User avatar
Ronaldlees
Posts: 294
Joined: Sat Apr 16, 2016 4:28 pm
Location: North Carolina, US
Contact: Website

Some uSD failures could be caused by this ...

Sun Nov 05, 2017 11:19 pm

I mentioned recently, on one of the threads in this section, that I'd had only one uSD disk go bad out of around twenty or so that I'd been using over the past couple years. I should have knocked on wood, because I had one of them die today. It was after I tried to insert the uSD into the slot, near a low clearance tight corner in the enclosure, using a small needle-nosed pliers. This particular enclosure was not laid out very well I admit.

When I later examined the disk, I discovered a hairline crack across the back of it. I wondered if the hairline could be an issue, and took another needle-nosed pliers to stress the crack only a little. It cracked easily, and I could see the cross-section of the wafer thin die (pun intended). The die itself is most of the thickness of the disk, or at least half of it. There's not much plastic, so I'd have to guess this same sort of thing could/is happening to others on these forums who've had "unexplained" uSD failures.

I didn't think I had used that much force to pinch the uSD with the pliers when I inserted it. I won't be using anything other than my fingers from now on ...
I am the Umbrella man

broe23
Posts: 902
Joined: Thu Jan 28, 2016 9:35 pm
Location: Central IL
Contact: Website

Re: Some uSD failures could be caused by this ...

Mon Nov 06, 2017 7:04 am

Always use tweezers if you have to remove the card or use a dremel to fix the slot.
Ren: Now listen, Cadet. I've got a job for you. See this button? Ren: Don't touch it! It's the History Eraser button, you fool! Stimpy: So what'll happen? Ren: That's just it. We don't know. Maybe something bad, maybe something good.

Heater
Posts: 17123
Joined: Tue Jul 17, 2012 3:02 pm

Re: Some uSD failures could be caused by this ...

Mon Nov 06, 2017 7:19 am

Interesting theory.

Have micro SD cards become a lot more fragile recently? Last time I tried to snap one in half, because it had annoyed me by becoming partially write protected whilst trying to us it in a Pi, it was not easy. That was some years ago mind.

However, I'm not sold on this theory. I would expect such physical damage to a card to cause total failure. What I have seen though is cards become fully or partially write protected, as in the above example. Sometimes they could even be formatted but failed to record data correctly afterwards. They were still operating to some extent.
Memory in C++ is a leaky abstraction .

User avatar
Ronaldlees
Posts: 294
Joined: Sat Apr 16, 2016 4:28 pm
Location: North Carolina, US
Contact: Website

Re: Some uSD failures could be caused by this ...

Mon Nov 06, 2017 2:24 pm

Broe23: Agreed that tweezers could be OK.

Heater: I suppose that what I had could have been that rare flukey disk. On a sample of one, I can't say much. I'd applied only normal pressure with the pliers. The hairline crack was so thin that at first I thought it was just a scratch, but a second pliers put onto the other half made it break easily right along the scratch line. It was a catastrophic failure - the device was not any longer detected by my OS after the insertion attempt, and it was not detected when I inserted it into a regular PC in order to test it.

Looking at the cross section, I could see that the die substrate was much thinner than when I worked in the industry (I spent eight years working in an IC wafer fab outfit). Maybe it was only half the old-time wafer thickness. The plastic on one side of the uSD is so thin that it just barely hides the topography of the metalization layer on the chip, and you can see a hint of the IC outline under it). I wonder if a little less pressure than what I applied could cause faulting in the memory, and lead to what you've observed (the read-only status) - but not leave a hairline crack telltale.

So, I'd suggest fingers or (as broe23 wrote) - a light tweezers for these ...
I am the Umbrella man

Return to “General discussion”