Author Topic: Doesn't happen every day.. IC package destroyed while hot air soldering.  (Read 3730 times)

0 Members and 1 Guest are viewing this topic.

Offline JPorticiTopic starter

  • Super Contributor
  • ***
  • Posts: 3573
  • Country: it
So i am assembling three prototypes where i used a MCP16311 buck converter. I had four left from a past project, and they were kept inside the tape, inside the antistatic bag.

None of them works. It's not a circuit problem, as i literally copy-pasted the power supply from the previous project, and triple checked.

Then this happened, i was removing one of the chips using hot air and tweezers. The package was destroyed.
I have soldered hundreds, even thousands of chips this way. Something like this never happened to me.

Out of ideas, all of the chips fails the same way (vfb doesn't rise above 0.2V, switcher doesn't switch, Vdd is 5V with a little pulse on it)

Semiconductor crime scene photos i guess

 

Offline Ranayna

  • Frequent Contributor
  • **
  • Posts: 983
  • Country: de
Re: Doesn't happen every day.. IC package destroyed while hot air soldering.
« Reply #1 on: September 04, 2018, 09:33:14 am »
The only thing I can think of would be moisture absorption. That could easily crack the casing when hot-air soldering.
Were these the oldest stock you ever used?
 

Offline JPorticiTopic starter

  • Super Contributor
  • ***
  • Posts: 3573
  • Country: it
Re: Doesn't happen every day.. IC package destroyed while hot air soldering.
« Reply #2 on: September 04, 2018, 09:38:16 am »
I am suspecting moisture as well. Now i will try and mount one of these in a working board and vice versa..

Another really nasty photo
 

Offline Benta

  • Super Contributor
  • ***
  • Posts: 6420
  • Country: de
Re: Doesn't happen every day.. IC package destroyed while hot air soldering.
« Reply #3 on: September 04, 2018, 12:09:21 pm »
Yep, moisture popcorning. You might be able to save them by baking.
 

Offline RoGeorge

  • Super Contributor
  • ***
  • Posts: 7011
  • Country: ro
Re: Doesn't happen every day.. IC package destroyed while hot air soldering.
« Reply #4 on: September 04, 2018, 12:25:49 pm »
After a while, the case absorbed moisture from air. In the original bag should be a moisture indicator.

To avoid future cases cracking, you need first to slowly bake the ICs (in order to slowly remove the moisture before proceeding to soldering), then solder the baked ICs with hot air or an oven.

Offline Bud

  • Super Contributor
  • ***
  • Posts: 7276
  • Country: ca
Re: Doesn't happen every day.. IC package destroyed while hot air soldering.
« Reply #5 on: September 04, 2018, 12:30:31 pm »
Was it an Aliexpress special?  :o
Facebook-free life and Rigol-free shack.
 

Offline JPorticiTopic starter

  • Super Contributor
  • ***
  • Posts: 3573
  • Country: it
Re: Doesn't happen every day.. IC package destroyed while hot air soldering.
« Reply #6 on: September 04, 2018, 06:18:21 pm »
 :-DD

nah, genuine MCP16311 from mouser
 

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 10172
  • Country: gb
Re: Doesn't happen every day.. IC package destroyed while hot air soldering.
« Reply #7 on: September 04, 2018, 07:15:46 pm »
An excellent demonstration that epoxy packages are not hermetic (and that you can get all sorts of ionic sh*t leaching in via the pins). I've not seen it on such a small package before though - plenty of times on PLCCs.
Best Regards, Chris
 

Offline RoGeorge

  • Super Contributor
  • ***
  • Posts: 7011
  • Country: ro
Re: Doesn't happen every day.. IC package destroyed while hot air soldering.
« Reply #8 on: September 04, 2018, 07:40:26 pm »
Look at the bright side, you just found a new method to instantly decap chips without nasty chemicals!  ;D

Can you take a photo of the decapped die under microscope, please?

