LGA 1155 Socket Burn: here we go again

Quote:

---

Last year, credible reports of a design defect with LGA1156 sockets made by a certain component manufacturer in particular, made headlines. Reviewers found that extreme overclocking using increase voltages, in test cases, caused certain pins of the socket to burn out, damaging both the board and processor. The defect was found to be caused by shorting between the pins and the socket causing tiny electrical arcs. Motherboard reviewer from TechReaction.net discovered that his Gigabyte GA-P67A-UD4 and GA-P67A-UD7 samples showed signs of socket burn. The publication is yet to receive a reply from the board manufacturer.

---

http://www.techpowerup.com/img/11-01-17/129a.jpg

http://www.techpowerup.com/img/11-01-17/129b.png

http://www.techpowerup.com/img/11-01-17/129c.jpg

http://www.techpowerup.com/img/11-01-17/129d.jpg

Quote:

---

LGA1155 and LGA1156 are very similar in physical pin layout, chip package, and retention clips, although the pin-maps are significantly different. TechReaction comments that while only "extreme conditions" cause such pin burnouts, 24/7-stable overclocked settings could, over a period of time, create similar conditions. Only time, and extensive testing will reveal if burnouts are a cause for concern to enthusiasts.

---

techPowerUp!

TechReaction

Now have got my fingers crossed!

i wonder how many volt's they where using on the OC

There is thread about that... He got it like that from Gigabyte, but again I doubt they used 1.6 - 1.7v

Just now waiting for my UD5 to arrive, and suddenly my OCD is kicking in like crazy. :shocked:

This is going to drive me insane lol.

Let's hope it's an isolated incident.

This is a major let down if it keeps going on for yet another socket.
Bet this will put off a lot of enthusiasts.

Wonder why everytime are gigabyte boards emerging in that

Lol @ gigabyte fanboys once they see this thread

Foxconn again????

whaaaaaaaaaaaaaa how about my UD7@!!!!

I sincerely hope this is an isolated case, as it seems.

how many volt??

1.6? ok...

1.4? not ok... :mad:

I think it's more a problem of the current than the volts ;) Looks to me as if they heated up too much. An indication for this is the discolouration.

There is no solution, is a curse for their "socket madness" strategy :D

Quote:

---

Originally Posted by FischOderAal

I think it's more a problem of the current than the volts ;) Looks to me as if they heated up too much.

---

agreed

higher voltage dosnt burn away metal, higher current which turns into higher temps does.

Wich mainbord on store(review mobo is not the same) have lotes socket?

Great, Gigabyte really needs to get their :banana::banana::banana::banana: together, or was it Foxconn or whoever is responsible for this mess really needs to shoot themselves for creating another disaster. First it was P55 boards and they have carried the exact same issue to P67, what a joke. I just invested in a P67 setup myself and I'm not happy to hear this.

And what is if it isn´t a socket problem?

Quote:

---

Originally Posted by miahallen

I have not used the board at all....I received it in this condition from Gigabyte. My M4E and UD7 are also Foxconn sockets, and have had heavy use without any issues :shrug:

---

maybe a production failure and bad quality check @ gigabyte..

the PINs are 1 VCCIO and 3 VSS.

My UD4 has shown no sign of this problem <yet>
And none of my P55 boards had this either.

Trying to remember but with the P55 sockets didn't this problem happen to other manufacturers motherboards also and was not limited to Gigabyte ?

Let's not blow this out of proportion yet, it's only been seen on 2 motherboards by the same guy, not a widespread issue. Before everyone attacks Gigabyte with their pitchforks in hand why don't we wait to see if this is a real problem or just an isolated incident.

socket can't take that many awesome gigahurts

Quote:

---

Originally Posted by BlueAqua

Let's not blow this out of proportion yet, it's only been seen on 2 motherboards by the same guy, not a widespread issue. Before everyone attacks Gigabyte with their pitchforks in hand why don't we wait to see if this is a real problem or just an isolated incident.

---

Agreed..:up:
Now I don't push my systems to the degree some of you Xtreme people do but I've benched the Intel board with 1.525vcore up to 5088MHz and seen no issues.
What I'd like to know is what voltages with what cooling this guy was using when this happenned.

