[SOLVED] Chances of pin resolder working?

[05WRX_Mark]

So today I took my computer to a specialist, will get a verdict tomorrow.

In the process of boxing up my two extra CPUs I dropped my Ryzen 7 from about 2 inches right back onto the cooler. The drop broke two pins and bent about 10.

I straightened the pins that weren't broken, and cleaned the board with 90% isopropyl.
I took a random wire from some junk lying around, split the ends, and figured out that three of them twisted are about the diameter of the other pins. So I fluxed the twisted wires, soldered the strands together, and soldered them to the board. In the process I accidentally fused two of the pins together, so I used a solder vacuum to get the solder off. It worked, but broke another pin. So after making another pin and getting both on, straightening everything out, and testing the pins for strength, I've got a CPU with all pins connected, but I'm not too sure how CPUs are put together in the first place. Is this going to work?

The new pins are about .05mm longer than the others, and one of them is a bit thick, too much solder on one side. If the thickness becomes a problem I can hopefully run the soldering iron up the pin and pick up the extra.

I made a monocle out of the eyepiece of a scope so I could see. Taped it to my head haha.

https://imgur.com/a/6vUzBCb

View: http://imgur.com/a/6vUzBCb

The third picture with the badly soldered and bent pin is before I took that one off and resoldered a new one, which can be seen in the first two pictures.

 

[Grobe]

Even if you somehow manage to fasten the broken pins (the solder have a tendency to attach to adjacent pins almost in an instance of touch), it is difficult/impractical to keep the cpu safe from esd, maybe if you have the original package, the little square bubble-antistatic-plastic that the CPU stand on, that you can cut one corner off so that at least most of the surface/pins are safe - together with an anti-static arm wrist.

 

[USAFRet]

Is this going to work?

Only way to know is to try.

 

[05WRX_Mark]

Even if you somehow manage to fasten the broken pins (the solder have a tendency to attach to adjacent pins almost in an instance of touch), it is difficult/impractical to keep the cpu safe from esd, maybe if you have the original package, the little square bubble-antistatic-plastic that the CPU stand on, that you can cut one corner off so that at least most of the surface/pins are safe - together with an anti-static arm wrist.

The strength test I did (no clue what a professional would do, I just came up with some BS) was to be able to lift the CPU by one pin only, and see how easily and where the pins bent relative to the original pins. I was able to pick the CPU up by all three with no give / bending at the solder joint, and the pins bent around the midpoint, instead of on the solder. And like I said in the post, I did manage to solder two pins together, but vacuumed the solder out. I'm 90% sure there is no short between the pins, unless there is such a thin layer of solder that it's not visible even under a magnifying glass.

If you're saying the soldering iron is more likely to create static and then spark on the CPU, then idk. I already soldered everything, so if it broke it broke. I really had nothing to lose at that point. Other than that, if you're saying the board is now more susceptible to ESD, this is my main CPU, so it should be living nice and safe in my computer, assuming everything still works, of course. It's currently in the plastic container it came in. I kept a bare foot on the computer case, plugged in. That should have provided ground, and beyond that today's humidity is almost 90%, so static isn't likely in the first place.

 

[cryoburner]

Rather than using random wires, it might have been better to use the original broken pins, or if that's not an option, similarly-sized pins taken off some old, nearly worthless CPU. In some cases, simply placing the broken pins into the corresponding holes in the motherboard socket and having them press against the CPU pads without soldering them has been known to work, though that's of course not guaranteed.

 

[05WRX_Mark]

Only way to know is to try.

Yep, waiting for the computer to come back tomorrow. Fingers are crossed.

 

[hotaru251]

as stated above..only way to find out is to try.

Not all the pins are required (some dont really do anything to begin with) but no real way tell if ones damaged are thsoe or not.

the only issue could be the height/thickness of the repaired pins. Can casue it to not physically fit the socket (and if height difference likely snap pin when tightening the cooler)

and yeah using repalcement pins from a dead cpu (can find some on ebay for cheap) is easier and higher chacne to work.

 

[05WRX_Mark]

Rather than using random wires, it might have been better to use the original broken pins, or if that's not an option, similarly-sized pins taken off some old, nearly worthless CPU. In some cases, simply placing the broken pins into the corresponding holes in the motherboard socket and having them press against the CPU pads without soldering them has been known to work, though that's of course not guaranteed.

I agree, unfortunately the pins were lost in carpet almost immediately. One of the pins also broke in half, so I don't know what I could have done about that one, either attaching a CPU pin, or putting one in the socket. One of my big concerns now is the connection on the bent pins. I think I saw part of the original solder joint lift on a couple of them when I was bending them back. I suppose if the CPU doesn't work I'll melt the solder on the pins (only about 10 in total) before I say it failed.

 

[05WRX_Mark]

as stated above..only way to find out is to try.

Not all the pins are required (some dont really do anything to begin with) but no real way tell if ones damaged are thsoe or not.

the only issue could be the height/thickness of the repaired pins. Can casue it to not physically fit the socket (and if height difference likely snap pin when tightening the cooler)

and yeah using repalcement pins from a dead cpu (can find some on ebay for cheap) is easier and higher chacne to work.

Thanks for your input. I was pissed at myself at the time and just wanted to try to fix it. The pins I made are certainly no CPU pins, but they're about the same strength and thickness. One problem could be the increased resistance of the copper wire and solder.

I'll definitely have to try to get the new pins even then, last thing I need is a broken pin stuck in the socket. To be honest I'm not too familiar with CPU and socket connections. Is the contact point on the side of the pin holes, or is it at the bottom? I ask because if its at the bottom I'll have to try to get the pins exactly the same height, but if it's on the side I might want to cut them a little short just to be safe.

 

[USAFRet]

Looking at the 3rd pic...I give the chances of working somewhere below 2%.

 

[getochkn]

I would have tried booting with the broken pins first. Might have gotten lucky and it just booted if it was just a ground pin or maybe find out a usb port doesn't work or something. But good luck. lol. I used to reball 360 and PS3 GPU's. That was a fun experiment in soldering work.

 

[05WRX_Mark]

Update: it worked. Fixed the issue with GPU / CPU (it ended up being the motherboard), and after redoing a couple pins that were too thick to fit into the socket, the computer booted right up. Thanks everyone.