Monday, January 17th 2011

Socket Pin Burnout Returns to Haunt LGA1155?

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.

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.

Source: TechReaction
Add your own comment

65 Comments on Socket Pin Burnout Returns to Haunt LGA1155?

#1
bear jesus
newtekie1 said:
So my guess would be that one of the previous reviewers were a little rough with removing and inserting the processor, and broke a few pins. This was still enough to make contact, but caused the arcing situation. Of course there is the other pin that isn't broken but is still burnt, but that can be caused by the extra stress of more power going through that one pin because the others are failing.
I had just assumed that maybe the pin was stuck to the CPU after the burn out and was broke when removing it but surly that would be something you would not miss as i thought when the retention bracket was open there was nothing holding the CPU in the socket thus any resistance would be very noticeable.
Posted on Reply
#2
Hayder_Master
why they lame INTEL, i know gigabyte sockets made by FOXCONN, if have gigabyte p55 series u can see it written on the side, and i read something exactly like this when was searching on bad things on Gigabyte p55 mobos.
LOTES still best socket made manufacture.
also they say new gigabyte A series sockets made by LOTES
Posted on Reply
#3
cadaveca
My name is Dave
newtekie1 said:
Some things should be cleared up.

1.) We do not know that he received the board this way from Gigabyte. He simply states that "upon arrival" there were burnt pins. However, he could have recieved it from another reviewer that tested it first, and the reviewer screwed up the pins. I know that when you review parts, they don't always come directly from the manufacturer. The manufacturer sends out the part to one person for review, then tells the reviewer where to send it when they are done.
lots of reviewers are like "if I don't get to keep the parts, I'm not very likely to do the review. Keeping the parts sent for review is how reviewers, in the most part, get paid. Guys selling lots of "new" parts, on a regular basis...reviewer.
Posted on Reply
#4
newtekie1
Semi-Retired Folder
cadaveca said:
lots of reviewers are like "if I don't get to keep the parts, I'm not very likely to do the review. Keeping the parts sent for review is how reviewers, in the most part, get paid. Guys selling lots of "new" parts, on a regular basis...reviewer.
Not how it works. And reviewers get paid by ad revenue generated by their review, not by keeping the parts to resell. Ask W1z if he gets to keep the cards he reviews and I bet he responds something like:

W1zzard said:
whether the sample goes back or not depends on the company.

i have no problem sending stuff back as long as i can keep one card for rebenching (which i'm doing right now. 5 days of non-stop benching so far and not even done with nvidia)

no problem sharing the coke, hookers will be decided on a case by case basis
And in a lot of cases they won't tell you to ship the part back, instead they will tell you to ship it to the next reviewer. Saves time and money on shipping the part back to the company only to have them ship it to the next reviewer themselves. Of course the down side is situations like this where the part was damaged by a previous reviewer.
Posted on Reply
#5
Link108
I would not assume that GB or intel are at fault when there isn't enough information in this blog. I've build two computers with GB mobos and have had nps. I'm also waiting on one for my new system. GB is top of the line quality just like Asus, MSI. There is not enough information here to make a logical choice as to what really happend.
Posted on Reply
#6
overclocking101
ALL new 1155 mobo reviews I've seen ALL have the LOTES hold down at least and most whole LOTES socket, Personally I think this is a freak accident as there is only one report before big time benchers all over were getting it and that is not the case now. Besides the foxconn issues were sorted out VERY soon after initial reports. I am using a 2nd gen foxconn socket on my P55 and NEVER any sighns of any socket burning, this is with 6 diff cpu's from 2 to 8 cores, ALL benched hard.
Posted on Reply
#7
erocker
newtekie1 said:
Not how it works. And reviewers get paid by ad revenue generated by their review, not by keeping the parts to resell. Ask W1z if he gets to keep the cards he reviews and I bet he responds something like:
Not how it works. Speaking in terms of the actual reviewer, many reviewers get paid by being able to keep the product. Many other reviewers get paid per review, keep the product, etc.
Posted on Reply
#8
Completely Bonkers
I'm really surprised that a low-voltage contact can arc. Yes, I would expect a higher-than-spec current through a small or bad contact could cause a "burnout" just like the way a fuse works, but I am really surprised that circa 1v can create an arc.