Isnt it a bit early to go bananas over this? It may just be a single incident, and may be related to other reasons...

jarle

Quote:

---

The defect was found to be caused by shorting between the pins and the socket causing tiny electrical arcs.

---

For the sake of accuracy, let's just get this straight: This is not the reason because "shorting between the pins" didn't happen. That's absurd. What would happen here is that excessive current would have indirectly caused some pins to overheat. Heat is the product of power dissipating in a material due to its resistance versus how much voltage is being applied (which determines how much current there is). The first thing to know is that physical contact points like that in any connector are high resistance joints versus plain metal (wire, pins, etc). The comparatively higher resistance causes more power to be dissipated at that contact point that in the lower part of the pin. This is further complicated by higher temperatures increasing the resistance of the metal. As either voltage is increased or resistance of the load is decreased (when the CPU is doing more work), more current flows. This causes increased heating proportional to each link in the chain's resistance provided the voltage is constant. In this particular case, the high resistance mechanical junction was the first to fail. More heat was being dissipated than the metal could handle and it essentially melted.

This same effect can be found on connections that are dirty but otherwise normal, such as extension cords. Dirty plugs make for hotter connections. If it gets bad enough, things will eventually burn up. This is also the reason why you have to select a wire that's big enough to handle the amount of current you intend to transfer across it instead of just using the same size of wire for everything from cell phone chargers to car engine starters. The metal has a little bit of resistance to it and you must select a conductor that will not overheat. This is also why you must use a bigger aluminum wire than you would a copper wire. Aluminum has more resistance per distance unit per cross-sectional unit, so it heats more for a given size and distance. If a metal had literally no resistance, you could quite literally use a hair-sized wire to power a city.

Going banana over every single incident is what people do on these forums nowadays, heh

Glad i've yet to of made any upgrade moves for a few years!!!

Surly if the pin is touching the pad on the CPU there should be no problem, but clearly it's not for it to ARC. However is it due to the high voltage, well more amp and poor socket pin material at fault

These forums? Pretty much everywhere...
Quick google search still only shows techreaction.

Problems like this may very well have been there even 5 years ago, though no1 discovered it then and was accepted like "faulty mobo", that's the only difference. Now if there's even as much as slight HINT to socket burn problem it'll be blown out of proportion. That's my view on this.

Also it's not Gigabyte's fault but the company that creates the socket.

Quote:

---

Originally Posted by Dahmer

These forums? Pretty much everywhere...
Quick google search still only shows techreaction.

---

Yeah, you are right.
More and more people are becoming victims of hypertension.

I thought that design flaw in socket design was the reason behind why they discontinued 1156 this hastily. If this proves not to be an isolated case, it would be kinda lame. Really.

my cpu contact pads are discolored..ive yet to check the socket.

Quote:

---

Originally Posted by Andrea deluxe

Wich mainbord on store(review mobo is not the same) have lotes socket?

---

If I remember right MSI and ASRock has Lotes for sure, ASUS has Foxconn I think

lol seems foxconn reused the sorted out S1556 and hoped noone would oc them... :rofl:

Quote:

---

Originally Posted by SubZero.it

If I remember right MSI and ASRock has Lotes for sure, ASUS has Foxconn I think

---

My Asus P8P67 Pro has a Lotes socket.

I experienced socket burn with LGA775 too :(

http://img195.imageshack.us/img195/2671/img4174wg.jpg

ASUS use LOTES & FOXCONN. It`s a lottery...

Quote:

---

Originally Posted by Hornet331

lol seems foxconn reused the sorted out S1556 and hoped noone would oc them... :rofl:

---

See, that's actually a good idea. But then why didn't they use the faulty sockets in H67 mobos which are pretty much guaranteed to be ran at stock clocks? :D Te only real advantage P67 has is namely overclocking IMO. :shrug:

actually there is no LOTES socket :shrug:

http://www.abload.de/img/socket0m8r.png

only the ILM (Independent Loading Mechanism) is from LOTES but the socket can be from every one ;)

These 2 incidents may be some isolated accidents based on user error, etc .. Specially after last round burn-outs, I bet they have learned a good lesson and won't repeat the same disaster.

