VOGONS


First post, by appiah4

User metadata
Rank l33t++
Rank
l33t++

The Athlon64 Socket 754 computer in my signature (I will list the parts again in a moment) was rehoused last night to a modern ATX case (because I want to use the Elan Vital T10 it is in for another purpose) and.. it would simply not POST.

Build:
ASUS K8V Deluxe motherboard
AMD Athlon64 3700+
2x1GB DDR400 Kingston RAM
Mitsumi 3.5" Floppy *
Samsung Spinpoint 120GB IDE
Samsung 20x DVD-RW *
HIS Radeon 1950PRO AGP 512MB
Sound Blaster Audigy 2 SB0350

All parts are from the original system aside from those marked with an asterisk, which are previously untested parts. However as you may read below, the system behavior did not change with these parts disconnected.

Unfortunately, I did not think to check and make sure it actually worked before taking things out of the old case, but it was working when moved to storage 6 months ago..

Here's the eccentric behavior it had last night:

  • After everything was installed, I booted the PC up. There was no vide fro either DVI port on the card. Fans spinned up and I even got the POST confirmation BEEP, but there was no further IDE activity light, it did not boot, possibly a POST error of some kind.
  • Thinking my X1950PRO died, I installed an X800XT PE instead, same behavior on either DVI or VGA.
  • I stripped the setup down to its barebones: CPU, RAM, Videocard. Not even a keyboard. No more POST beep, no video.
  • Wondering something may be shorting in the case, so I removed everything and set it up as an open bench, still no video.
  • I thought I would test the PSU for voltage; 5.05V and 11.80V on the molex connector.
  • Then something possessed me to check for shorts on the PSU, and this is where it gets weird: When the PSU is NOT connected to the motherboard, 12V and 5V on the molex have no continuity to Ground. When the PSU is connected to the motherboard, checking for continuity between 5V and Ground results in an alert beep from the multimeter for about half a second, but then it stops, almost as if something is cutting the 5V rail shortly after.

So what do I make of this? Is the molex behavior normal? Could it be the PSU is dying out? Could it be I somehow managed to damage the motherboard? Why could I not be getting any video out from any graphics card?

Last edited by appiah4 on 2019-12-09, 07:20. Edited 1 time in total.

Retronautics: A digital gallery of my retro computers, hardware and projects.

Reply 1 of 24, by PCBONEZ

User metadata
Rank Oldbie
Rank
Oldbie

Reset the BIOS. The settings may have gotten buggered in storage.

All through DDR & DDR2 Asus was fond of junk non-Japanese caps.

Disconnect the PSU and do the 5v continuity check directly on the pins in the mobo power connector.

Did you connect any drives when you connected the motherboard? A short in a drive might do this.
.

GRUMPY OLD FART - On Hiatus, sort'a
Mann-Made Global Warming. - We should be more concerned about the Intellectual Climate.
You can teach a man to fish and feed him for life, but if he can't handle sushi you must also teach him to cook.

Reply 2 of 24, by appiah4

User metadata
Rank l33t++
Rank
l33t++
PCBONEZ wrote:
Reset the BIOS. The settings may have gotten buggered in storage. […]
Show full quote

Reset the BIOS. The settings may have gotten buggered in storage.

All through DDR & DDR2 Asus was fond of junk non-Japanese caps.

Disconnect the PSU and do the 5v continuity check directly on the pins in the mobo power connector.

Did you connect any drives when you connected the motherboard? A short in a drive might do this.
.

I did try resetting the CMOS, no go.

All the caps on the board look fine, although I realize visual inspection is not a guarantee. Finding what is gone may be hard, but I will check all electrolytics for shorts on the circuit tonight.

I checked the 5V and G sockets on the 20-pin ATX connector on the motherboard for shorts, there were none. I also checked the PSU for a short between 5V and G while connected to Power but without a motherboard connected, and there is no short there on its own either.) It's confusing.

I had a new DVD-ROM and a new Floppy connected to the system, these are the only new parts on the PC (let me not ethat down in the original post now that I think of it..). However, it continued to exhibit the behavior even after removing them. Maybe a short on one of them fried the motherboard or a fuse in the PSU?

Retronautics: A digital gallery of my retro computers, hardware and projects.

Reply 3 of 24, by PCBONEZ

User metadata
Rank Oldbie
Rank
Oldbie
appiah4 wrote:

I checked the 5V and G sockets on the 20-pin ATX connector on the motherboard for shorts, there were none. I also checked the PSU for a short between 5V and G while connected to Power but without a motherboard connected, and there is no short there on its own either.) It's confusing.

