NVIDIA Updates Chip Package Materials, 55nm GPUs subject to Changes too

[btarunr]

A product change notification (PCN) document by NVIDIA The Inquirer claims to have access to, indicates changes to the bump materials of several NVIDIA graphics processors (GPUs). Affected by this change are popular GPUs such as G92 and G92b (55nm). Changes include replacement of a High-Pb solder (95% Pb / 5% Sn) bump material with Eutectic Solder (63% Sn / 37% Pb). Bumps are those parts of the die that establish electrical contact with the leads/pins of the FC-BGA package. Failures of these bumps are irreparable leading to permanent damage. This follows several events that lead to NVIDIA owning up defects in certain mobile graphics and MCP parts.

Implications of this PCN are:

• Current G92 and G92b are weaker and could be subject to failures similar to those products already diagnosed with failing packages and official announcements issued.
• This could just be a precautionary measure by NVIDIA since these changes according to the PCN are aimed "to increase supply and enhance package robustness" according to the PCN. In other words, better safe than sorry.
• Sourcing bump-processing services from a single provider is more economical than several providers doing it, as was the case with the recent mobile GPU failure fiasco. Mobile GPUs aren't much different from regular ones per say.

A list of affected products, according to the PCN is provided. It carries the PCN number of PCN0346A. It has the "PCN Submit Date" of June 13, 2008, "Planned Implementation Date" of July 28, 2008, and a "Proposed First Ship Date for change" of August 17, 2008 which makes it likely that none of the products available in the market as of now use this "robust" silicon packaging material. The G92 and G92b processors go into making popular graphics cards such as GeForce 8800 GT, 8800 GTS 512M, 9800 GT, 9800 GTX, 9800 GTX+, etc.



View at TechPowerUp Main Site

 

[OnBoard]

So any added warranty, or means to prevent this? Those stock cooler early versions (8800GT) are in more risk, if heat is a factor?

 

[tkpenalty]

High-PB? I smell BS here, since Nvidia, and their partners are under RoHS, which means NO lead in anything that they use.

 

[largon]

RoHS "Lead-free" does not mean the product doesn't contain lead...
FCPGA junction solder is specifically exempted in the directives.

 

[nafets]

Here's a quote from a user at Slashdot, regarding this issue, which may clear up things;

"Yes there is an exception in RoHS for lead solder that has a high melting point. However, the official RoHS rule is that while lead solders in general are prohibited, there is an exception allowing for the use of lead solder that contains at least 90% lead. The idea being that solder with at least 90% lead melted at a higher temperature and was at least somewhat safer if disposed of improperly. Otherwise, potentially there may also have been no replacements for high lead content solders that performed as well when the first RoHS directives were drawn up in 2003. Currently (2008), however, there are lead-free solders that would work, but the lead free solders are more expensive than lead based solders (by roughly three times). Using a lead-free solder with a significantly different composition may also require a new packaging design and another extensive round of qualification, too. I am not totally sure how this would be done.

It get worse, the new solder to be used by nVidia mentioned in this Inq article states that it will only contain 63% lead and 37% tin, making nVidia based cards with this solder not saleable to consumers in the EU according to RoHS directives. The replacement 63Pb/37Sn solder has a somewhat better tensile strength and a lower coefficient of thermal expansion than the older 95Pb/5Sn solder, which may be why nVidia chose this route to fix the problem. Whether nV will be selling very many products in the EU with this fix and whether this will correct the problems, is another issue."

Whether or not it's correct or not, I am not sure, but it seems fairly logical...

 

[largon]

Clearly, the random user rolleyes at Slashdot is too confused about this for it to be worth to pay them any attention...
Anyone can download the RoHS directives & see there's no such silly minimum lead percentage as the random user at /. claims.

 

[tkpenalty]

Clearly, the random user rolleyes at Slashdot is too confused about this for it to be worth to pay them any attention...
Anyone can download the RoHS directives & see there's no such silly minimum lead percentage as the random user at /. claims.

Someone knows his stuff.

RoHS means NO LEAD.

 

[largon]

RoHS means NO LEAD. Except for the following applications:
[huge list]

Fixed.

Guidance on the specific applications of lead, mercury, cadmium, hexavalent chromium and PBDE that are exempt from the requirements of the RoHS Regulations:

(...)

