[O/C]Socket 1156 (i5/i7) Burning Issues

Socket 1156 (i5/i7) Burning Issues
by Earthdog

There has been a lot of discussion lately across many sites regarding Socket 1156 burning up and charring to the socket and pin area. The rare problems we have seen are on the Foxconn made socket and tend to only occur when using extreme voltages, clock speeds, and extreme cooling. Recently there has been one publicly posted occurrence of the Lotes socket showing the same failure, under the same extreme circumstances. While this is an obvious issue for extreme benchers, a valid question was raised regarding moderate overclocks on air or water, and whether this problem could happen to them.

Click to expand...

Click here to continue reading.


Discuss this article below. If you are interested in contributing to the front page (www.overclockers.com), please feel free to contact me.


Thanks,
sno.lcn

very useful information
I still think that its a condensation/cooling/volt issue that requires many factors to be present b4 you will duplicate the problem.
I'm not trying to debunk this problem but I think its over hyped for a few isolated cases, clear and unbiased reports ( such as E.D's)will tell in time how serious a problem it is.
I run my rig at 4.0GHz for 24/7 but I bench it at 4.3 + with some high volts and last socket check everything was 100%, I hope to have my water chiller running this weekend and then I will push harder

Grnfinger, in discussion I've seen it's only happened when running at/over 4.5. I'm not the most familiar person with the issue however, but I do believe it's a legitimate current/voltage issue with the socket - not cooling or condensation related.

Like you said tho, more clear and unbiased reports will help to clarify down the road, and I'm hoping we'll see another article from Earthdog once the overclocking community figures this out.

Well done piece Earthdog.

I concur about this being a current/voltage/socket issue from what I have seen as well. These guys that are cooking their equipment on this socket are pros. Not to say they wont make mistakes or that particular mistake allowing condensation on the board, but one would think when that happens, they would mention it and not black ball/worry about an entire socket setup.

Grn - Im looking forward to your extreme testing as well. Please keep us all posted!

There will undoubtedly be follow up article on this issue as this pans out.

Looking forward to it!

How much is "too much" voltage?

I don't think it's been clarified yet, but people are trying to figure things out. Some think it's a design problem with the socket, and too much current going thru certain pins at extreme overclocks. Time will tell afaik.

I can't remember where but there was a close up picture of this issue, and one person showed the CPU after being installed in a Foxconn socket then taking it out to show pressure marks. A few contacts on the CPU showed no pressure mark.

Ah here: http://www.anandtech.com/mb/showdoc.aspx?i=3661

I've bench at 4.52Ghz as in my sig "non-Turbo".....and a few times 4.62GHz both on air with UD3R.....and I think it has a Foxconn socket (not sure).
I've no intention of disassemble the entire unit just to check the issue.....but so far, I have no problem of what so ever. Matter of facts, it runs like a dream.
Like they said, don't fix until it breaks.

dgb303 said:

How much is "too much" voltage?

Click to expand...

There are several factors that determine the ceiling for what voltage you can feed your system....and I say systems as it is not just the CPU that can suffer from too much voltage.

For instance if on a X58 or P55 you put too much voltage to your IOH you can start having video issues as IOH feeds the DMI channel between the PCIe and your CPU.
If you feed too much to your ICH, essentially the southbridge chip, you can have drive or USB issues as ICH voltage feeds your chip that regulates the data flow between your peripherals (USB/SATA etc) and your CPU.

Then off course cooling plays a major roll, now I'm not taliking about keeping it below temp limits, when you put your CPU under LN2 the electronics can tolerate much more voltage as when you have it on air, yes, because the temps are low but also because the microelectronic pathways and billions of transistors that make up the various chips including your CPU, behave very different under extreme cold.

then last but not least your micro-architecture of the CPU or chip-set makes a big difference. The 65nm CPU can take a load of voltage , more than the 45nm CPU's and then the 32nm CPU's are much more sensitive than it's predecessors. I have never killed a CPu until I started OCing a 32nm CPU and believe me you actually can kill it with voltage, not with temps contrary to popular belief.

hope this hels

There is much more to this but is is a start.