Offline mcinque

  • Supporter
  • ****
  • Posts: 1131
  • Country: it
  • I know that I know nothing
Re: Doesn't happen every day.. IC package destroyed while hot air soldering.
« Reply #9 on: September 04, 2018, 08:05:20 pm »
Was it an Aliexpress special?  :o
For sure!  :-DD

Yep, moisture popcorning. You might be able to save them by baking.
I heard the baking should be very long (i.e. 90°C for 24h), don't know if it's right.
have you had any experience or maybe do you have any time/temperature combination to suggest?
 

Online Ian.M

  • Super Contributor
  • ***
  • Posts: 13216
Re: Doesn't happen every day.. IC package destroyed while hot air soldering.
« Reply #10 on: September 04, 2018, 09:11:35 pm »
Google: JEDEC bakeout

Sub 100°C the bakeout times can be *very* long e.g. 10 days for a 4.5mm thick package at 90°C & maintained <5% RH.  At 125°C, the max bakeout time is a far more reasonable 48H.

A DIY oven for the purpose of baking out small quantities of loose parts could be as simple as a 40W incandescent light bulb in a high temperature holder mounted in a tin can with a ring of holes round the edge of the base for airflow, then a shallow aluminum tray, to hold the chips suspended above the bulb, with space for air circulation at the edge, and a bonded on thermocouple for temperature control, a lid with a vent to control the air-flow due to convection, and a programmable temperature controller capable of following a temperature profile, controlling the bulb power via a TRIAC.  PID proportional control would be preferable. The can exterior and the lid should be thermally insulated.

If the components cant withstand a 125°C bakeout, then you'd need something fancier, with a drier on its incoming air.   For components of uncertain provenance, I would suggest holding at 95°C long enough to significantly reduce the moisture level before the 125°C bakeout.
« Last Edit: September 04, 2018, 09:20:54 pm by Ian.M »
 

Offline JPorticiTopic starter

  • Super Contributor
  • ***
  • Posts: 3573
  • Country: it
Re: Doesn't happen every day.. IC package destroyed while hot air soldering.
« Reply #11 on: September 04, 2018, 09:22:01 pm »
Look at the bright side, you just found a new method to instantly decap chips without nasty chemicals!  ;D

Can you take a photo of the decapped die under microscope, please?

wish i had one.. I actually forgot to ask my father if he has one at his lab (he's a chemist)
the most i can zoom in is with the crappy usb microscope i used above
 

Offline HwAoRrDk

  • Super Contributor
  • ***
  • Posts: 1607
  • Country: gb
Re: Doesn't happen every day.. IC package destroyed while hot air soldering.
« Reply #12 on: September 04, 2018, 11:33:00 pm »
According to the Digikey website, the MCP16311 is classified as Moisture Sensitivity Level (MSL) 3, which is only 168 hours, or 7 days. Sounds like moisture exposure could indeed be a culprit.
 

Offline Eka

  • Regular Contributor
  • *
  • Posts: 162
  • Country: us
Re: Doesn't happen every day.. IC package destroyed while hot air soldering.
« Reply #13 on: September 05, 2018, 03:30:08 am »
Yep, moisture popcorning. You might be able to save them by baking.
I heard the baking should be very long (i.e. 90°C for 24h), don't know if it's right.
have you had any experience or maybe do you have any time/temperature combination to suggest?
It can even be many days or weeks. All depends on the package. Go read to the specifications sheet for that chip to see what is recommended for that package.
 

Offline westfw

  • Super Contributor
  • ***
  • Posts: 4349
  • Country: us
Re: Doesn't happen every day.. IC package destroyed while hot air soldering.
« Reply #14 on: September 06, 2018, 08:22:01 am »
I didn't realize that moisture would end up being a problem with "hand" soldering of SMT components :-(I knew it was an issue with full industrial reflow equipment...)
 

Offline Eka

  • Regular Contributor
  • *
  • Posts: 162
  • Country: us
Re: Doesn't happen every day.. IC package destroyed while hot air soldering.
« Reply #15 on: September 06, 2018, 06:38:58 pm »
I didn't realize that moisture would end up being a problem with "hand" soldering of SMT components :-(I knew it was an issue with full industrial reflow equipment...)
Think about the physics of it. Using hot air around and over the case of chip to desolder or solder it is just like putting it into a reflow oven. It will use up one of the reflow cycles the chip is rated for, and usually they are rated for 1 to 3 cycles. I've never seen higher. Also any moisture gain the chip had will need to be countered if the case is moisture sensitive. So baking out the moisture is needed before soldering or desoldering with hot air. Thermal shock and differential temperatures in the case is also an issue. It is best to warm up the whole board first. That way the temperature difference between the top and bottom of the case is less when it is reflowed.

