Tale of two GA-K8VM800M

[pfrcom]

Both Rev.1 and example of mass production variation

First has MBZs for VRM input and KZGs for output - one KZG blown, shown in pic

[pfrcom]

Re: Tale of two GA-K8VM800M

Second has KZGs for input and MBZs for output - all look OK, pic

[pfrcom]

Re: Tale of two GA-K8VM800M

Originally in filthy condition - cleaned in water with Simple Green, then dried outdoors in Australian Summer sun

Afterwards neither would start - sprayed with Isopropyl alcohol and left to dry outdoors again

Then they started, and so did my frustration

With the first, I replaced KZGs with recycled HMs and thin MBZs - deprecated practice, but board's worth very little

[pfrcom]

Re: Tale of two GA-K8VM800M

The first did an install of Windows XP with one issue - Code 10 for Sound - it's Realtek ALC655, using a well proven driver

In Details it has SUBSYS_00000000 - compared to SUBSYS_A0021458 for one without issue

No issues at all with second - no exclamations in Device Manager, including Realtek AC97 - but no sound either

To verify, I swapped in my Rev.2 K8VM800M - and after XP adjusted for its differences, there was sound

I've tried different drivers, and older BIOSes, all to no avail

Along the way, its Realtek AC97 device vanished - don't even see Sound allocated an IRQ, in list shown by BIOS during startup

I've even tried replacing original three 100uf 10v Teapo near ALC655 chip, with 100uf 16v KMA recycled from a defunct Intel board

Shown in this pic - note default jumpers on F_AUDIO, been caught before by that gotcha for no sound

[pfrcom]

Re: Tale of two GA-K8VM800M

Amount of time I've spent on these, I should have left them in the computers left out for rubbish collection

[pfrcom]

Re: Tale of two GA-K8VM800M

With more searching I found a driver for the first board, accepted by XP for Realtek AC97

But still no sound produced

After these, plus an Asus A8V Deluxe that also produces no sound even though driver shows no issues, I reckon on-board Realtek sound is just plain fragile

[momaka]

Re: Tale of two GA-K8VM800M

That could be it. Or perhaps something specific with those chips, their drivers, and XP. Would be interesting to try another OS. Maybe Windows 7?

An HP DC7700 motherboard did something similar to me recently where the AC97 just disappeared randomly from Device Manager and no more sound after that. But that was with a SoundMax onboard audio, so I can't imagine this to be the same problem. However, I will note that both the HP website and Windows Update had some specific patch for that that fixed my problem.

So that's why I am suggesting a different (and possibly newer) OS - just for an experiment.

[pfrcom]

Re: Tale of two GA-K8VM800M

Would be interesting to try another OS. Maybe Windows 7?

Not optimistic, seeing second board no longer shows a sound device in the IRQ list produced by Award BIOS

Speaking of which, it's F8, the latest for a Rev.1 board

BTW, the first board also had sound issues when I booted from a Knoppix CD

I'll try the second board with Windows 7, and post an update

[stj]

Re: Tale of two GA-K8VM800M

are you sure you dont need jumpers on the front-panel conector to route the sound to the jack sockets?
i have a few boards like that.

[stj]

Re: Tale of two GA-K8VM800M

just checked my similar K8ns-pro
if you have a 10pin front audio connector, you need to put a jumper across 5-6 and 9-10 to route the autput to the back.

[pfrcom]

Re: Tale of two GA-K8VM800M

you need to put a jumper across 5-6 and 9-10 to route the autput to the back.

Thanks, but in post #4

note default jumpers on F_AUDIO, been caught before by that gotcha for no sound

[momaka]

Re: Tale of two GA-K8VM800M

Amount of time I've spent on these, I should have left them in the computers left out for rubbish collection

True.
But you also miss out an opportunity to improve your troubleshooting skills, or learn something new perhaps.