But it doesn't hurt to take it easier on these new CPUs until we know more about this.

Quote:

---

Originally Posted by Chri$ch

actually there is no LOTES socket :shrug:

http://www.abload.de/img/socket0m8r.png

only the ILM (Independent Loading Mechanism) is from LOTES but the socket can be from every one ;)

---

Very interesting...

wasnt it the same way for the 1156 stuff? lotes made the top part and the actual socket was still foxconn?

Quote:

---

Originally Posted by Tetrafluorometh

Glad i've yet to of made any upgrade moves for a few years!!!

Surly if the pin is touching the pad on the CPU there should be no problem, but clearly it's not for it to ARC. However is it due to the high voltage, well more amp and poor socket pin material at fault

---

i still wanna see how small of an airgap there is to have a 1v arc...

Caused by more current and less current pads than the 1366 socket AFAIK.

So the potential still exists... people stopped talking about this 4 months after the LGA 1156 was released so I wouldn't worry about the LGA 1155 much.

Quote:

---

Originally Posted by RoadconeTuning

i still wanna see how small of an airgap there is to have a 1v arc...

---

Heh. You're wasting your breath. I explained what happened in detail on the last page and nobody cared.

Quote:

---

Originally Posted by Particle

Heh. You're wasting your breath. I explained what happened in detail on the last page and nobody cared.

---

I read it!

People also ignored:

Quote:

---

He got it like that from Gigabyte

---

Quote:

---

Originally Posted by Chri$ch

actually there is no LOTES socket :shrug:
only the ILM (Independent Loading Mechanism) is from LOTES but the socket can be from every one ;)

---

Exactly, Lotes dont have the permission to fab the sockets

Huum, as i think buy a 2600K on the end of the month, and as i have only 3-4 boards possible choice ( PCIex slot spaces ), UD7, Maximus IV, P8P67 WS Revolution. I hope get more infos soon.

Here's a good idea, STOP USING FOXCONN SOCKETS! I love it how gigabyte was even so low as to put foxconn sockets underneath lotes brackets on their top end motherboards. Stunts like that is the main reason some people swore it wasn't just foxconn sockets when in truth it was just slimy mobo makers trying to hide the shame once again. Using whatever socket is cheapest is fine in budget enthusiast boards, but not top end boards with monster pwm to the socket.

so I guess I am one of the only people who read that he GOT THE BOARD LIKE THIS FROM GIGABYTE. when someone gets a bad kit of RAM you don't see people saying that all RAM ever is bad...

I MIGHT be concerned about this if it was widespread like the 1156 problem was but this is one case and this is how he GOT IT. besides 1155 chips take overall much less power so the chances of this happening are even lower

So, was it a brand new board he got, or maybe was it a review sample that got back for another reviewer that "forgot" to mention, he burned the socket, huh?

Quote:

---

Originally Posted by Dainas

Here's a good idea, STOP USING FOXCONN SOCKETS! I love it how gigabyte was even so low as to put foxconn sockets underneath lotes brackets on their top end motherboards. Stunts like that is the main reason some people swore it wasn't just foxconn sockets when in truth it was just slimy mobo makers trying to hide the shame once again. Using whatever socket is cheapest is fine in budget enthusiast boards, but not top end boards with monster pwm to the socket.

---

Protip: read thread:

Quote:

---

Originally Posted by STEvil

Protip: read thread:

---

Did beforehand; the socket was burnt somehow. Nevermind the fact that you can clearly see it was burnt with a processor in it :P

So tell me this; how often does that happen and when has burnt socket been a common topic before foxconn started being complete s**t. I've head of all sorts of problems with boards being shipped with bent pins on 775, but never socket burn to the point of molten pins. This was caused by foxconn cutting corners no-matter how you spin it

Quote:

---

Originally Posted by Dainas

Here's a good idea, STOP USING FOXCONN SOCKETS! I love it how gigabyte was even so low as to put foxconn sockets underneath lotes brackets on their top end motherboards. Stunts like that is the main reason some people swore it wasn't just foxconn sockets when in truth it was just slimy mobo makers trying to hide the shame once again. Using whatever socket is cheapest is fine in budget enthusiast boards, but not top end boards with monster pwm to the socket.