You did not somehow end up with one of those proprietary Dell P3 20-pin PSUs did you?

If that PSU came with the new case or you haven't used it before check that the wires did not get switched in the plug.
If no, then check that they aren't switched on the PCB inside - which you can by voltage checks at the plug if you don't want to pop the cover.
Some PSUs won't start or stay running without some load. If you run into that, plug a hard drive in.
You don't need to mess with the data cables when you use a hard drive for load. Better if the drive has no important data on it.

appiah4 wrote:

I had a new DVD-ROM and a new Floppy connected to the system, these are the only new parts on the PC (let me not ethat down in the original post now that I think of it..). However, it continued to exhibit the behavior even after removing them. Maybe a short on one of them fried the motherboard or a fuse in the PSU?

Usually PSUs only have one fuse and if it blows it takes out the whole thing.

The data cables are not in the power path.
A shorted drive might burn something in the PSU but it's not likely to hurt the motherboard.
(Except in the unusual case where drives get power from a special jack on the motherboard. Not common.)
.

GRUMPY OLD FART - On Hiatus, sort'a
Mann-Made Global Warming. - We should be more concerned about the Intellectual Climate.
You can teach a man to fish and feed him for life, but if he can't handle sushi you must also teach him to cook.

Reply 4 of 24, by appiah4

User metadata
Rank l33t++
Rank
l33t++

The PSU is a known and fully working High Power 500W that I have previously used in my Athlon FX system as well as this very same Socket 754 motherboard. It is (was) fully functionaly and ATX compliant.

It is weird that there is a short (or seemingly there is one, there is a continuity beep for half a second then it stops..) only when the motherboard is connected makes the motherboard suspect to me but there is no short on the 5V on the motherboard without the PSU connected so I do not know what to think of it.

I will use a spare AOpen 300W PSU I had lying about tonight. That means I'll probably have to downgrade to the X800XT-PE but that's OK, it' s more period correct and allows for a Windows 98 dual boot anyway.

Retronautics: A digital gallery of my retro computers, hardware and projects.

Reply 5 of 24, by PCBONEZ

User metadata
Rank Oldbie
Rank
Oldbie

Capacitors fail in several different ways.
One that is almost but not quite 100% shorted might take just enough charge after power is applied for a bit to turn of the beep.
.
It's not good to jump on that one as the problem without looking at other possibilities though.
.

GRUMPY OLD FART - On Hiatus, sort'a
Mann-Made Global Warming. - We should be more concerned about the Intellectual Climate.
You can teach a man to fish and feed him for life, but if he can't handle sushi you must also teach him to cook.

Reply 6 of 24, by appiah4

User metadata
Rank l33t++
Rank
l33t++

Just when you thought it couldn't get any weirder.. It did.

Now, I brought home an AOpen 300W PSU. First things first: I did NOT think to test this PSU before hooking it up to the board, but had no reason to believe it was faulty (especially in the same way).

So I hooked the PSU up to the mainboard and then to the power outlet. Then, without turning the PC on, checked the 5V to Ground for continuity. And.. It also had continuity!

Video: https://www.youtube.com/watch?v=qPR74CJkmko

Then I pulled the setup apart, and tested the PSU alone.. Not connected to the motherboard or power, nothing. And.. 5V to Ground still had continuity! Whatever the motherboard did to the PSU, it seems to have caused permanent damage to it (the more modern High Power 500W PSU did not exhibit this behavior, but then when I checked that 5V shorted to G for just half a second and stopped - probably some kind of overvoltage protection in the modern unit..)

Video: https://www.youtube.com/watch?v=EshQ8IyOzF8

So I thought, OK, this mainboard is FUCKED, let's see if I can find any shorts.. I tried one of the VRM capacitors, and.. It was shorted (in circuit, of course - but still not the behavior I should be seeing?)

Video: https://www.youtube.com/watch?v=ODDKkawJcAo

I'm now in the market for a new Socket 754 to replace this one (adding the photo in case someone can notice anything physically wrong that I couldn't):

ASUS K8V Deluxe Damaged.jpg
Filename
ASUS K8V Deluxe Damaged.jpg
File size
1.63 MiB
Views
1570 views
File license
Fair use/fair dealing exception

Is the motherboard dead? What can explain all this? What are the chances that the motherboard also killed my precious Athlon64 3700+ as well? 🙁

Retronautics: A digital gallery of my retro computers, hardware and projects.

