• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

What's Kosher when shipping a CPU?

GeezerMan

GeezerMan

Platinum Member
This is kinda a rant and WTF about shipping CPUs, and maybe it's just me. When I ship a CPU, I use the CPU clamshell, either the original one or something like Newegg uses, then bubble wrap, then a box.

I must be the strange one, because I ordered two Intel CPUs to upgrade two old Dell towers at work from two different ebay sellers with very good feedback.

Seller one, ships the bare CPU in standard bubble wrap, the small bubble size, then in a small padded envelope. No antistatic bag, no antistatic bubble wrap. Seller two, even worse, Bare CPU wrapped in paper, then stuck in a small padded envelope.

Am I the crazy, overly-cautious one here? i realize the odds of electrostatic shock are slim, but still, paper?
 
The odds of electrostatic shock aren't slim. I work at Intel and I've personally nuked a fair number of CPU's over the years just from picking them up. I remember once, we had a CPU that we were using to calibrate a tester, and we pulled it out of one tester and my fellow engineer just passed it to me and it tested bad for two pins. So I messed with it a bit and couldn't figure it out, and I handed it back and the same two pins tested bad on his tester... we decided it was static since the odds of a CPU failing opens-shorts on two pins just randomly is pretty much non-existent.

I remember we were having high fall-out during packaging and there was a short bit where the CPU's would go over a bit of rubber belt before being packaged and we were losing a small but significant percent of parts from just being on a rubber map briefly. One last example was once I was working on a very high priority issue and we had a prototype sample that we'd rewired using a focused-ion beam to resolve the issue. I tested it and it was all clean on the high volume tester, and was carrying it off to another part of the building covered in aluminum foil. I handed it off to a co-worker and he asked me for the ID number on it, so I took the aluminum foil off, read the number out to him, and handed it to him and there was a loud snap of static electricity and that was it for that sample. I remember going into a meeting an hour later and having to admit that I had screwed up and not followed ESD guidelines for the sample in front of a large crowd... not a good time.

If anything, I tend to think that people underestimate how much of a problem it is.

So, yeah, Geezerman, I think people shipping CPU's around in paper is silly. I think generally people outside the industry don't understand that a trivial thing like static can kill a CPU worse than jumping up and down on it. I think nowdays you could drive a car over the top of something like a socket 1155 CPU and it would likely be ok, but if you walked across a carpet in winter and picked it up by the bare pin contacts, you'd likely destroy it.

Electrostatic protection circuitry on CPU's - and other integrated electronics - has gotten significantly better over the years, but it's still a real issue. Whenever I read a post of someone who talks about a CPU - particularly an Intel CPU because I'm very familiar with our testing procedures - being dead-on-arrival, my first thought is that they weren't careful about static. But people generally seem to blame the manufacturer or the distributor (ie. Newegg).

Patrick Mahoney
Senior Design Engineer
Intel Corp.
* not a spokesperson for Intel Corp. *
 
Last edited:
i use a static bag first.. then i use a pink static bubbler, and then i mail it out in a bubble envelope.

The only thing is you dont want the die to crack, and if its not an AMD cpu, u have no pins to worry about.

So all your worried about is impact when the package gets dropped.

The smaller the package, the less weight it has overall, which means less impact at drop.

I have not lost 1 intel cpu from mailing yet that way.... And thats how my sponsors would mail it out to me.
 
pm said:
The odds of electrostatic shock aren't slim. I work at Intel and I've personally nuked a fair number of CPU's over the years just from picking them up. I remember once, we had a CPU that we were using to calibrate a tester, and we pulled it out of one tester and my fellow engineer just passed it to me and it tested bad for two pins. So I messed with it a bit and couldn't figure it out, and I handed it back and the same two pins tested bad on his tester... we decided it was static since the odds of a CPU failing opens-shorts on two pins just randomly is pretty much non-existent.

I remember we were having high fall-out during packaging and there was a short bit where the CPU's would go over a bit of rubber belt before being packaged and we were losing a small but significant percent of parts from just being on a rubber map briefly. One last example was once I was working on a very high priority issue and we had a prototype sample that we'd rewired using a focused-ion beam to resolve the issue. I tested it and it was all clean on the high volume tester, and was carrying it off to another part of the building covered in aluminum foil. I handed it off to a co-worker and he asked me for the ID number on it, so I took the aluminum foil off, read the number out to him, and handed it to him and there was a loud snap of static electricity and that was it for that sample. I remember going into a meeting an hour later and having to admit that I had screwed up and not followed ESD guidelines for the sample in front of a large crowd... not a good time.

If anything, I tend to think that people underestimate how much of a problem it is.

So, yeah, Geezerman, I think people shipping CPU's around in paper is silly. I think generally people outside the industry don't understand that a trivial thing like static can kill a CPU worse than jumping up and down on it. I think nowdays you could drive a car over the top of a CPU and it would likely be ok, but if you walked across a carpet in winter and picked it up by the bare pins, you'd likely destroy it.