---

Take 2:

http://www.abload.de/img/socket0m8r.png

Quote:

---

Originally Posted by BlueAqua

Let's not blow this out of proportion yet, it's only been seen on 2 motherboards by the same guy, not a widespread issue. Before everyone attacks Gigabyte with their pitchforks in hand why don't we wait to see if this is a real problem or just an isolated incident.

---

Only on one motherboard :rolleyes:

Quote:

---

Originally Posted by Hondacity

my cpu contact pads are discolored..ive yet to check the socket.

---

Keep us posted :up:

Quote:

---

Originally Posted by M.Beier

I experienced socket burn with LGA775 too :(

http://img195.imageshack.us/img195/2671/img4174wg.jpg

---

:rofl:

Quote:

---

Originally Posted by DarthShader

So, was it a brand new board he got, or maybe was it a review sample that got back for another reviewer that "forgot" to mention, he burned the socket, huh?

---

It came from Gigabyte HQ in TW....I do not know what it was put through before I got it. I'm not planning to test it (unless Gigabyte askes me too).
Gigabyte is sending me a replacement now.

I'm still waiting for more detail about how it was used before I got it. :yepp:

Quote:

---

Originally Posted by Petra

My Asus P8P67 Pro has a Lotes socket.

---

Same, full Lotes, on my ASUS P8P67-M PRO too.

Quote:

---

Originally Posted by Dainas

Did beforehand; the socket was burnt somehow. Nevermind the fact that you can clearly see it was burnt with a processor in it :P

So tell me this; how often does that happen and when has burnt socket burn been a common topic before foxconn started being complete s**t. I've head of all sorts of problems with boards being shipped with bent pins on 775, but never socket burn to the point of molten pins. This was caused by foxconn cutting corners no-matter how you spin it

---

Protip: Read thread again.

well I gotta say I never had problems with gigabyte MoBo's so far, however you never know what the future holds. But gigabyte mobo's have other advantages, But the boards made lately are too expensive to use crappy parts that do this, gigabyte needs to test better and kick suppliers out that are not up to par.
My :2cents:
terramir

Is this only with giga p67 boards?

ugh I'm not going to bother arguing over it. I know the details and do not care. My problem with is that this 1155 socket has the same godawful hair thin easly bent and distorted pins that were in the 1156 foxconn socket, and ultimately thats what makes these problems happen. I have personally seen a lotes 1156 socket next to a foxconn 1156 and the difference in pin integrity is night and day.

Quote:

---

Originally Posted by Dainas

ugh I'm not going to bother arguing over it. I know the details and do not care. My problem with is that this 1155 socket has the same godawful hair thin easly bent and distorted pins that were in the 1156 foxconn socket, and ultimately thats what makes these problems happen. I have personally seen a lotes 1156 socket next to a foxconn 1156 and the difference in pin integrity is night and day.

---

Lotes doesnt make a socket.

There, now you dont have to read it a 3rd time.

Quote:

---

Originally Posted by Chri$ch

actually there is no LOTES socket :shrug:

http://www.abload.de/img/socket0m8r.png

only the ILM (Independent Loading Mechanism) is from LOTES but the socket can be from every one ;)

---

But maybe having the bracket from Lotes gives the right pressure or something else. Correct me if I'm wrong

MSI P67 GD65 , LOTES as far as I remember, instaled it there two days ago, was the 'no plastic edition' phew ... but still running on 1.3V and with all C1E and Downclocking stuff, I think I'm safe.

Quote:

---

Originally Posted by STEvil

Lotes doesnt make a socket.

There, now you dont have to read it a 3rd time.

---

I admire your patience :ROTF:

Calm down people, it was an one-time thing, maybe it won't even happen again, ever! No need to jump at false conclusions about Foxconn sockets and Gigabyte boards.

Quote:

---

Originally Posted by STEvil

Lotes doesnt make a socket.

There, now you dont have to read it a 3rd time.

