VOGONS


Keep GeForce 4 to cards from dying

Topic actions

Reply 20 of 71, by adalbert

User metadata
Rank Oldbie
Rank
Oldbie
Miphee wrote on 2021-02-16, 07:16:

ESD is not a serial-killing monster, I haven't lost anything to ESD and I haven't used an ESD bag in years.

I lost hardware to ESD twice: first time it was SiS 530 mainboard, it stopped working right after a spark went on some data lines, and second time it was Commodore C64 CIA chip, which had one pin damaged because of a spark going to joystick port, thus disabling the functioning one joystick button. That's not much hardware, but ESD was the direct trigger of failure there.

BTW I have one Ti 4200 with some slight artifacts now, will try transplanting memory from FX 5200 and see the results.

Repair/electronic stuff videos: https://www.youtube.com/c/adalbertfix
ISA Wi-fi + USB in T3200SXC: https://www.youtube.com/watch?v=WX30t3lYezs
GUI programming for Windows 3.11 (the easy way): https://www.youtube.com/watch?v=d6L272OApVg

Reply 21 of 71, by Miphee

User metadata
Rank Oldbie
Rank
Oldbie

I also "discharge myself" sometimes but my point stands. A single bag is useless when it's off the hardware and there is nothing that discharges your body. That current has to go somewhere and that will be the hardware you touch if there is no efficient grounding present. The ESD bag is like an RCD without proper grounding, won't work when needed.
That's the reason I don't use any, it's pricey and space-consuming. I have like 800 pieces of bigger hardware and ~2000 smaller, it would cost me $2-3000 to seal everything I have. It's just a hobby, not my day job. They are expensive enough without me having to buy protective equipment because a spark MIGHT damage something on a 20-30 years old hardware.
I agree that ESD protection is better than no ESD protection but the average hobbyist won't ever know what really killed the hardware. They are old and most of them were tossed around in a recycling center for weeks before a reseller finally sold it to them. That's why they're all scratched up.
There could be so many reasons why they fail and somewhere in that line there is ESD.
But I dare anyone to show us a picture of their ESD-protected room where everything is built according to the strict ESD specs (read them, it's crazy, most of those documents are not free either) and everything is neatly packed in sealed ESD package.
Because that's the only way to effectively minimize that kind of damage.
I don't really know how people know that ESD was the reason a certain hardware failed. Do all of them work in a government science lab to confirm that?
Or is it just a 20-30 years old piece of junk that was never intended to work beyond 10ish years (planned obsolescence)?
We all know that elcos only have like 2000 hours of life @ maximum temperature and these cards are packed with it. People won't check them because they don't have the equipment for it and they don't want to desolder all caps on their shiny "new" retro stuff to test their health. They start using them and benching them and these old elcos simply give out especially on graphics cards.
Okay, it's also just a theory like ESD but that's what I believe. I guess we all have certain beliefs about these things that could be right or could be wrong.

Reply 23 of 71, by bloodem

User metadata
Rank Oldbie
Rank
Oldbie
Miphee wrote on 2021-02-16, 11:12:

How did you confirm it was due to ESD?

I guess <working device> ---> *SPARK* ---> <dead device> was his first clue 😀

1 x PLCC-68 / 2 x PGA132 / 5 x Skt 3 / 9 x Skt 7 / 12 x SS7 / 1 x Skt 8 / 14 x Slot 1 / 5 x Slot A
5 x Skt 370 / 8 x Skt A / 2 x Skt 478 / 2 x Skt 754 / 3 x Skt 939 / 7 x LGA775 / 1 x LGA1155
Current PC: Ryzen 7 5800X3D
Backup PC: Core i7 7700k

Reply 24 of 71, by Miphee

User metadata
Rank Oldbie
Rank
Oldbie

Not much info there, just a spark going places. Was the hardware under power or not? It just proves that under certain circumstances a device can be killed by ESD when it's not in the protective bag. It doesn't prove that ESD is common or that it can kill a hardware "much later" or that it can shorten the lifespan of components.
I see this topic come up all the time but I haven't seen anyone actually test the failed components with serious equipment to prove that ESD is a real threat.
Hell, I've even seen multiple videos where the guys kept "discharging" on computer components. They couldn't break anything no matter how hard they tried and even if the spark shut the computer down it booted right up after restart like nothing happened.
So what's up with that? Why didn't those components die? Why didn't the computer die when the visible spark hit the mainboard when it was on? You have to touch the hardware a 1000 times before it fails or what?
At least a comparative test should be done on multiple parts to see what happens. ESD the hell out of a few boards, RAMs and expansion cards and see if they still work or not.