Electrostatic protection circuitry on CPU's - and other integrated electronics - has gotten significantly better over the years, but it's still there. Whenever I read a post of someone who talks about a CPU - particularly an Intel CPU because I'm very familiar with our testing procedures - being dead-on-arrival, my first thought is that they weren't careful about static. But people generally seem to blame the manufacturer or the distributor (ie. Newegg).
Click to expand...

So, is there a consumer based program I can run on the CPU to test for 100% operation? other than a Prime95 type of program? or is it a case where if it gets shocked, it usually dies completely?
 
Last edited:
GeezerMan said:
So, is there a consumer based program I can run on the CPU to test for 100% operation? other than a Prime95 type of program?
Click to expand...

I think something like Prime95 or LinX is about the best you can do at home. Prime95 is a good one for checking the CPU largely in isolation, but Linx is pretty good at testing the CPU with the memory together. The CPU test in the recent 3DMark benchmarks is pretty good at testing the CPU, video card and memory all together. My personal favorite for the new 2nd generation core CPU's (ie. Sandybridge) is LinX.

I find that ESD (electrostatic discharge) problems are just about the only thing that I ever saw in the lab that could cause a CPU to be flaky but that was extremely rare and a long time ago. Most of the time it's just dead and doesn't do much of anything. It won't get past link training power-on. But very rarely in the past - back when we had shared bus I/O's, static could cause an oxide breakdown issue on just one side of one of the output drivers on an IO pin in such a way that it still sort of worked. But that was rare and nowadays with the high-speed IO that we have, I can't think it would train the link successfully and so nowadays I think it would be totally dead.
 
Last edited:
I think I'll do a "You should have warned them that coffee can be hot" type of thing here and sue Intel and say they should stamp warning labels on their CPUs to assist ebay sellers..:whiste:

pm said:
I think something like Prime95 or LinX is about the best you can do at home. Prime95 is a good one for checking the CPU largely in isolation, but Linx is pretty good at testing the CPU with the memory together. The CPU test in the recent 3DMark benchmarks is pretty good at testing the CPU, video card and memory all together. My personal favorite for the new 2nd generation core CPU's (ie. Sandybridge) is LinX.

I find that ESD (electrostatic discharge) problems are just about the only thing that tend to cause a CPU to be flaky. But most of the time it's just dead and doesn't do much of anything. It won't get past link training power-on. But very rarely in the past - back when we had shared bus I/O's, static could cause an oxide breakdown issue on just one side of one of the output drivers on an IO pin in such a way that it still sort of worked. But that was rare and nowadays with the high-speed IO that we have, I can't think it would train the link successfully and so nowadays I think it would be totally dead.
Click to expand...
 
pm said:
Electrostatic protection circuitry on CPU's - and other integrated electronics - has gotten significantly better over the years, but it's still a real issue. Whenever I read a post of someone who talks about a CPU - particularly an Intel CPU because I'm very familiar with our testing procedures - being dead-on-arrival, my first thought is that they weren't careful about static. But people generally seem to blame the manufacturer or the distributor (ie. Newegg).
Click to expand...

So grounding myself to the metal fridge and working in my boxers and a pair of rubber soled shoes isn't going to cut it? 😕😛
 
at least it wasn't like this:

pict0029hd0.jpg
 
I don't know why anyone would buy any CPU off flea-bay. If it isn't a new PIB, with all valid serial numbers, then ... well... it is flea-bay after all. Many of my friends have been burned on CPUs trying to save a few bucks, and they regretted it.
One idiot put a Athlon chip in a regular envelope, and all the pins were bent by the mail processing machine. Another one, it had scorch marks on it, and the 3rd one the heatspreader came off.
The only exception is if it is a really old CPU you are looking for, and they don't make them anymore.
 
Elixer said:
I don't know why anyone would buy any CPU off flea-bay. If it isn't a new PIB, with all valid serial numbers, then ... well... it is flea-bay after all. Many of my friends have been burned on CPUs trying to save a few bucks, and they regretted it.
One idiot put a Athlon chip in a regular envelope, and all the pins were bent by the mail processing machine. Another one, it had scorch marks on it, and the 3rd one the heatspreader came off.
The only exception is if it is a really old CPU you are looking for, and they don't make them anymore.
Click to expand...

I have had my hand in many a Dell Precision 380 and 390, and I can tell you Dell uses massive CPU heatsinks in them. You can get away with slow CPU fans when the heatsink is huge, and the CPU is running stock. This results in usually decent used CPUs. Why am I telling you this? because that's what my need was for, a 5 year old Dell socket 775 desktop that only takes Conroe cores. The Ebay sellers who dissassemble the Dell machines usually have decent parts, since you know the CPUs have not been overvolted, unless the board went nuts.

Not all of ebay is bad, as my personal feedback of over 1600, all positive, will attest. You might get a better CPU there than here, where some hot rodder has overvolted everything. I'll probably be selling my personal Q6600s soon, both SLACRs, always aftermarket cooled, and both never overclocked, and I'll probably put them on Ebay.
 
I always package the CPU in the original retail box before shipping. I have plenty of space to keep my boxes though.
 
Static is a big problem for sure. Some people are just more prone to carrying higher static charges and can kill parts right off. I always use a static strap or touch the psu to ground myself. A long time back I killed a sound oor network card when I did not ground myself.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top