After these, plus an Asus A8V Deluxe that also produces no sound even though driver shows no issues, I reckon on-board Realtek sound is just plain fragile

Could be that too. It seems strange that the sound went out on both boards and no device in Dev. Manager. Certainly doesn't sound like an SB BGA issue, especially since those VIA chipsets likely don't get hot at all to cause that.

However, now I also recall having a computer with an Asus A8V-MX where the sound disappeared is a similar manner (I don't remember if the sound devices disappeared from Dev. Manager, but they probably did because I remember trying to reinstall the drivers several times). Long story short, I found that the problem was caused by a wrong driver for some other device I tried to install on the board. I don't remember how I spotted that, though. Also, that board had an ALC653 chip, which could be similar to what you have.

If that's not the issue, then I wonder if Gigabyte just did something wrong with the board design to cause that. Check AVDD (pins 25 and 38), DVDD (pins 1 and 9), and VRef (pin 27) voltages on the Realtec ALC655 chip on your board and make sure they are in spec.

if you have a 10pin front audio connector, you need to put a jumper across 5-6 and 9-10 to route the autput to the back.

That shouldn't make the audio disappear from Device Manager, even if he didn't have those jumpers.

[pfrcom]

Re: Tale of two GA-K8VM800M

seems strange that the sound went out on both boards and no device in Dev. Manager

First board does have Realtek AC97 in Device Manager - but produced no sound after I found a driver to satisfy it

That shouldn't make the audio disappear from Device Manager, even if he didn't have those jumpers.

Extreme example with those jumpers missing was an Intel D845GERG2 which wouldn't even power up

Bought it as faulty - pleased with the fix required

[stj]

Re: Tale of two GA-K8VM800M

ah yes, power.
the audio pre and output stages need seperate power.
get some pinouts and go digging.
it may be a bad inductor or regulator.

[pfrcom]

Re: Tale of two GA-K8VM800M

Would be interesting to try another OS. Maybe Windows 7?

I did try an install of Windows7, from scratch, on the second board - not surprised that it ended up saying: No sound device installed

And to add insult to injury, the second board has now decided to lose CMOS settings when power is switched off - with a CR2032 measuring 3.06v on DMM

[mockingbird]

Re: Tale of two GA-K8VM800M

Maybe there's some water creeping beneath some BGA? What I always do when I clean PCBs with water is hold a gasolene funnel to the end of my shop vac hose (shop vac hose hooked on exhaust of course) and then blow the heck out of every crevice. You'll see droplets of water flying out at light speed. You've got to have a good powerful shopvac, like the ones construction people use with a powerful motor.

[pfrcom]

Re: Tale of two GA-K8VM800M

Maybe there's some water creeping beneath some BGA?

I wondered that, especially the second board with its escalating failures

So, prior to the Windows7 test, I fired up my air compressor and tried blowing out its various nooks & crannies, albeit with air pressure fairly low

This thing's giving me the sh*ts so much, I might repeat with the air pressure dialed up

[pfrcom]

Re: Tale of two GA-K8VM800M

Maybe there's some water creeping beneath some BGA?

Mockingbird is on the money

Before having another go with compressed air turned all the way up to Eleven, I decided to remove the northbridge heatsink

To find a smear of water down one side of the chip - right side in the attached pic

So I sprayed it all over again with Isopropyl alcohol, then finished off with compressed air, a bit higher than previous but nowhere near maximum

Afterwards, the problem of losing CMOS settings was fixed

Unfortunately on-board sound hasn't come back - maybe I'll have a look at voltages on the ALC655 later, although fineness of its pins is a bit off-putting

So lesson learned - static air drying is insufficient after a bath - need moving air as per https://www.badcaps.net/forum/showth...hlight=smokers

[stj]

Re: Tale of two GA-K8VM800M

actually, you should have removed all heatsinks first.
because otherwise you could have washed away any paste without knowing.

[pfrcom]

Re: Tale of two GA-K8VM800M

you could have washed away any paste without knowing.

No chance of that happening, as it turned out

Not an atom of heat sink compound on the northbridge originally - has AS5 now

Or maybe I did wash it away in the first place