For anyone with a physics background, I have just learned about Paschen curves and field emission. Here is an easy read for those interested: http://www.lorentzcenter.nl/lc/web/2010/424/PROBLEMS/NXP/OverviewWallashspie.pdf
Posted on Reply
#9
Steevo
This isn't in a vacum, however that looks more like the pins were carrying more than they could handle. Arcing should occur on the pads as they are physically closer than the pins.
Posted on Reply
#10
HillBeast
Glad I'm on 1366. That's all I have to say. Oh and 1156/1155 fanbois who flame me saying 'OH IT WAS JUST A FEW BOARDS THAT DID IT! 1156/1155 IS STILL GOOD!!1one'. STFU. Your socket is for mainstream. Stop trying to use it for high end. Get 1366 and stop being a cheapskate.
Posted on Reply
#11
newtekie1
Semi-Retired Folder
HillBeast said:
Glad I'm on 1366. That's all I have to say. Oh and 1156/1155 fanbois who flame me saying 'OH IT WAS JUST A FEW BOARDS THAT DID IT! 1156/1155 IS STILL GOOD!!1one'. STFU. Your socket is for mainstream. Stop trying to use it for high end. Get 1366 and stop being a cheapskate.
You can go on about your 1366 being so much more extreme than my 1156, but when my 1156 is outperforming your 1366 for half the price it doesn't really hold up.;)
Posted on Reply
#12
rickss69
My bud on the bot team has been banging his Gigabyte (GA-P67A-UD5) for a week now at 5.3GHz on water and reports no problems.
Posted on Reply
#13
miahallen
newtekie1 said:
Some things should be cleared up.

1.) We do not know that he received the board this way from Gigabyte. He simply states that "upon arrival" there were burnt pins. However, he could have recieved it from another reviewer that tested it first, and the reviewer screwed up the pins. I know that when you review parts, they don't always come directly from the manufacturer. The manufacturer sends out the part to one person for review, then tells the reviewer where to send it when they are done.

2.) The reviewer posting about the problem has no idea what voltage was used, because he didn't burn the pins.

Now, as for my own observations from the pictures, it seems at least one of the pins is actually broken off. So my guess would be that one of the previous reviewers were a little rough with removing and inserting the processor, and broke a few pins. This was still enough to make contact, but caused the arcing situation. Of course there is the other pin that isn't broken but is still burnt, but that can be caused by the extra stress of more power going through that one pin because the others are failing.
The board was sent to me directly from Gigabyte HQ in TW...but it appeared to have been used previously.
Posted on Reply
#14
buggalugs
miahallen said:
The board was sent to me directly from Gigabyte HQ in TW...but it appeared to have been used previously.
That makes sense, if they are loan/test boards they most likely have been tested by several people. Why give out brand new boards for you guys to wreck. Hahaha

No doubt the first thing the testers do is push the board to its limits and someone went a little too far. When its not your board and you didnt pay for it who cares right?

This publicity could be a little unfair to Gigabyte, if a tester pumped 1.8v into it, its hardly Gigabyte's fault its wrecked.
Posted on Reply
#15
miahallen
The boards I get from Gigabyte are mine for keeps....
I don't think the boards have been available long enough to have sent them to other reviewers before me....but I don't know that for fact.
Posted on Reply
#16
HillBeast
newtekie1 said:
You can go on about your 1366 being so much more extreme than my 1156, but when my 1156 is outperforming your 1366 for half the price it doesn't really hold up.;)
Pics or it didn't happen.
Posted on Reply
#17
LAN_deRf_HA
HillBeast said:
Glad I'm on 1366. That's all I have to say. Oh and 1156/1155 fanbois who flame me saying 'OH IT WAS JUST A FEW BOARDS THAT DID IT! 1156/1155 IS STILL GOOD!!1one'. STFU. Your socket is for mainstream. Stop trying to use it for high end. Get 1366 and stop being a cheapskate.
Yeah, I'm sure all the 5Ghz sandybridge chips out there are trembling in their boots from fear of your old hot and hungry 1366.
Posted on Reply
#18
HillBeast
LAN_deRf_HA said:
Yeah, I'm sure all the 5Ghz sandybridge chips out there are trembling in their boots from fear of your old hot and hungry 1366.
LGA2011. Durr.

EDIT: Also you're bashing 1366 when your system has a 1366 CPU.
Posted on Reply
#19
DrPepper
The Doctor is in the house
HillBeast said:
LGA2011. Durr.

EDIT: Also you're bashing 1366 when your system has a 1366 CPU.
I think he's aware of the fact he has a 1366.
Posted on Reply
#20
pr0n Inspector
HillBeast said:
LGA2011. Durr.

EDIT: Also you're bashing 1366 when your system has a 1366 CPU.
well duh, why can't he do that?
Posted on Reply
#21
HillBeast
DrPepper said:
I think he's aware of the fact he has a 1366.
pr0n Inspector said:
well duh, why can't he do that?
I'm not pointing out to him he has 1366, I'm pointing out it's stupid to be saying...