Reply 7 of 24, by appiah4

User metadata
Rank l33t++
Rank
l33t++

OK, so nobody has any ideas about what's going on here? 😢

Retronautics: A digital gallery of my retro computers, hardware and projects.

Reply 8 of 24, by kikipcs

User metadata
Rank Newbie
Rank
Newbie

While I don’t have any concrete suggestions, I do have food for thought: one of my Socket 7 boards just would not boot outside a case. The only half-logical explanation I have for it was that maybe ground was around the screw holes, or something? Haven’t had time to read the whole problem-in-progress (pardon me), but if it worked just fine in the old case, can you try putting the mobo in that old case and see if it works? Also, are you sure you didn’t damage any traces or bumped any capacitors when swapping the board?
Another tidbit of info for you: I once rescued a friend of a friend of my uncle’s homebuilt PC. Turned out that the panel connections were wired wrong; that was the only thing preventing it from operating properly. If you have access to a manual, it’d be advisable to check afery jumper or pin scheme. And take it easy!

Reply 9 of 24, by appiah4

User metadata
Rank l33t++
Rank
l33t++
kikipcs wrote:

While I don’t have any concrete suggestions, I do have food for thought: one of my Socket 7 boards just would not boot outside a case. The only half-logical explanation I have for it was that maybe ground was around the screw holes, or something? Haven’t had time to read the whole problem-in-progress (pardon me), but if it worked just fine in the old case, can you try putting the mobo in that old case and see if it works? Also, are you sure you didn’t damage any traces or bumped any capacitors when swapping the board?
Another tidbit of info for you: I once rescued a friend of a friend of my uncle’s homebuilt PC. Turned out that the panel connections were wired wrong; that was the only thing preventing it from operating properly. If you have access to a manual, it’d be advisable to check afery jumper or pin scheme. And take it easy!

Thanks for the suggestions. In order;

- It does not boot in a modern case, and it does not boot on an open bench. I could consider moving it back into an old ATX/AT case, but how would that explain the strange 5V/Ground short_
- I don't *think* I damaged the board, but it is possible after all. I can't locate any scratches visible to the eye, but it could have been ESD damage all the same.
- It still does not work even with no front panel connectors connected..

Retronautics: A digital gallery of my retro computers, hardware and projects.

Reply 10 of 24, by TheMobRules

User metadata
Rank Oldbie
Rank
Oldbie

If there is a short on the motherboard the PSU protection should kick-in as soon as you turn it on. If at least the fans start and the PSU voltages look OK as you mentioned, then the problem may be somewhere else (be sure to also check the +3.3V line).

appiah4 wrote:

After everything was installed, I booted the PC up. There was no vide fro either DVI port on the card. Fans spinned up and I even got the POST confirmation BEEP, but there was no further IDE activity light, it did not boot, possibly a POST error of some kind.

This caught my attention since I had the exact same problem with a 775 motherboard a while back. The first few times I would hear the POST beeps but no video... I initially thought it may be the video card, but after discarding that and trying a few more times I didn't even get beeps any more, only the fans running. It turned out to be bad capacitors, specifically KZG from Chemi-con, a known problematic series that is almost guaranteed to have an early demise.

Which leads me to the picture you posted: those 3 large caps on the VRM input (near the parallel port connector), they look like Chemi-con, can you tell what series they are? If it's KZG or KZJ I would bet that's your problem, even if they appear to be fine. My MSI K8N-SLI had 3 KZGs on the same spot, all 3 failed (though only one was bloated).

Reply 11 of 24, by appiah4

User metadata
Rank l33t++
Rank
l33t++

Capacitors o nthe motherboard are from a shit brand called.. KXG or KXE, or some such - I forget, what exactly - but they are shit..

Time to order replacement caps I guess! In the meantime I ordered an nForce3 Socket 754 board (Gigabyte GA-K8NS) to try out and make sure everything else works and the issue is indeed the motherboard.

The board also has started to turn on automatically without an on/off button attached when the 20-pin ATX connector is connected..

Retronautics: A digital gallery of my retro computers, hardware and projects.

Reply 12 of 24, by imi

User metadata
Rank l33t
Rank
l33t

KXG and KXE are capacitor series from nippon chemicon, absolutely not a shit brand :p
https://en.wikipedia.org/wiki/Nippon_Chemi-Con
there should be a small outline of the logo next to the letters

Reply 14 of 24, by kikipcs

User metadata
Rank Newbie
Rank
Newbie

Alright, I have a couple more suggestions you might want to consider on your wonderful journey of troubleshooting.