Reply 25 of 71, by adalbert

User metadata
Rank Oldbie
Rank
Oldbie

Working powered mainboard, spark off my finger, no image, dead 😜 that happened in my case

BTW this was a cheap OEM board. Properly designed hardware usually haves some kind of ESD protection, either in internal chip design, or externally like TVS diodes which can absorb ESD. Of course it can still fail, but yeah, it is not that easy to kill it with normal handling.

There was a LTT video where they killed some hardware with ESD gun, one mainboard was killed in first try, and the rest required absolutely hardcore amount of zapping.

Repair/electronic stuff videos: https://www.youtube.com/c/adalbertfix
ISA Wi-fi + USB in T3200SXC: https://www.youtube.com/watch?v=WX30t3lYezs
GUI programming for Windows 3.11 (the easy way): https://www.youtube.com/watch?v=d6L272OApVg

Reply 26 of 71, by Miphee

User metadata
Rank Oldbie
Rank
Oldbie

Agreed. I'm not 100% against ESD protection, just like 90% against it. 😁
I simply don't believe that handling a component will damage it in any way and cause it to fail one by one like it happened to OP.
I handle mine without any protection because I don't want to make an expensive thing even more expensive.
I could "ground myself" with an ESD strip easily but it's uncomfortable. I already bound my metal shelf system to the EBG line and that's what matters most of the time. Proper grounding.

Reply 27 of 71, by imi

User metadata
Rank l33t
Rank
l33t

I have a grounded ESD mat to work on and store everything in ESD bags if possible, and yes I always try to discharge myself after having walked around a lot, I might not have a fully ESD compliant environment, but I try to reduce the risk at least ^^

I have infact damaged hardware by ESD before, though more modern stuff.

but the issue at hand can also be bad solder connections, this era of BGA chips is notorious for failing over time.

Reply 28 of 71, by pentiumspeed

User metadata
Rank l33t
Rank
l33t

I discharged into parallel port blowing the super i/o chip on the FX chipset pentium Asus motherboard back in the day.

Later on repaired the motherboard by replacing the chip. Hard way. I made a fine wire with tiny L hook tool and heat each leg slide the hook between legs and turn 90, then gently pull (sliding along pad) under each lifting each leg. Works well.

Cheers,

Great Northern aka Canada.

Reply 29 of 71, by bloodem

User metadata
Rank Oldbie
Rank
Oldbie
pentiumspeed wrote on 2021-02-16, 18:07:

I discharged into parallel port blowing the super i/o chip on the FX chipset pentium Asus motherboard back in the day.

Later on repaired the motherboard by replacing the chip. Hard way. I made a fine wire with tiny L hook tool and heat each leg slide the hook between legs and turn 90, then gently pull (sliding along pad) under each lifting each leg. Works well.

Cheers,

Do you have proof? How do you know that it was ESD that killed the super i/o chip? How do you know that it didn't die of old age at that precise moment? How, how, how?
</sarcasm> 😁

1 x PLCC-68 / 2 x PGA132 / 5 x Skt 3 / 9 x Skt 7 / 12 x SS7 / 1 x Skt 8 / 14 x Slot 1 / 5 x Slot A
5 x Skt 370 / 8 x Skt A / 2 x Skt 478 / 2 x Skt 754 / 3 x Skt 939 / 7 x LGA775 / 1 x LGA1155
Current PC: Ryzen 7 5800X3D
Backup PC: Core i7 7700k

Reply 30 of 71, by pentiumspeed

User metadata
Rank l33t
Rank
l33t

Getting good contact on the old style BGA chipset with molded in round metal IHS or plain early BGA packages is impossible as I researched by doing some experiments testing flatness of the top and discovered they have concave top. Best fix is lap them to get high spots sanded off which are corners.

Cheers,

Great Northern aka Canada.

Reply 31 of 71, by pentiumspeed

User metadata
Rank l33t
Rank
l33t
bloodem wrote on 2021-02-16, 18:09:
pentiumspeed wrote on 2021-02-16, 18:07:

I discharged into parallel port blowing the super i/o chip on the FX chipset pentium Asus motherboard back in the day.

Later on repaired the motherboard by replacing the chip. Hard way. I made a fine wire with tiny L hook tool and heat each leg slide the hook between legs and turn 90, then gently pull (sliding along pad) under each lifting each leg. Works well.

Cheers,

Do you have proof? How do you know that it was ESD that killed the super i/o chip? How do you know that it didn't die of old age at that precise moment? How, how, how?
</sarcasm> 😁

I was hooking my computer up and the case was desktop factor form by reaching over it blind. Zapped from my finger.

Cheers,

Great Northern aka Canada.

Reply 33 of 71, by bloodem

User metadata
Rank Oldbie
Rank
Oldbie

Can't take a joke, huh? Also the name calling was... uncalled for.