LAN_deRf_HA said:
old hot and hungry 1366
...when you are running it yourself implying 1366 is a piece of garbage. 1366 is still the best platform for any CPU. Period. Until 2011 comes out, 1366 will remain the best. It has the best features, best performance, best overclocking, best reliability, best CPUs and most importantly: most pins meaning less chance for arcing because of people using mainstream to do high end stuff. Leave mainstream for mainstream. Buy 1366 if you want to overclock, and if you want to get bang for buck then why didn't you buy AMD? Lynnfield is only there to fill a market hole between Clarkdale and Bloomfield/Gulftown.
Posted on Reply
#22
bear jesus
HillBeast said:

...when you are running it yourself implying 1366 is a piece of garbage. 1366 is still the best platform for any CPU. Period. Until 2011 comes out, 1366 will remain the best. It has the best features, best performance, best overclocking, best reliability, best CPUs and most importantly: most pins meaning less chance for arcing because of people using mainstream to do high end stuff. Leave mainstream for mainstream. Buy 1366 if you want to overclock, and if you want to get bang for buck then why didn't you buy AMD? Lynnfield is only there to fill a market hole between Clarkdale and Bloomfield/Gulftown.
Just because something is old and runs hot does not mean it is garbage but when when new mid range single threaded quad cores can trade blows with the old dual threaded 6 core then it's not exactly the best thing out there meaning your assertion of it being the best performance is flawed, the new cores can overclock to around or over 5ghz on air so by far the current best overclockers, reliability wise is there a reason why you think the new CPU's would be less reliable? and why do you think more pins decreases the risk of arcing? surly the more pins you have the higher the risk one of them could arc due to a bad contact? even though it was just caused by low quality sockets because of a company's bad product so hardly something to effect all boards with sockets from all company's.

People buy mainstream parts and push them because they can and they can handle it otherwise there would be a lot of unhappy members on here.
Posted on Reply
#23
HillBeast
bear jesus said:
Just because something is old and runs hot does not mean it is garbage but when when new mid range single threaded quad cores can trade blows with the old dual threaded 6 core then it's not exactly the best thing out there meaning your assertion of it being the best performance is flawed
http://www.anandtech.com/bench/Product/142?vs=192 <- Need I say more? It isn't trading blows. It's not even close. Also I never said old and hot = garbage, I said he implied it.

bear jesus said:
the new cores can overclock to around or over 5ghz on air so by far the current best overclockers
http://www.techpowerup.com/forums/showthread.php?t=118081 <- Um yeah. 5GHz...

bear jesus said:
reliability wise is there a reason why you think the new CPU's would be less reliable?
Socket Pin Burnout Returns to Haunt LGA1155? - Pa... <- You're looking at it.


bear jesus said:
and why do you think more pins decreases the risk of arcing? surly the more pins you have the higher the risk one of them could arc due to a bad contact?
I am going to assume you never learnt about electronics before? The more contacts, the less resistance something has. It's common sense. If you have two pipes feeding a house with water, it's not going to be as good as three pipes. Two pipes will need to feed more water in and it is more strain on the pipes. Do I need to pull up Wikipedia to prove this?

bear jesus said:
even though it was just caused by low quality sockets because of a company's bad product so hardly something to effect all boards with sockets from all company's.
But this is the bad press and bad press sticks. I don't recall seeing anywhere near as much bad press about 1366 arcing as 1156 or 1155 have, let alone at all. I'm probably wrong on that one though because Google can be unreliable.

bear jesus said:
People buy mainstream parts and push them because they can and they can handle it otherwise there would be a lot of unhappy members on here.
Or people could stop being cheapskates, pay more and not whinge when it craps out because it is much higher quality.
Posted on Reply
#24
Frick
Fishfaced Nincompoop
HillBeast said:

Or people could stop being cheapskates, pay more and not whinge when it craps out because it is much higher quality.
Wow, have you been in contact with humanity lately? :laugh:
Posted on Reply
#25
bear jesus
HillBeast said:

I am going to assume you never learnt about electronics before? The more contacts, the less resistance something has. It's common sense. If you have two pipes feeding a house with water, it's not going to be as good as three pipes. Two pipes will need to feed more water in and it is more strain on the pipes. Do I need to pull up Wikipedia to prove this?

Or people could stop being cheapskates, pay more and not whinge when it craps out because it is much higher quality.
I forgot to say when overclocked the 2600k trades blows with the 980x in specific tests, a suicide run on a 980x is not exactly a good way to point out that it can hit 5ghz, I'm sure if i kept my 965 cool enough to push around 1.7v through it before v droop it could hit 5ghz.

I as far as i knew the sockets burning out was all to do with bad contacts on low quality sockets made by foxconn and nothing to do with resistance thus none of your example or anything you could pull from any site would apply.

As far as the comment on people being cheapskates just for going for mid range parts and then overclocking why have you not gone for a 980x and a better board with faster and more ram? obviously as you cheeped out you bought lower quality components, oh wait you did not.
Just because you chose mid range parts does not mean you chose low quality parts.

With so many members here buying mid range parts and pushing them then surly there would be a lot more people complaining either that or people are not complaining when things die due to pushing them too far so really calling people cheapskates and whingers is uncalled for and inaccurate.

But this is starting to get off topic, so far a single board has been found to have a messed up socket and what caused it is still unknown so we shoudl be looking for more examples not arguing about random things :laugh:
Posted on Reply
Add your own comment