- The first thing that came to my mind after a further read is that the BIOS might have somehow got damaged/corrupted/whatever. Could be by touching one of the chip's legs with another metal thing or by ESD (don't quote me on that, though, I'm not an expert).
One thing you should definitely try is attach both a hard drive, and a floppy drive. Boot up and observe the behavior.
Take a look at the manual for your board, too: https://dlcdnets.asus.com/pub/ASUS/mb/sock754 … _k8v_deluxe.pdf

It mentions a "CrashFree BIOS 2". It might be a tool to try out in case something did in fact happen to the BIOS.

- You should note that based on the manual, your mobo has a dual-function power switch. It could be that it is stuck in one mode or switches immediately into another mode, preventing you from boot?

- Apparently your board came with what ASUS calls a "POST Reporter". It is supposed to provide vocal (!) info as to the state of the POSTs, like it would verbally abuse you if you were to install memory the wrong way, I assume? Attach a speaker (not a tiny piezo one, a speaker speaker with a cone and all) and check if the board says anything worthwhile. If it doesn't, it may be that somebody changed/erased these messages (they are customizable with a OS-based editor)

- What is the behavior of the on-board LED?

There is also a "DIY Troubleshooting Guide" for your mobo on the Asus page, but sadly it's in Traditional Chinese...

Reply 15 of 24, by appiah4

User metadata
Rank l33t++
Rank
l33t++
imi wrote:

KXG and KXE are capacitor series from nippon chemicon, absolutely not a shit brand :p
https://en.wikipedia.org/wiki/Nippon_Chemi-Con
there should be a small outline of the logo next to the letters

I googled K8V Deluxe capacitor and was confronted with a gazillion websites complaining about failing capacitors on this motherboard.. Eventually, I was led to an ASUS press release on this problem: https://web.archive.org/web/20040401083247/ht … spx?KB_ID=85257

I have heard some rumors saying that the appearance of some capacitors on K8V Deluxe may deform soon after use. Is that true? Ho […]
Show full quote

I have heard some rumors saying that the appearance of some capacitors on K8V Deluxe may deform soon after use. Is that true? How do I check if my K8V Deluxe is affected, and what should I do next if mine happened to be like that?

It has been discovered recently that the nine electrolytic capacitors CE18, CE19, CE20, CE21, CE22, CE25 , CE38, CE39, CE45 (see Figure1, used at power stabilization circuitry) may expand and fracture. This problem has been confirmed to result from quality control of RLZ 0333 capacitors, and proven to happen only on a small number of such capacitors which can be easily distinguished from the RLZ 0333 marking (see Figure 2) on them.

If you need further support after finding RLZ 0333 capacitors at these nine locations, or experiencing symptom as described above on K8V Deluxe motherboards, please kindly contact your place of purchase, supplier, or the nearest ASUS service centre for immediate follow-up.

(December 04, 2003)

Figure 1.jpeg
Filename
Figure 1.jpeg
File size
68.67 KiB
Views
1444 views
File license
Fair use/fair dealing exception

Figure 1

Figure 2.jpeg
Filename
Figure 2.jpeg
File size
6.86 KiB
Views
1444 views
File license
Fair use/fair dealing exception

Figure 2

RLZ is probably the brand of capacitors on my board that I can't remember for sure right now. I think we found the culprit 🙁

This will be a long recap job, and I HATE desoldering stuff.. Let's hope it didn't kill my 3700+..

an81 wrote:

I had a similar problem with M3A32-MVP Deluxe, posting on initial check and then not posting at all after rehousing and not being able to figure it out at all. In the end it turned out to be cracked north bridge ic from that shitty cement like thermal paste asus uses on their boards. The board would flex as I pulled it out and with all the heatpipes connecting the vrm and nb and the sb, that stress had to vent somewhere. Yours is a lone nb radiator only, but it wouldn't hurt to check it out anyway, just be sure to preheat it to ~100° C before trying to pull it off.

Entirely possible but not very likely, I will not lift the cooler until it is my last resort, but this is now at the back of my mind as well.

kikipcs wrote:
Alright, I have a couple more suggestions you might want to consider on your wonderful journey of troubleshooting. […]
Show full quote

Alright, I have a couple more suggestions you might want to consider on your wonderful journey of troubleshooting.