1 x PLCC-68 / 2 x PGA132 / 5 x Skt 3 / 9 x Skt 7 / 12 x SS7 / 1 x Skt 8 / 14 x Slot 1 / 5 x Slot A
5 x Skt 370 / 8 x Skt A / 2 x Skt 478 / 2 x Skt 754 / 3 x Skt 939 / 7 x LGA775 / 1 x LGA1155
Current PC: Ryzen 7 5800X3D
Backup PC: Core i7 7700k

Reply 34 of 71, by Tetrium

User metadata
Rank l33t++
Rank
l33t++
Miphee wrote on 2021-02-16, 11:12:
I also "discharge myself" sometimes but my point stands. A single bag is useless when it's off the hardware and there is nothing […]
Show full quote

I also "discharge myself" sometimes but my point stands. A single bag is useless when it's off the hardware and there is nothing that discharges your body. That current has to go somewhere and that will be the hardware you touch if there is no efficient grounding present. The ESD bag is like an RCD without proper grounding, won't work when needed.
That's the reason I don't use any, it's pricey and space-consuming. I have like 800 pieces of bigger hardware and ~2000 smaller, it would cost me $2-3000 to seal everything I have. It's just a hobby, not my day job. They are expensive enough without me having to buy protective equipment because a spark MIGHT damage something on a 20-30 years old hardware.
I agree that ESD protection is better than no ESD protection but the average hobbyist won't ever know what really killed the hardware. They are old and most of them were tossed around in a recycling center for weeks before a reseller finally sold it to them. That's why they're all scratched up.
There could be so many reasons why they fail and somewhere in that line there is ESD.
But I dare anyone to show us a picture of their ESD-protected room where everything is built according to the strict ESD specs (read them, it's crazy, most of those documents are not free either) and everything is neatly packed in sealed ESD package.
Because that's the only way to effectively minimize that kind of damage.
I don't really know how people know that ESD was the reason a certain hardware failed. Do all of them work in a government science lab to confirm that?
Or is it just a 20-30 years old piece of junk that was never intended to work beyond 10ish years (planned obsolescence)?
We all know that elcos only have like 2000 hours of life @ maximum temperature and these cards are packed with it. People won't check them because they don't have the equipment for it and they don't want to desolder all caps on their shiny "new" retro stuff to test their health. They start using them and benching them and these old elcos simply give out especially on graphics cards.
Okay, it's also just a theory like ESD but that's what I believe. I guess we all have certain beliefs about these things that could be right or could be wrong.

There wouldn't be any need to seal anything. Dunno where you are getting that from.

ESD damage is not a theory, like was previously mentioned. But as you can imagine, it's one of those things that is hard to prove in most situations due to the equipment and training required to actually produce the evidence (like having access to an electron microscope and the needed training to operate it and have the knowhow of what exactly to look for). Your 'demands' for evidence of this is quite unreasonable. Elcos have little to do with ESD damage. But ESD damage is all but just a theory, as you seem to be claiming.

And perhaps you are just playing ignorance but we all know that this 20 or 30 year old hardware can be quite expensive, which make ESD packaging look like much more of a good investment and you know this.

Whats missing in your collections?
My retro rigs (old topic)
Interesting Vogons threads (links to Vogonswiki)
Report spammers here!

Reply 35 of 71, by imi

User metadata
Rank l33t
Rank
l33t
Miphee wrote on 2021-02-16, 11:12:

I also "discharge myself" sometimes but my point stands. A single bag is useless when it's off the hardware and there is nothing that discharges your body. That current has to go somewhere and that will be the hardware you touch if there is no efficient grounding present. The ESD bag is like an RCD without proper grounding, won't work when needed.

I don't really know how people know that ESD was the reason a certain hardware failed. Do all of them work in a government science lab to confirm that?

well I felt and saw it happen in real time with my own eyes, idk what more I can tell you not sure if I need lab equipment for that, if you don't believe that go ahead :p

even unsealed ESD bags will offer you some benefit because they don't build up charge themselves when storing stuff together, imagine having all your cards in normal plastic bags next to each other, just moving stuff around would constantly build up charge, but yes for discharge protection an ESD bag has to be sealed, but that can be as simple as folding the top over.

Reply 36 of 71, by auron

User metadata
Rank Oldbie
Rank
Oldbie
bloodem wrote on 2021-02-16, 05:56:
In general, there is no need to pry off the heatsink, because it’s not usually glued to the chip. The only cards I came across w […]
Show full quote
auron wrote on 2021-02-16, 01:18:

it's not always a good move; in fact when doing it results in prying off the heatsink with force and damaging the BGA, it's not a good move at all. there's been posts with examples of this here and in general it has probably happened more than people care to admit, because this TIM replacement thing is stuck as "common sense" now, as indicated by your post.

