Sign in with
Sign up | Sign in
Your question
Solved

CPU Socket pins broken _is their a solution.

Last response: in CPUs
Share
February 10, 2011 12:58:13 AM

Hi..I recently bought an Asus P7P55LX Socket 1156 on auction with "bent pins"... These are no problem as they can be straightened but I have found 5 pins that apparantly are broken .....I initially thought they may have been compressed into the socket but using tweezers I grabbed what was protruding and gently used upward pressure but I have concluded that they must be broken ..... I would be extremely grateful for any advice on a method I can use to make the M/B usable...Now as I understand not all the pins on a 1156 socket are actually used and I wonder how I can verify this...
As I see it then I need to know alternatives that I can perhaps use .....Is soldering an option ? or some other method .... Thanks HK
a b à CPUs
February 10, 2011 1:16:15 AM

There are several pins in the spec that are grouped together. Several dozen are all grounds, several dozen are Vcore, etc. I can't seem to find a readily available pinout for 1156 though. Soldering is out of the question, since all the pins are mounted in a plastic carrier, and the entire assembly is soldered to the board. You can't just replace individual pins without replacing the entire socket base. If anyone has an 1156 pinout, i'm curious to see it myself. I can't seem to find one *anywhere*


m
0
l
Related resources
a b à CPUs
February 10, 2011 1:39:20 AM

I had just stumbled upon that PDF myself. There are some pins marked as reserved that may be unused.
m
0
l
February 10, 2011 1:56:57 AM

Hi..This is absolutely outstanding ..Many thanks for your prompt replies ... My knowledge is limited on the socket structure but I understand exactly what you are saying... I will download the data sheet and check it out ..I might be in luck.. There is one question that I would like an opinion on ...The pins are flattened ..quite extraordinary but I made a small tool and have straightened out in a short time about 15%.....I then came across the 5 pins that I see as a real problem..Is it possible for pins to be compressed into the plastic socket carrier ....?
I actually took the M/B to a local shop and they were staggered to see how the pins had been damaged and could not understand how it could have been done...the tecnician did say that it was unlikely that pins were broken ?? Of course he had never looked at it with magnification as I have...Anyway at this point I would like to establish if compressed pins are a possibility...Thanks HK
m
0
l
a b à CPUs
February 10, 2011 2:09:59 AM

It may be possible that the socket pins were pushed down into the packaging. From photos and first hand observation it appears as if the pins are connected to a substrate below the plastic bottom of the socket, and the pins could have been pushed down into the gap between the bottom of the socket cavity and the substrate to which the pins are attached. The only way to know for sure is to try to decompress those pins and get them standing back up again.
m
0
l
a b à CPUs
a b Ĉ ASUS
February 10, 2011 2:12:03 AM

I really hope you can get the board to work, but that situation sounds like something for an RMA to the manufacturer. You bought it from an auction (bad idea) so you can't do that. Most likely what you have there is an expensive paperweight.

I wish you good luck in trying to get it fixed...
m
0
l
a b à CPUs
February 10, 2011 2:29:37 AM

LGA sockets are generally bad news when they get damaged.... At least with PGA chips you could straighten pins or replace broken pins. The LGA sockets are just too easy to screw up. One accidental brush against the pins and you destroy a socket.
m
0
l
February 10, 2011 5:15:25 AM

Hi..Regards the Auction..the MB was listed as faulty with bent pins so I took a punt on fixing it so I cannot blame anybody but myself ..I can always auction it off again....Anyway.....Heres where I am at....I downloaded the PDF and as long as I am looking at the socket from the right direction ? I have identified these pins as possibly broken or compressed..
AG38= PSI# Asynch Cmos 0
AF37 = TDI_M Tap I
AE35 = VTT_ sense analog
AB = 34 and 35 VSS Ground
F32 = VSS ground.

