Overclock.net - Overclocking.net
     
 
Home Gallery Reviews Blogs Register Today's Posts Mark Forums Read Members List


Go Back   Overclock.net - Overclocking.net > Graphics Cards > NVIDIA

Reply
 
LinkBack Thread Tools
Old 09-03-08   #1 (permalink)
Commodore 64
 
ReoEagle's Avatar
 
intel

Join Date: May 2008
Posts: 618

Rep: 45 ReoEagle is acknowledged by some
Unique Rep: 42
Trader Rating: 0
Default High lead? How can that be the problem

Ok, out of this whole confusion with nVidia solder job being the problem. I have heard from a friend that High-lead is a amazing and a major standard and why the RoHS has actually has an exempt on it. Now to be honest, I don't fully understand why the process could be bad, and I have a very short attention span.
But I really would like an explanation about this
Quote:
Originally Posted by ReoEagle View Post
I thought that to be RoHS compliant you need to use a lead-free solder or minimal amount of lead? It's exempt

This states that it's been an industry standard for quite some time
I looked at it closer and eutectic solders have a lower melting point than high lead solder, I have a short attention span

Apparently it only has one lead-free equal and it's not large in the market place.
IBM intros C4NP

Please correct me if I'm wrong about this but how could a tried and true process thats been used for years, (and only recently found it's equal with no lead) and is one of the few exempt process from the RoHS with high lead-content be the absolute failure to the chip?

Quote:
Originally Posted by DuckieHo View Post
ATI switch to eutectic pads already...

http://www.tgdaily.com/content/view/39045/135
To be honest, all I can gather of what a eutectic mixture is just one that cools all at about the same time, with tin and lead being the most common, and I believe is currently allowed, though there better quality ones, they're just far more expensive.

Please tell me if I'm being dumb about this, as it's kind of hard for me to believe that High lead soldering is the problem as it is a defacto standard in the industry. =(
__________________
System: Birdies4
CPU
E8400 @ 3.6 v1.26 Q808
Motherboard
eVGA 750i FTW
Memory
Corsair DHX 800
Graphics Card
SLI 8800 GTS (G92)
Sound Card
X-Fi XtremeGamer
Power Supply
750 Toughpower
Case
=(
CPU cooling
9700LED
OS
Haiku / XP Pro
Monitor
P1110 Aperture grille = <3
ReoEagle is offline   Reply With Quote
Old 09-03-08   #2 (permalink)
Commodore 64
 
ReoEagle's Avatar
 
intel

Join Date: May 2008
Posts: 618

Rep: 45 ReoEagle is acknowledged by some
Unique Rep: 42
Trader Rating: 0
Default

Seriously if anyone could explain it to me, I would really like it. As I really don't like people bashing in nVidia and giving a technical reasoning that is flawed.
__________________
System: Birdies4
CPU
E8400 @ 3.6 v1.26 Q808
Motherboard
eVGA 750i FTW
Memory
Corsair DHX 800
Graphics Card
SLI 8800 GTS (G92)
Sound Card
X-Fi XtremeGamer
Power Supply
750 Toughpower
Case
=(
CPU cooling
9700LED
OS
Haiku / XP Pro
Monitor
P1110 Aperture grille = <3
ReoEagle is offline   Reply With Quote
Old 09-03-08   #3 (permalink)
Overclocker in Training
 
Raiden's Avatar
 
intel nvidia

Join Date: Jul 2008
Posts: 7

Rep: 0 Raiden Unknown
Unique Rep: 0
Trader Rating: 0
Default

I am definitely no chemist, but lead+tin are used to create a conductive metal that melts easily when enough thermal heat is applied. Usually they have rosin as well to make solder "stick" to metal contacts that you would find on a circuit board. Non-lead (and thus RoHS compliant?) solder requires higher temperatures, but also helps out the environment, and gives better results at higher temperatures where solder can melt away. (The article I read said that this was needed for high temperature automotive components.)
__________________
System: Custom Build Something!
CPU
E8500 Wolfdale 3.16GHz
Motherboard
GIGABYTE GA-EP35-DS3L
Memory
G.SKILL 4GB (2 x 2GB) DDR2 SDRAM DDR2 1066
Graphics Card
EVGA GeForce 9800 GTX+ 512MB SSC Edition
Hard Drive
HITACHI 500GB 7200 RPM SATA 3.0Gb/s
Sound Card
Sound Blaster Audigy SE 7.1 Channels
Power Supply
Antec TPQ-850 850W 80PLUS Bronze Certified
Case
Antec Nine Hundread Black Steel
CPU cooling
ARCTIC COOLING Freezer 7 Pro 92mm
OS
Vista Ultimate SP1 64 bit
Monitor
Acer Black 22" 5ms Widescreen LCD Monitor
Raiden is offline   Reply With Quote
Old 09-04-08   #4 (permalink)
Commodore 64
 
ReoEagle's Avatar
 
intel

Join Date: May 2008
Posts: 618

Rep: 45 ReoEagle is acknowledged by some
Unique Rep: 42
Trader Rating: 0
Default

Quote:
Originally Posted by Raiden View Post
I am definitely no chemist, but lead+tin are used to create a conductive metal that melts easily when enough thermal heat is applied. Usually they have rosin as well to make solder "stick" to metal contacts that you would find on a circuit board. Non-lead (and thus RoHS compliant?) solder requires higher temperatures, but also helps out the environment, and gives better results at higher temperatures where solder can melt away. (The article I read said that this was needed for high temperature automotive components.)
High-lead solder melts at over 300C, which is a lot more than most alloy solders, but is radically more than your average run of the mill low-lead (or light lead) solder at your local store or used for most non-computerized parts.