In general, there is no need to pry off the heatsink, because it’s not usually glued to the chip. The only cards I came across with glued heatsinks are Voodoo 3 (and custom Voodoo 1 / Voodoo 2), some early GeForce 256 / Creative & Visiontek GeForce 2 GTS and a Gainward GeForce FX 5900.
I also typically replace the thermal paste (and sometimes the cooler altogether) on all of my cards immediately after the initial testing, and it’s usually a pretty easy process that takes 10 minutes.
Now, when it comes to glued heatsinks, I usually don’t bother. I find that the heat transfer is still very good anyway, so I just add a fan or replace/lubricate the existing one.

excluding flip-chip open die parts, and heatsinks that are glued on without any other support, the actually sensible thing to use on old low TDP hardware is a suitable thermal pad, perhaps - it's easier to apply, keeps the surface clean and you won't have to worry about your replacement thermal paste drying out in the future. i use one on my k6-233 which is a nearly 30W part, and it heats up the s370 cooler on there quite well, so i'd say there's no reason why it wouldn't work even better on lower TDP parts, such as these graphics cards.

i wouldn't be surprised if that (crappy) thermal paste on some of that old PC hardware was just a cost cutting measurement. for reference, sony used thermal pads on at least some of the ps2 consoles - same ballpark TDP and BGA package.

Reply 37 of 71, by Velociraptor

User metadata
Rank Member
Rank
Member

We're still not really any further forward in working out what kills these cards. I've seen it put to any of the following, but finding out what's likely and unlikely is difficult.

* Memory chips failing over time
* BGA soldering failing
* Other solder joints failing
* ICs like the GPU just dying
* ESD causing damage there and then or that later manifests
* Heat

Anecdotally it seems to be the higher value cards from the mid 00s that get it worse. The 9800pros, the 4200 ti and faster on AGP. But that doesn't mean it is only them, and it doesn't mean it's because they're taking more volts/watts than the other cards, although it might. It could just be that people aren't too interested in the cheap, abundant value cards and only interested in the higher end ones so only comment when a higher end card dies.

I will have a couple of these cards running in systems I've built recently and I'd very much like these cards to last for as long as possible. So for me that means I'll remove the heatsink, so long as it's not glued, replace the stock cooler with a big passive Zalman one, strap a fan to it so it runs cold, put heatsinks on the memory, eventually get a thermal imaging device of some kind and find hotspots, make sure cases have good airflow, not overclock the cards - perhaps even underclock.

I do use a grounded ESD mat. I've been grounding myself before I do anything, and when I take things from an anti-static bag they go onto the mat. I'm upgrading to proper earth bonding not crocodile clips. It's not terribly expensive and won't make any difference to how difficult things are. I have also ordered ESD gloves but I have a suspicion they might be too awkward.

Reply 38 of 71, by douglar

User metadata
Rank Oldbie
Rank
Oldbie

ESD isn’t so much a problem for me in the summer, but I get nervous about it in the winter when humidity drops. I can imagine that people who live in warmer, humid climates might not see the issue at all. For me, I had to stop using my rolling chair becuase it was a little van der graaf generator. I replaced it with grounded old Tower case. That seems to assist with the static reduction pretty well before I touch anything on the workbench. I do need to get mosr static bags though, for the dust if nothing else.

I think the most ESD vulnderable components I’ve seen were the nforce 6100 mtx motherboards. The usb controller on that chipset used to fail with the slightest provocation.

For the Geforce 4xx series, back in the day, mine were prone to slow death by over heating. The didn’t last without improved cooling.

Reply 39 of 71, by Miphee

User metadata
Rank Oldbie
Rank
Oldbie
imi wrote on 2021-02-16, 21:20:

well I felt and saw it happen in real time with my own eyes, idk what more I can tell you not sure if I need lab equipment for that, if you don't believe that go ahead :p

I haven't seen it ever and I handle a lot of stuff. But as I mentioned before, I'm only 90% against all that craze that surrounds ESD.
An ESD bag is good as a physical protection against scratches but useless against ESD when the hardware needs to be used.
You also bought an ESD mat because of it, right? The excess current has to be grounded or it will go to whatever you touch.
And that's what I'm saying, ESD protection is useless without grounding.
I don't see why it's so hard to accept. People buying these bags by the hundreds is a waste of money.
OP said he used those cards before they failed one by one so if it was really ESD then a bag wouldn't have protected those cards either.
Equipotential bonding with excellent grounding (less than 5 Ω if possible) is the key and it's mandatory in every household and workshop (here).