So I know what ground is and these may not cause a problem but I presume the others are crucial.. Thanks HK
m
0
l
a c 83 à CPUs
a b Ĉ ASUS
February 10, 2011 5:25:46 AM

Agreed. They look crucial to me. I get the feeling if all the board needed was a few pins straightened the guy wouldn't have sold it.
m
0
l
a b à CPUs
February 10, 2011 8:20:38 AM

Sounds like it's time to punt again.
m
0
l
February 10, 2011 4:14:56 PM

You should try working it out with the seller and get him to agree to take it back. Bent pins are way different than broken or missing.
m
0
l
February 10, 2011 5:27:58 PM

OK...Thanks for all your comments ...HK
m
0
l
February 10, 2011 9:10:41 PM

Just on an off chance ..I have isolated what 3 pins are for ...Of course meaningless to me but possibly somebody might know how crucial they are to the motherboards operation.. Thanks HK


(1) TDI_M (Test Data In) transfers serial test data into the
processor. TDI_M provides the serial input needed for
JTAG specification support.

(2) PSI# Asynch Cmos
Processor Power Status Indicator: This signal is asserted
when maximum possible processor core current
consumption is less than 15 A. Assertion of this signal is an
indication that the VR controller does not currently need to
be able to provide ICC above 15 A, and the VR controller
can use this information to move to more efficient
operating point. This signal will de-assert at least 3.3 ms
before the current consumption will exceed 15 A. The
minimum PSI# assertion and de-assertion time is 1 BCLK.
O Asynch
CMOS

(3) VTT_SENSE Analog
VTT_SENSE and VSS_SENSE_VTT provide an isolated,
low impedance connection to the processor VTT voltage
and ground. They can be used to sense or measure
voltage near the silicon.
m
0
l
a b à CPUs
February 11, 2011 3:06:51 AM

It would seem that: (1) is only for testing, and therefore unnecessary; (2) is for letting the voltage regulator run more efficiently at low loads, and that if it was missing would simply make it run for full load all the time; and (3) is probably necessary for the voltage regulator to function properly, as it uses it to identify what the voltage supplied to the chip is, although it may be possible to operate without it but have significant decreases in voltage when under load.
m
0
l
February 11, 2011 7:02:15 AM

PreferLinux said:
It would seem that: (1) is only for testing, and therefore unnecessary; (2) is for letting the voltage regulator run more efficiently at low loads, and that if it was missing would simply make it run for full load all the time; and (3) is probably necessary for the voltage regulator to function properly, as it uses it to identify what the voltage supplied to the chip is, although it may be possible to operate without it but have significant decreases in voltage when under load.



Hi..Thats Great information .... I have looked at my options and completely ruled out soldering but I did read somewhere when I first Googled re the problem that it was possible to use a conductive adhesive epoxy to basicly glue pins in....There is such an epoxy available here but at a cost that would make such an exercise (cost wise ) stupid...however I am asking round in case one of the local tech shops has some in a partly used state...I might be able to buy a "blob" off them.....There is another electronic adhesive called JB Weld but I do not know if its conductive..it retails at a far lower price... I presume one could use a suitable wire..Put a small bit of epoxy on the end and push it in beside the broken pin which I am sure would make contact between the two ...wait for the epoxy to set and just put a bit more epoxy round them...Then snip off the wire at the right height... with my very limited knowledge this seems to be the only option I have apart from re selling it...
May I just say this is a great forum .....Cheers HK
m
0
l

Best solution

a b à CPUs
February 11, 2011 12:41:42 PM

JB Weld is not electrically conductive (once it completely cures). And no, I would not even TRY to use epoxies for the pins in an LGA socket. First of all, the epoxy dries rock hard, and these pins need to flex when the socket is clamped down. Second, you're talking about pins that only a few thousandths of an inch thick, spaced a few hundredths of an inch apart. How can you use peanut butter consistency epoxy on parts that small effectively?

My opinion, ditch the board, and consider it a lesson learned.

Share
February 19, 2011 3:12:24 AM

Best answer selected by hairykiore.
m
0
l
!