Accurately locating SOT-23 parts

Come one now, never heard of wireless?

Jamie

I just checked the NXP spec for sot23 all they spec is a maximum of 0.1mm

-Lasse

That would make sense with the footprint Joerg linked before (.7mm wide pad).

Den onsdag den 18. december 2013 02.16.51 UTC+1 skrev snipped-for-privacy@attt.bizz:

the 0.1mm is max distance (height) from bottom of pin to bottom of package, how does that relate to the pads size?

-Lasse

It doesn't. I have examples right here.

Nope, it all woiks. Of course, this will never happen in a regular high-tech assembly house but in the world of consumer goods every penny counts.

If I understand what you're saying, the resulting joint will consist of just a layer of solder between the pad and the pin. It seems to me that that will result in stress concentration around the edge, with an increased likelihood of a fatigue failure of the joint.

Sylvia.

Nonsense. If you place a .5mm leg on a .7mm pad, with a .5mm offset, the leg *IS* off the pad. Fix your process.

Absolute nonsense. You'd never get any yield from a process that crappy.

As I said, I have examples. Not that I'd advocate such sloppy placement but for stuff that isn't mission-critical one can see it on occasion.

Anyone thinking they can make money making junk like that is crazy. "Mission critical" has nothing to do with it.

Go to [insert just about any kids toy department here], buy some of the

Not buying it. *Anything* with small parts that has manufacturing tolerances looser than the lead sizes isn't going to be manufacturable. Sorry, not buying your fairy tales.

Only if their process isn't as bad as you claim.

Looked at one of my high-volume boards (assembled by low bidder in China) and the SOT23s are all centered quite well, but about 30% are rotated slightly. They're actually EIAJ SC-59A, which is pretty close to an SOT-23.

I'm talking about reflow soldered boards here- wave soldered boards use different optimal pad sizes, and have a specific orientation of the footprint relative to motion through the wave.

--sp

As I said, you are welcome to swing by and I'll prove it to you. In fact, one board with out-of-aligment SOT23 is from a very large production volume cordless phone.

See above.

I have commercial high-volume boards here where SOT23, SC75 and similar parts are so far out of alignment that sections of some pins are off the pad. Yet the solder connection is solid. Keith may not believe it but that's life.

Another thing I notice a lot is MELFs having rolled 0.5mm or so out of center. It is one reason why I do not use that package in my designs.

Yes. All that's needed is to get things close enough for surface tension of the molten solder to take over. The final position will be where those forces came to balance.

Joe Gwinn

Yes, physics is on our side, for once. No, I don't believe Joerg's assertion that people actually (for profit) boards with processes so sloppy that they can't even get the parts on the pads. Their waste must be astronomic if they're that bad. It must be hard work to make things that badly.

Den torsdag den 19. december 2013 01.23.27 UTC+1 skrev snipped-for-privacy@attt.bizz:

could be manual mounting done in a rush

-Lasse

Toys? They aren't making any money there, even with Chinese labor.

Think about "why (just) around the edge"? Methinks whatever stress you are thinking of,it would be roughly uniform; the area is so small..