---

Why did you have to do that? I'm sitting here with half a bag of popcorn, and you just nailed the entertainment :mad:

First P55 Socket Problem was an issue from the Foxconn socket. Pin's was not solid enough and made short circuit.

Second problem was that some P55 Mobo's could deliver more Amps (~160a - some said 130a are enough to damage socket) than socket could handle - that is a fact and even Francois told us that: http://www.xtremesystems.org/forums/...=234723&page=3

Cause of that fact most time it happened with the overclocked i7 chips + HT + full load (and not only with foxconn socket - if i remember right some Tyco got also burned).

I think P67 will not get that big problem coz 2600k is only chip with HT and i heared (never tested one) that even @5GHz+ is only around ~110a

now where do all thos people get their lotes sockets from?

i thought lotes didn't produce intel LGA 1155 sockets?

each socket is made of 4 parts: backplate; top plate; mounting gear (hinges for the top plate + lever) and the socket itself


check the side of the socket itself to see if you really have a lotes socket; i'd be suprised if you find lotes written onto the socket itself

Quote:

---

Originally Posted by generics_user

now where do all thos people get their lotes sockets from?

i thought lotes didn't produce intel LGA 1155 sockets?

each socket is made of 4 parts: backplate; top plate; mounting gear (hinges for the top plate + lever) and the socket itself


check the side of the socket itself to see if you really have a lotes socket; i'd be suprised if you find lotes written onto the socket itself

---

I am willing to bet most people are seeing it on the top plate.... not the socket

THe LOTES is on the Metal plate that holds the CPU in place. So I presume the piece of metal is from Lotes and the rest is from someone else ?
But on the Top of the SOCET is engraved 155X LM , Which fits the description of LOTES from the Table ....
Also http://images.bit-tech.net/content_i...67a-gd65-1.jpg
I'm Fairly sure I See 115X LM.

Quote:

---

Originally Posted by Kondik

THe LOTES is on the Metal plate that holds the CPU in place. So I presume the piece of metal is from Lotes and the rest is from someone else ?
But on the Top of the SOCET is engraved 155X LM , Which fits the description of LOTES from the Table ....
Also http://images.bit-tech.net/content_i...67a-gd65-1.jpg
I'm Fairly sure I See 115X LM.

---

On 1155 LOTES only makes the ILM, not the socket.

Quote:

---

Originally Posted by Kondik

THe LOTES is on the Metal plate that holds the CPU in place. So I presume the piece of metal is from Lotes and the rest is from someone else ?
But on the Top of the SOCET is engraved 155X LM , Which fits the description of LOTES from the Table ....
Also http://images.bit-tech.net/content_i...67a-gd65-1.jpg
I'm Fairly sure I See 115X LM.

---

Thats not the socket.. its the ILM, the socket is under the ILM.. you know the thing with the pins... :p:

And i bet foxconn offers the cheapest socket of the 4 listed manufacturers...

Maybe a bad production week / contermination in the metal the pins are made of?

for me even testing "under water" with single stage - just a little corrosion, nothing more :D

http://img263.imageshack.us/img263/5899/44476096.jpg

Quote:

---

Originally Posted by Hornet331

Thats not the socket.. its the ILM, the socket is under the ILM.. you know the thing with the pins... :p:

And i bet foxconn offers the cheapest socket of the 4 listed manufacturers...

---

Then it makes sense , thanks for clearing things up.

Quote:

---

Originally Posted by alucasa

Going banana over every single incident is what people do on these forums nowadays, heh

---

Exactly. I've had the opportunity to test multiple LGA1155 Gigabyte motherboards, for myself and for systems I'm burning in as we speak (for customers). Checking for any issues with the pins has come up with nothing.

All the motherboards were packed well and without any issues some people and sites are reporting. Even for the review samples.

While this doesn't mean much compared to the amount of motherboards they are shipping, I would say it's even less of an issue.

Quote:

---

Originally Posted by BlueAqua

Let's not blow this out of proportion yet, it's only been seen on 2 motherboards by the same guy, not a widespread issue. Before everyone attacks Gigabyte with their pitchforks in hand why don't we wait to see if this is a real problem or just an isolated incident.

