Never had a card die till yesterday - bad timing

jmc111

Posts: 12   +1
(On second PC) - Radeon 290

Working away and the image jumped.
Then after a bit it did it again...and... again.
Monitor started going blank after image jump and finally...no image came back.

Found an old Radeon 6950 and have nice 4/3 low rez image.

Knew I would have to pay a lot more for a Radeon 6600XT but the only single one on Newegg was $900.
No way in BLEEP is that happening.
Oh, and it ships FROM China.

ALL my other videocards ran till retired and would need adapters to (maybe) function on a new monitor.
Just hope my Vega56 keeps going.

I don't want anymore "firsts"!
 
Last edited:
I remember years ago one of my GTX 480s' croaked. It is rare for a GPU\CPU to just up and die. It was still under warranty though, so EVGA replaced it free.
 
Last edited:
I remember years ago one of my GTX 480s' croaked. It is rare for a GPU\CPU to just up and die.
At the end of last year, I decided that one of my machines needed more memory, and, being under the influence of alcohol and some very mild opiates, I decided to make that so.

I proceed to open the case, and have at it. During the process, I had to cut cur away part of the HSF shroud. By now I was sweating profusely,. I tried to keep my head outside of the case, but alas, I sweated all over the board..

I wiped away all the droplets from the board's chipset, feeling I had almost surely short circuited the board, and it was doomed.

I finally got the junk back together,and tried to boot it up. The system hung at the Windows flashing light and refused to budge from there.

So, I did what any "sane individual" would have done, I yanked the GTX-750 ti, to save it for a future build..

That done, I tried to reboot the system on the IGP, reasoning I had nothing to lose. "Wah-lah", rhe system booted.perfectly.

OK, so we're working in an 80 degree room at 60% RH, I surely didn't kill the card with a static.discharge.

I heaved the card. To this day I wonder if I should have reinstalled the video drivers. But I couldn't fathom how shoving in 2 extra sticks of RAM, could have trashed the driver software, especially with the system unplugged.

Oh well, live and learn, and what's done is done..

Besides, that card was nothing but trouble from the start. First, no video, and I had to call EVGA support, who fed me an obscure BIOS setting to change. Then, since I put the card in without an internet connection, M$ started sending me messages about, "how my copy of Windows might be stolen". :mad:.
 
Hundreds & hundreds of V blogs on YouTube on reflowing & reworking
GPU's. Most likely flexing it in the PCIe slot fractured the chip from the PCB.
Laptop computers also have this issue of SMT reflow work needed. Most likely from handling it roughly.
A superUSER does not have to buy electronic tools and a hot air gun usually. But if you heated it up quickly to 300 C, plus, you may be able to
save your GPU. Reflow liquid flux is used. It would take over an hour to get it prepared. I see many cracks on the bottom connector when you wiggle it out
and do not press down that tab and get frustrated.
Moving a motherboard to another case and you may damage the motherboard and have DRAM LEDs show that is the issue. Depressing
when we damage our equipment and warranty will not cover this damage.
Many balled solder connections under the CPU holder may fracture and be like a cold solder joint thereafter. Handle your circuit cards very gently.
 
I had an NVidia 8600 GT card that died. It was obvious what the problem was because capacitors on its board were bulging. It looked repairable and I put it aside. As time went by I realised the repair was never going to happen so I threw it away.
 
I have never thought that a capacitor has ever bulged and went bad. That must have been when you were young in 1898 and they whistled in your short wave radio. Since the 1970s there were only good electrolytic capacitors and if put in a well-designed circuit, never poped or bulged. Put them in with revered polarity and they bulge and pop.

Just your 8600 GT is not good enough with today's standards.
Cold solder joints have always been what I disliked. Sometimes you may see a solder joint that does not shine. If the product was originally manufactured right with no cold solder joints and defects the product lasts almost forever until you tinker with it and are too rough.