The reflow oven temperature curves all have a more gentle warm up and soak period before a relatively fast rise to solder melting temperatures, then cool down after. The time spent at solder melt temperatures is an issue. Often it is limited to only 15 to 45 seconds.

As for baking out, it can be done at various temperatures, and that effects how long it takes. I saw in one TI doc that 40C was the maximum bake out temperature for parts in reals, and trays. Sometimes the manufacturer will have different reflow curves and maximum temperatures for different moisture exposure levels.

This TI doc show the requirements for their power modules. It even has an overall time limit for start to end of reflow soldering cycle.
http://www.ti.com/lit/an/slva840/slva840.pdf

Then there are the temperature handling issues of the other parts on the PCB. Lithium batteries are something that can't even be heated to bake out temperatures, let alone reflow temperatures. So they need to be removed before baking out the moisture. They also need shields to protect them if using hot air near them.
 

Offline Yansi

  • Super Contributor
  • ***
  • Posts: 3930
  • Country: 00
  • STM32, STM8, AVR, 8051
Re: Doesn't happen every day.. IC package destroyed while hot air soldering.
« Reply #16 on: September 06, 2018, 06:42:34 pm »
Me and my friend have once been dismantling an old 48 port ethernet switch, removing all interesting components from the board. There was a metric shit ton of synchronous RAM chips and when heating one of these, it popped right in half with a hole in the middle.  :-DD Never experience this before. We were baffled by that.
 

Offline exe

  • Supporter
  • ****
  • Posts: 2647
  • Country: nl
  • self-educated hobbyist
Re: Doesn't happen every day.. IC package destroyed while hot air soldering.
« Reply #17 on: September 06, 2018, 07:56:39 pm »
> they were kept inside the tape, inside the antistatic bag.

That's a bummer, I store all my stuff almost like you and hoped I was protected from pocorn problem. Worse yet, I thought it's enough to bake just for 1h at 90C, didn't know it might not be enough.

I wonder if IC can be damaged this way, but continue to work somehow (and fail or malfunction later). Perhaps, this is possible.
 

Offline CatalinaWOW

  • Super Contributor
  • ***
  • Posts: 5569
  • Country: us
Re: Doesn't happen every day.. IC package destroyed while hot air soldering.
« Reply #18 on: September 06, 2018, 08:10:09 pm »
ICs can be damaged this way and continue to work for a while, but it is relatively rare.  You were lucky in a way that one of them blew right open and made the diagnosis obvious.  It is pretty common for popcorning to kill the chip without visibly damaging the package.  Diagnosis can be a real head scratcher if you don't have a chip failure diagnosis lab at your disposal.

Anti-static bags often provide no meaningful seal once opened.  Even the ones with a ziplock fastening can leak like a sieve.  A dessicant bag can help, but can saturate fairly quickly.  A lot depends on your environment. 

Manufacturers usually provide a bakeout procedure for the packages they use.  In some cases it may vary from the JEDEC standard.
 

Offline JS

  • Frequent Contributor
  • **
  • Posts: 947
  • Country: ar
Re: Doesn't happen every day.. IC package destroyed while hot air soldering.
« Reply #19 on: September 06, 2018, 08:49:38 pm »
Could this be a problem while soldering with an iron?
While reworking on a PCB, for removing a dead IC with hot air, what should be done to prevent this from happening to the surrounding parts? Bake the PCB before reworking? What about hot air reflowing, with an oven or a reballing process?

I was aware of this, I just don't know how to deal properly in repair process, a lot of info about this for manufacturing process but I haven't seen much about reworking...

JS

EDIT: just found this http://www.ti.com/lit/an/slva439a/slva439a.pdf
« Last Edit: September 06, 2018, 09:22:25 pm by JS »
If I don't know how it works, I prefer not to turn it on.
 
The following users thanked this post: CatalinaWOW


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf