the insertion side of the power supply connector as it was on connected to the motherboard. > > 2.377--3.398--GND--5.08--GND--5.07--GND--4.62--GND---12.26 > 3.390--11.88--GND--0.222-GND--GND---GND 5.08--5.10 > > The hard drive connector was good also. > > I'm guessing the motherboard is toast, because I can't think of anything else to do. > > Thanks. > > Darren Harris > Staten Island, New York. So, let's redraw the pinout and your voltage measurements. Pin 1 is upper left. Pin 1
8 is blank (no pin). Name Name ---- ---- 2.377 3.3V X X 3.3V 3.390 3.398 3.3V X X -12V -11.88 GND COM X X COM GND 5.08 +5V X X PS_ON# 0.222 (logic low, is OK) GND COM X X COM GND 5.07 +5V X X COM GND 4.62 PWR_OK X X <-- used to be a -5V pin here (no pin now) ??? +5VSB X X +5V 5.08 12.26 +12V X X +5V 5.10 Page 30 from one of these copies of the spec. The primary site (first link) doesn't seem to be working right. http://www.formfactors.org/developer/specs/atx/ATX12V_1_3dg.pdf http://web.archive.org/web/20090529...tors.org/developer/specs/atx/ATX12V_1_3dg.pdf ******* The 2.377 reading is wrong. Check why that is happening. Maybe you meant 3.377 ? Note that, power supplies use "remote feedback" on one of the 3.3V pins. On the end of the main power cable, on the supply, you may see two wires entering a single pin. The thinner wire, caries a "copy" of the voltage level, back to the supply. That is how the supply measures the 3.3V more carefully, to make sure
it is adjusted right. When you add an adapter cable on the end of a supply, it does not properly extend the sense wire back to the supply. And what that means, is any 3.3V drop in the extension cable, is not compensated for by the supply. This isn't normally a big deal, as long as the extension cable uses a decent heavy gauge wire. You have two 3.3V pins, where the readings are great. Only one of the pins is way off. Take another reading. Only if that 2.377 reading is actual, do you have a problem. Normally, all the 3.3V pins are joined to a common copper plane on the motherboard, so the voltages can't differ all that much, unless a pin is snapped off. The 4.62V on PWK_OK, is a logic level coming out of an IC. So it doesn't have to be exactly 5V in practice. Similarly, the PS_ON# is 0.222V, and that is a valid logic low going from the motherboard back to the PSU. Since the PSU is running, that looks good. In some cases, if you had around 1.2V on PS_ON#, it is possible for an
ATX power supply to be "half-on", in which case the outputs will deliver less than the normal flow of current (supply is "weak"). And a bad level on PS_ON# can do that. This is because the PSU does not use a conventional logic IC to check the level on PS_ON#, and it's likely to be some kind of transistor circuit with poor logic characteristics. You need good clean levels on PS_ON#, because the power supply won't be using a Schmidt Trigger on its end. http://en.wikipedia.org/wiki/Schmidt_trigger So the problem might not be power. As long as that 2.377 is a typing error :-) Paul
Yes, that should have been 3.377V.

There are no positions that have two leads in the connector.

And the adaopter I have actually has thicker guage leads than the wires comiong from the power supply.

Since this is the Mini-ITX board that has been killing my hard drives I can only assume that it is a problem buried deep in the software/chips, as odd as that sounds. (Unless there are power spikes from the power supply that I'm not catching).

Thanks.

Darren Harris
Staten Island, New York.

OK, here's an idea. If the native disk interface is IDE,
connect an IDE to SATA adapter, and use a SATA drive instead.
The adapter then, would "take the abuse" from the motherboard.
See if there is still a problem after doing that.

For example, this one is bidirectional, so you have to
wire it up correctly, to go from IDE motherboard to
SATA hard drive. Using this, is just a means of
digitally isolating the motherboard from the hard drive,
and see if the "damage" source remains the motherboard.
If the adapter is "blown out", then you know there is
something seriously screwy on the IDE cable.

http://www.newegg.com/Product/Product.aspx?Item=N82E16812186078

Now, this kind of adapter, you can put two of these on an IDE cable.
(The previous one, might only allow one drive per IDE motherboard
connector.) This adapter has a "Master/Slave" jumper, and if using
two adapters, you assign them unique identifiers (one master, the
other slave). If you had a single IDE ribbon, a hard drive and
an optical drive, two of this adapter would allow both to be adapted.

http://www.newegg.com/Product/Product.aspx?Item=N82E16812200156

If there was a spare PCI slot, you could plug in a PCI to IDE
controller board. As a means of bypassing the motherboard path.
That would be another way to do it. You'd need to install the
driver in the OS, for the card, before moving the cable over
from the motherboard IDE to the card IDE. Then, it should be
able to boot from the storage card.

Paul
 
Back
Top