If it is not broken, then do not start tinkering with anything. 20 minutes later the fractured (cat's whisker thick) solder joint connection will fail when flexed. Took me hours to come to this and my observation of what I have caused when everything was working fine, I thought to make it better to the last detail and then I disappointed myself.
 
I have never thought that a capacitor has ever bulged and went bad. That must have been when you were young in 1898 and they whistled in your short wave radio. Since the 1970s there were only good electrolytic capacitors and if put in a well-designed circuit, never poped or bulged. Put them in with revered polarity and they bulge and pop.
I hate to be the, "thrower of wet blankets":, on such an eloquent rant, but you have no idea what you're talking about.

Why do you suppose Gigabyte puts, "only solid capacitors", on every box of every motherboard they shovel out the door? Why do other makers take the trouble to announce, "only Japanese capacitors? Because some of the parts coming out of China are still trash. And no, it's not a case of "putting them in backwards". They fail because they're crap, plain and simple

So, some greedy numskull manufacturers used fluid electrolytics at least through the mid 90's, and possibly longer.

The now defunct "eMachines" brand, used the cheapest fluid capacitors they could possibly find in the PSUs of their computers. The net result was, the caps failed, and with no circuit protection, the mobos very often failed when the PSUs did. And since you couldn't install the OEM OS into another board, all you had left, was the proverbial "brick".

So, we had a "power user, expert repairman" with a long tenure of service, who had "a barn full", of these bricks, and was going to sue the living daylights out of eMachines, and their parent company, Gateway. It pained me to inform him that the statute of limitations on TORT claims, was 2 (two), years.

After Vista appeared with its native ACHI drivers, he still insisted that you "had to set the BIOS to run the HDDs as "IDE"..

Oh well, what do I know, I'm just a simple, pedestrian, "hobbyist"..
 
To: Captian Crankey: You "flamed" me that time.

I noticed ever since the 286/386 CPU had pins to be pressed down into a holder. The CPU needed cooling fans after the original CPU's. They never
sold us anything but eMachines that also had problems when I had to press
those pins back down on only a small cooling fan, so I could continue. Always
a weak point was the CPU heating up, and recently what grand CPU heat sinks & cooling fans to solve the problem.
The surface mount technology is never what I attempted to get into previously. Different soldering equipment is needed. Recently I thought I
could do a little, but maybe only the heating up and reflow using liquid flux.
I know it is this when I press down on the cooling fan and it improves
a little or loosen and tighten the cooling fan screws that it is this.
Originally I asked, "What did I do wrong here?" I accepted that it was me
and not a warranty situation. From Captian Cranky, what did I do wrong?
Handling the motherboard by the cooling heat sink & fan? I also had the
motherboard mounting standoffs come up, so next time I put them in more
snug. A new motherboard and I did this to two, one right after another thinking it was the Bios in the H.P. Z390 Omen, and not this problem. It also was this problem and then I got an MSI MAG B365M Mortar, and I disappointed myself, and this can not continue when I get the replacement that I ordered to 3 in a row.

2 in a row, and the next thing that happens is I get flamed by Captain Cranky.
 
To: Captian Crankey: You "flamed" me that time.
I only "flamed" you on the single point of your stated date after which true fluid electrolytics were installed or manufactured.

Case in point, I had a Pioneer stand alone DVD recorder, Circa 2006 (give or take).

It remained plugged in "forever". Then one day, the clock quit. I never bothered to yank it apart, but I figure it's a good possibility, that the PSU caps, "bled out", or, "fizzled out", if you prefer.

I know it wasn't just the clock, because the machine will no longer power on.

To your other points about "reflowing", that might be the issue, or "problem", we're discussing in my, "memory too much" thread.
2 in a row, and the next thing that happens is I get flamed by Captain Cranky.
Ya gotta admit, that the colloquial definition of, "insanity", is doing the same thing over, and expecting a different result.

Not that I'm even remotely thinking that could possibly apply to you.
 
Last edited:
Back