---

:up:

On the board in question, it has a Lotes ILM and a Foxconn socket :)

Oh, and TPU mis-quoted me....it was only 1 board with an issue, not two.....it's always wise to read the source. :up:
http://www.techreaction.net/2011/01/...-sandy-bridge/

what about this case? (Giga)

http://image.ohozaa.com/i/642/15012011147.jpg

http://www.overclockzone.com/forums/showthread.php/1158304

Quote:

---

Originally Posted by M.Beier

Why did you have to do that? I'm sitting here with half a bag of popcorn, and you just nailed the entertainment :mad:

---

i knew that too, but its not as if foxconn is the only maker of sockets, that would be worth screaming about if true. :p:

hm just for the record i have tested most h67 and p67 gigabyte boards and have not had any issues,

i think techreaction have jumped into an article without really considering what might be going on here,

the pins that are affected here at vtt/ground areas whereas on the p55 board it was vcore related pins iirc...

lets see if this continues to happen, my bet is this is an isolated incident from a review sample that might have been touched by multiple people...

Quote:

---

Originally Posted by youngpro

i think techreaction have jumped into an article without really considering what might be going on here

---

What article James? I just made a few observations. :rolleyes:

I'm really relieved that after bringing this up there hasn't been a single other incident reported (to my knowledge), which is slowly convincing me this is an isolated case....and that's what I'm hoping for! :up:

But I think it's to early to write it off completely. Like I said in my blog, "Time will tell" :yepp:

Quote:

---

Originally Posted by miahallen

What article James? I just made a few observations. :rolleyes:

I'm really relieved that after bringing this up there hasn't been a single other incident reported (to my knowledge), which is slowly convincing me this is an isolated case....and that's what I'm hoping for! :up:

But I think it's to early to write it off completely. Like I said in my blog, "Time will tell" :yepp:

---

http://www.xtremesystems.org/forums/...2&postcount=75

I cant read it due to language barrier but..

Quote:

---

Originally Posted by youngpro

hm just for the record i have tested most h67 and p67 gigabyte boards and have not had any issues,

i think techreaction have jumped into an article without really considering what might be going on here,

the pins that are affected here at vtt/ground areas whereas on the p55 board it was vcore related pins iirc...

lets see if this continues to happen, my bet is this is an isolated incident from a review sample that might have been touched by multiple people...

---

Reminds me of the VTT panic back in the wolfdale days.... everyone was screwming like mad, when there where only a view cases and most of them where achived with unrealistic values anyway... :p:

Quote:

---

Gigabyte has responded to the reports that its LGA1155 motherboards have a habit of emulating Bonfire Night, by claiming that the company sent the wrong boards out to reviewers.

Gigabyte's US team sent TechReaction two boards: one UD7 (brand new) and one UD4 (with an open box). However, Gigabyte explained that the UD4 was already a faulty component before it was sent out to TechReactor, and shouldn't have left the company's building.

'We suspect that we messed up and sent him a board that we had damaged during our own testing and have sent him another new one,' claims the company.

'Our testing involves overclocking and other various tests to which no motherboard should normally be exposed – we’re basically testing durability, and durability is determined when the board dies.

We’re still waiting to check the board and trace where it has been, to try and figure out how a damaged board was sent as a media sample and prevent this from happening again.'

Gigabyte didn't detail the exact limits of its boards' durability, but it did claim that there were no reports of retail boards exhibiting the same problems.

---

Source: Bit-Tech

Calm down people. The sky isn't falling and your sockets aren't burning up. I know I've been blasting ~1.7vcore through my P67 board/CPU for ~8 weeks and I haven't seen any signs of corrosion or degradation. Considering how many other people have been doing the same thing you'd think we'd have more than one story about this. Doesn't help that the socket wasn't even damaged by the writer of the article but it came from the manufacturer damaged. Case frigging closed.

Quote:

---

Originally Posted by Chri$ch

Source: Bit-Tech

---

/thread

Quote:

---

Originally Posted by Aberration

/thread

---

haha not the only person who think's like me

I wonder why ppl doing pin power transfer, it will be much better to get in buttom of cpu's + plate and - plate to power up cpu.
Other conection as normal, IO pins to control cpu/mainboard