- The first thing that came to my mind after a further read is that the BIOS might have somehow got damaged/corrupted/whatever. Could be by touching one of the chip's legs with another metal thing or by ESD (don't quote me on that, though, I'm not an expert).
One thing you should definitely try is attach both a hard drive, and a floppy drive. Boot up and observe the behavior.
Take a look at the manual for your board, too: https://dlcdnets.asus.com/pub/ASUS/mb/sock754 … _k8v_deluxe.pdf

It mentions a "CrashFree BIOS 2". It might be a tool to try out in case something did in fact happen to the BIOS.

- You should note that based on the manual, your mobo has a dual-function power switch. It could be that it is stuck in one mode or switches immediately into another mode, preventing you from boot?

- Apparently your board came with what ASUS calls a "POST Reporter". It is supposed to provide vocal (!) info as to the state of the POSTs, like it would verbally abuse you if you were to install memory the wrong way, I assume? Attach a speaker (not a tiny piezo one, a speaker speaker with a cone and all) and check if the board says anything worthwhile. If it doesn't, it may be that somebody changed/erased these messages (they are customizable with a OS-based editor)

- What is the behavior of the on-board LED?

There is also a "DIY Troubleshooting Guide" for your mobo on the Asus page, but sadly it's in Traditional Chinese...

- I have attached IDE and floppy devices to the system, and at first it gave me a POST beep but no video, but now it no longer produces any beeps whatsoever, and when I connect a PSU to it, that PSU's 5V and G appear to be shorted.

- I can't really verify the BIOS's integrity at this point

- The POST reporter is actually a BIOS function and I never enabled it.

- The onboard LED lights up as soon as a PSU is connected, and remains ON (Stand-By) regardless of whether the system is powered on or off.

Retronautics: A digital gallery of my retro computers, hardware and projects.

Reply 17 of 24, by kikipcs

User metadata
Rank Newbie
Rank
Newbie

Mmmm... I'm still keen to push the possibility of bad BIOS, mainly because I've had a pretty similar problem and bad BIOS was the exact cause, hence I was asking for the LED status. The more light you shine on the capacitors, though, the more of a suspect they become. they might as well have dried up or cracked inside, with no visible signs of failure. Well then, If you're more inclined towards recapping I stand by you. (buuuut bad BIOS is still a possibility, though 😁 )

Take it easy and keep in touch with us in regards to updates!

Reply 18 of 24, by appiah4

User metadata
Rank l33t++
Rank
l33t++

I checked my board and all caps marked on ASUS FAQ are 1500uF 6.3V parts; however mine are not RLZ 0333; the ones by the VRM are RLX 0344 (again by OST, maker of the faulty RLZ caps) and the ones near the RAM are a super suspect Elite brand. I think it is caps all the same..

More on OST caps: https://badcaps.net/forum/showthread.php?t=1585

Retronautics: A digital gallery of my retro computers, hardware and projects.

Reply 19 of 24, by PCBONEZ

User metadata
Rank Oldbie
Rank
Oldbie

I am not active at badcaps.net any more but when I was I ran up over 10k posts over there.
That was when rebuilding motherboards was my 'main thing'.

The brand is OST
RLX and RLZ are series of OST. They differ by ESR and Ripple ratings.
The 0333 and 0344 are just date codes. Year 2003, 33rd and 44th week.

OST are FAMOUS for failing with no visible signs. Not the only brand or series, but a common one.
Elite aren't good either but they are more common in LCD monitors and PSUs than on mobos.

About the time P4 and DDR RAM came to be Asus was using them (and other crap caps) by the boatload.
IMHO Asus became a crap brand at that point. I used to say "Friends don't let friends do Asus".
Years ago over at BCN I coined the expression "ASDS".
"Asus Sudden Death Syndrome". It related directly to their extensive use of crap caps that invisibly fail.
(PC133 and earlier Asus were usually good quality boards.)

In 2003 pretty much everyone had those problems but Gigabyte and Abit quickly cleaned up their act by switching to all Jap caps.
Asus and most others didn't.
In the same time frame Nichicon (a Jap brand) HM and HN series had a manufacturing defect which confused everyone (still to this day). HM and HM with date codes 2005 and older are suspect but it seems not all batches had the problem because only one or two plants had the procedure wrong.

In all likely hood a full recap will bring your board back.
With OST don't skip any of them. Not even the little ones.

I can't be here much but there are several vogons members that are long-time BCN members.
If none find this then ask at BCN. A lot of great techs over there.
.

GRUMPY OLD FART - On Hiatus, sort'a
Mann-Made Global Warming. - We should be more concerned about the Intellectual Climate.
You can teach a man to fish and feed him for life, but if he can't handle sushi you must also teach him to cook.