I looked at the skillet PCB technology and I am curious

Hi,

I looked at the skillet PCB technology and I am curious – don’t the ICs get fried by the high temperature that they are continuously exposed to during the baking process?

In most spec sheets I read about exposure in the domain of few seconds so pretty much when soldering by hand one can’t sneeze without some $19 part getting fried. How does that apply to the skillet process?

Thanks

~B

Because when you’re hand soldering the heat is directly on the IC. When you skillet it the board is heated up and he IC doesn’t get the heat directly.

TSPRAP:
Because when you’re hand soldering the heat is directly on the IC. When you skillet it the board is heated up and he IC doesn’t get the heat directly.

Cool thanks, does this also apply to rework station (hot air) meaning its still OK to heat just as in the skillet scenario?

Thanks

Boyan

Boyan,

The skillet approach is totally safe. I can’t believe I used to hand solder anything before now. I have a hot air reflow from SFE as well. Although good with IC’s large 44 pin tqfp packages are difficult but not impossible with only the hot air gun. For small resistor and caps the hot air gun is a snap. But if you are laying out a whole board why not heat up the whole pcb and “zap” all your parts at once? Recommend checking out www.zeph.com for solder paste and some other interesting tutorials.

My $0.02 - the mfgs who write the datasheets are just covering their asses for production runs.

You can do some really evil things to ICs (over heat, super cool, run too fast, etc) without too much problems. I’ve over clocked PICs to twice their spcd speed without damage. I’ve run 5V parts at 3.3V without problems. But these are just flukes that I do with non-production stuff. So you can solder to IC’s and do a mean job of it, and still get the IC to work. But for production runs (say 1000pcs for example), the mfg has to be able to stand behind their product, so they’ll limit what stresses you can put on the IC and still get good yields. I’ve never really looked at the maximum heating per pin/per second ratings - I just solder it up.

Blah, blah, blah. The Skillet rules. Hot-air rework is the only way to fly (if you need to salvage an IC or a PCB). Good luck!

-Nathan

ohararp:
The skillet approach is totally safe. I can’t believe I used to hand solder anything before now. I have a hot air reflow from SFE as well. Although good with IC’s large 44 pin tqfp packages are difficult but not impossible with only the hot air gun. For small resistor and caps the hot air gun is a snap. But if you are laying out a whole board why not heat up the whole pcb and “zap” all your parts at once? Recommend checking out www.zeph.com for solder paste and some other interesting tutorials.

Thanks everyone,

After few trial runs I am convinced - I was able to harvest a beutiful 208 pin package by 3DFX (some kind of Asic) and it looks like it has never been used before - no little solder fragments and no pin breaching, take a look:

http://www.gokrystal.com/images/DSC00832.JPG

http://www.gokrystal.com/images/DSC00833.JPG

I think there are a lot of things you can do once (or twice or just a few times) to an IC, but if you do it repeatedly you’ll get some sort of thermal-cycling failure and so the data sheet says don’t do it. But if you look at the same data sheet’s recommended reflow profile, sometimes it’ll show the temperature going outside the “absolute maximum ratings”.

Also, as a hobbyist I’m willing to accept a 1-failure-in-100 risk from mistreating my chips, but that’d be an unacceptable failure rate on a production line (since a board may have many chips…)