I had a long discussion with an engineer, he said it seemed unlikely but entirely plausible that it could be the problem, it would certainly be really nice to see someone that isn't charlie talk about it and show with direct evidence that there is a problem at that level of heat and electromiagration.

Also most things need to have minimalist amounts of lead to qualify, high-lead solder because of it's efficiency with computer parts is SPECIFICALLY exempt by the RoHS.

Read the papers I put in the quote please, and you'll see what I mean, if you'd like me to get more information about it, I'll gladly get some more. But I may have to talk to a few more engineering friends about this and make an counter article because it's starting to seem unlikely, but like I said, this isn't extactly my area of expertise, and I could be explaining it wrong to my friends in the first place.
__________________
System: Birdies4
CPU
E8400 @ 3.6 v1.26 Q808
Motherboard
eVGA 750i FTW
Memory
Corsair DHX 800
Graphics Card
SLI 8800 GTS (G92)
Sound Card
X-Fi XtremeGamer
Power Supply
750 Toughpower
Case
=(
CPU cooling
9700LED
OS
Haiku / XP Pro
Monitor
P1110 Aperture grille = <3

Last edited by ReoEagle : 09-04-08 at 05:45 AM
ReoEagle is offline   Reply With Quote
Old 09-04-08   #5 (permalink)
Commodore 64
 
ReoEagle's Avatar
 
intel

Join Date: May 2008
Posts: 618

Rep: 45 ReoEagle is acknowledged by some
Unique Rep: 42
Trader Rating: 0
Default

So case in point. nVidia chips somehow maybe flawed but the reasoning that Charlie gave was a load, unless someone can prove otherwise. =)
__________________
System: Birdies4
CPU
E8400 @ 3.6 v1.26 Q808
Motherboard
eVGA 750i FTW
Memory
Corsair DHX 800
Graphics Card
SLI 8800 GTS (G92)
Sound Card
X-Fi XtremeGamer
Power Supply
750 Toughpower
Case
=(
CPU cooling
9700LED
OS
Haiku / XP Pro
Monitor
P1110 Aperture grille = <3
ReoEagle is offline   Reply With Quote
Old 09-04-08   #6 (permalink)
Boinc Boinc Boinc
 
grunion's Avatar
 
intel ati

Join Date: Sep 2006
Location: Indiana
Posts: 18,024

Folding Team Rank: 29
Trader Rating: 68
Default

Quote:
Originally Posted by ReoEagle View Post
So case in point. nVidia chips somehow maybe flawed but the reasoning that Charlie gave was a load, unless someone can prove otherwise. =)

Who cares what Charlie writes, seriously?

Fact of the matter is Nvidia took a $196 million charge, proof enough that there is an issue.
Until Nvidia states the facts I suggest you don't take Charlies rants personally.

System: Emergency Deep
CPU
E8400/500*9
Motherboard
DFI LT X48-T2R
Memory
G.Skill PI
Graphics Card
XFX 5870/XFX 295 Rocking PhysX by NV
Hard Drive
RaptorX/RAID0/Blacks/RAID0
Sound Card
X-FI Platinum
Power Supply
Ultra X3 1000
Case
Armor Jr.
CPU cooling
Megahalems > All
GPU cooling
Stock
OS
7 Ultima
Monitor
204b
Overclock.net - 2009 Chimp Challenge Champions 6 Million+ Folding at Home points
grunion is offline I fold for Overclock.net Overclocked Account   Reply With Quote
Old 09-04-08   #7 (permalink)
Tigers > Lions
 
intel nvidia

Join Date: Apr 2006
Location: In fear of banhammers
Posts: 3,881

Rep: 344 TheSubtleKnife is a proven memberTheSubtleKnife is a proven memberTheSubtleKnife is a proven memberTheSubtleKnife is a proven member
Unique Rep: 263
FAQs Submitted: 4
Trader Rating: 12
Default

I think it might have to do withe the fact thay are using High-Lead balls, but eutectic pads, so the eutectic pads expand faster than the lead, but they are bonded, so it causes the lead to crack? and maybe all the other applications are high lead on high lead connections, so thats why they dont fail...

just my $.02
__________________
Need a quick question answered or you just want to hang out? Why dont you hit me up in the OCN CHAT?