17. Lead in solders to complete a viable electrical connection between semiconductor die and carrier within integrated circuit Flip Chip packages.
Flip chips are attached to their packages or PCBs using very small solder bumps and many types use solder bumps containing lead. Lead is used for two main reasons. Its ductility reduces the risk of damage to brittle parts of flip chip circuitry. Lead also protects against the possibility of thermal fatigue, which results from cyclic temperature changes and is not well understood with lead-free solders. High melting point solder bumps are attached using solder containing typically 37 – 40% lead to the package because this combination has a high resistance to a phenomenon called “electromigration” which in higher power flip chip packages would otherwise cause premature failure of the device. The solder connections to the chip are known as level 1 and level 1 flip-chip connections may contain lead. (...)

Quote from the site of the regulatory department in UK as official EU site (rohs.eu) is broken atm.

 

[DaedalusHelios]

I knew solder had exemptions but..... huge list. Wow, I thought solder was the only exemption covering lead. I will visit rohs.eu soon.

Thanks for the linkage.

 

[btarunr]

So any added warranty, or means to prevent this? Those stock cooler early versions (8800GT) are in more risk, if heat is a factor?

If there are 1000 GeForce users and 100 Radeon users, if 20 Radeon cards fail, it becomes a big issue, cause for concern...but if 20 GeForce cards fail, it isn't as big, a smaller proportion failed. Part of Charlie's usual poop-o-rama was that NV goes bankrupt if it's made to replace existing G92 parts. Take it with a crate of salt.

Be optimistic, my G92 survived the Indian summer of 45 C room-temp (when the AC failed for a week). It's full of heart.

 

[[I.R.A]_FBi]

full of love or full of ?

 

[largon]

The thing is, there's no reason whatsoever to believe the PCN would have anything to with any type of defect. It's all in Charlie's head.

 

[btarunr]

The thing is, there's no reason whatsoever to believe the PCN would have anything to with any type of defect. It's all in Charlie's head.

I feel the same, it's just precautionary or enhancement. Instead it's an incentive, they're using superior materials now on. I've never heard of a G92 card fail. I remember Charlie whine even back when it was announced 8800 GT cards would use PCBs with lesser layers (because they weren't required, might as well cut costs).

 

[tkpenalty]

I see the G92s dying for reasons OTHER than the core itself... yeah I have to agree, its in charlie's head =_=

 

[largon]

(...) they're using superior materials now on. (...)

Actually, they're moving to tecnically inferior bump solder. The higher the amount of lead, the better.
The new stuff which has less Pb and more Sn:
- melts at a lower temp
- is more brittle
- has comparably worse thermal chars

 

[btarunr]

Actually, they're moving to tecnically inferior bump solder. The higher the amount of lead, the better.
The new stuff which has less Pb and more Sn:
- melts at a lower temp
- is more brittle
- has comparably worse thermal chars

Then what makes the product robust

Why did they switch to this?

 

[largon]

The internet tells me such solder flows more readily thus is easier to play with. And I guess they're aiming to reduce lead usage - just like the whole industry.

 

[TheGuruStud]

Too hot = joints come apart from the solder weakening and/or too much expansion and contraction with temp fluctuation.
Which is why the laptop ones are vulnerable b/c no one cools them for shit.

Nvidia's is changing it to prevent this from happening (stronger solder).


That's what I gather from all of this.

 

[PP Mguire]

My 8800GS runs 80-85c 24/7 and i havent had any problems. My room is usualy about 90f all the time. Charlies head i say.

 

[OnBoard]

Be optimistic, my G92 survived the Indian summer of 45 C room-temp (when the AC failed for a week). It's full of heart.

Ouch, that's hot Mine ides at 47C and 5 mins with FurMark heats it up to 74C (the higher of GPU-Z temps). Not going to die on me then, not that I was worried, but never hurts to be cautions

 

[btarunr]

Ouch, that's hot Mine ides at 47C and 5 mins with FurMark heats it up to 74C (the higher of GPU-Z temps). Not going to die on me then, not that I was worried, but never hurts to be cautions

With the reference fan at 100%, the GPU would idle (rendering Microsoft Windows) at 65 C. Our summers can be as harsh as your winters

 

[largon]

Too hot = joints come apart from the solder weakening and/or too much expansion and contraction with temp fluctuation.
Which is why the laptop ones are vulnerable b/c no one cools them for shit.

Nvidia's is changing it to prevent this from happening (stronger solder).

It's the exact opposite. They're moving into solder that melts at lower temp.

 

[OnBoard]

Our summers can be as harsh as your winters

Bad for people, good for overclocks

 

[TheGuruStud]

It's the exact opposite. They're moving into solder that melts at lower temp.

Then it's more flexible and won't break away. I didn't say melting, anyway

 

[DarkMatter]

I must agree with the guru. Probably it follows the same principle as anti-hearthquake building structures. More flexible, somehow "weaker" but it doesn't break and you get no (micro)fisures.