System: PURE ACTION!
CPU
E7200 3.8Ghz
Motherboard
Asus P5Q Pro
Memory
2GB Mushkin + 2GB Ballistix
Graphics Card
eVGA 8800GTS 800/1950/1050
Hard Drive
320 + 120 + 80 GB
Sound Card
8 Channel
Power Supply
700W GameXstream
Case
Antec 1200
CPU cooling
Stock
GPU cooling
Stock
OS
Windows 7 x64
Monitor
22" Acer WS
TheSubtleKnife is offline Overclocked Account   Reply With Quote
Old 09-04-08   #8 (permalink)
PC Gamer
 
Bunnywinkles's Avatar
 
amd ati

Join Date: Jul 2008
Location: Ohio
Posts: 1,258

Rep: 57 Bunnywinkles is acknowledged by some
Unique Rep: 55
Trader Rating: 13
Default

Quote:
Originally Posted by TheSubtleKnife View Post
I think it might have to do withe the fact thay are using High-Lead balls, but eutectic pads, so the eutectic pads expand faster than the lead, but they are bonded, so it causes the lead to crack? and maybe all the other applications are high lead on high lead connections, so thats why they dont fail...

just my $.02
hes right, they didnt use the same mixture, so when the card heats up the chip expands unevenly, which in turn, can crack the lead balls.

http://www.overclock.net/hardware-ne...explained.html

read. i know its INQ but all seems right..
__________________

Quote:
Originally Posted by lordikon View Post
In summary, it is a GTX250/9800GTX+/OC'd 9800GTX/55nm OC'd 8800GTS g92/55nm OC'd 8800GT. Nothing confusing about it.

System: Elga Lynn Borschavich
CPU
AMD Phenom II 710 X3 *3.6Ghz*
Motherboard
Gigabyte 790X
Memory
2x2GB OCZ DDR3 1600
Graphics Card
HIS 4850 ICEQ4 1GB
Hard Drive
2x Segate Barracuda 500GB 7200.11 *RAID 0*
Sound Card
Onboard Dolby Digital
Power Supply
PC P&C 610w
Case
Antec Twelve-Hundred
CPU cooling
HDT-S1283 *Push/Pull*
GPU cooling
Stock
OS
Windows 7 Professional 64-bit and Ubuntu 9.10
Monitor
LG 22" WS
Bunnywinkles is offline   Reply With Quote
Old 09-04-08   #9 (permalink)
Commodore 64
 
ReoEagle's Avatar
 
intel

Join Date: May 2008
Posts: 618

Rep: 45 ReoEagle is acknowledged by some
Unique Rep: 42
Trader Rating: 0
Default

Quote:
Originally Posted by TheSubtleKnife View Post
I think it might have to do withe the fact thay are using High-Lead balls, but eutectic pads, so the eutectic pads expand faster than the lead, but they are bonded, so it causes the lead to crack? and maybe all the other applications are high lead on high lead connections, so thats why they dont fail...

just my $.02
It's common enough the two are mixed, and usually within tolerance, if it was really a problem, you'd heard about it with other BGA chips

System: Birdies4
CPU
E8400 @ 3.6 v1.26 Q808
Motherboard
eVGA 750i FTW
Memory
Corsair DHX 800
Graphics Card
SLI 8800 GTS (G92)
Sound Card
X-Fi XtremeGamer
Power Supply
750 Toughpower
Case
=(
CPU cooling
9700LED
OS
Haiku / XP Pro
Monitor
P1110 Aperture grille = <3
ReoEagle is offline   Reply With Quote
Old 09-04-08   #10 (permalink)
Commodore 64
 
ReoEagle's Avatar
 
intel

Join Date: May 2008
Posts: 618

Rep: 45 ReoEagle is acknowledged by some
Unique Rep: 42
Trader Rating: 0
Default

Quote:
Originally Posted by grunion View Post
Who cares what Charlie writes, seriously?

Fact of the matter is Nvidia took a $196 million charge, proof enough that there is an issue.
Until Nvidia states the facts I suggest you don't take Charlies rants personally.
It's just not me, I'm just getting sick of talking to some people that say all nVidia chips are gonna die, and I feel like everyone on here including people that like nVidia, especially the ATI enthusiast, feel that are as well. I just kind of want evidence to state otherwise, because I see people just taking it as pure fact.

I have nothing against ATI, just their stuff with me doesn't seem to do all too well (Fusing to my Molex connector for one! =P)

Though, their ATI's motherboards, I gotta say were some of the most rocksolid boards I've ever used, (Though I've only used Intel ones). Too bad AMD bought 'em =(

System: Birdies4
CPU
E8400 @ 3.6 v1.26 Q808
Motherboard
eVGA 750i FTW
Memory
Corsair DHX 800
Graphics Card
SLI 8800 GTS (G92)
Sound Card
X-Fi XtremeGamer
Power Supply
750 Toughpower
Case
=(
CPU cooling
9700LED
OS
Haiku / XP Pro
Monitor
P1110 Aperture grille = <3
ReoEagle is offline   Reply With Quote
Reply


Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools



All times are GMT -5. The time now is 05:48 AM.


Overclock.net is a Carbon Neutral Site Creative Commons License

Terms of Service / Forum Rules | Privacy Policy | DMCA Info | Advertising | Become an Official Vendor
Copyright © 2009 Shogun Interactive Development. Most rights reserved.
Page generated in 0.18916 seconds with 8 queries