Solder Paste with/without silver

Responding to your questions ...

RELIABILITY OF 1% SILVER SOLDER PASTE: It�s as good as the next solder paste, all other things the same. Look at your back articles by Jennie Hang in SMT magazine. I want to say she wrote an article on the topic in the past year. HELP WITH 0402 TOMB-STONING: None. Search the fine SMTnet Archives on tomb* and 0402 EUTECTIC SOLDER PASTE SnPb USE: Most people do not use solder pastes with silver. We�re used Sn62 since the beginning of time. We think it's a little shinier.

Expanding further ...

From "Soldering in Electronics" by Klein Wassink:

"Note on solder alloy composition: Solder paste in hybrid circuit technology (for thick-film circuits) usually has a metal composition of tin62-lead36-silver2. It should be realised that the addition of silver is not at all necessary for the silver (-palladium) metallisation of the components, but for the much thinner silver-palladium conductors on the thick-film substrates. The use of the more expensive silver containing alloy, instead of the common tin60-lead40 alloy, for (reflow) soldering on printed boards with copper solder lands is not based on technological necessity, but sometimes on better availability of this alloy (in the form of paste) and in most cases merely on habit. For the rest: with silver loaded alloy no harm is done !"

Which is better - Sn62 or Sn63? You�ll find two very distinct opinions: 1 No difference between the two solder alloys and they are identical/interchangeable. 2 Sn62 has better wetting characteristics and better thermal fatigue life characteristics.

Metallurgically, the Sn62 alloy has slightly: * Better surface tension values than Sn63 (376 dynes/cm versus 490 dynes/cm). * Lower solidus temperature than Sn63 (179�C versus 183�C). * Higher strength than Sn63, but properly wetted connections of either solder have more than sufficient strength. So this difference is immaterial in reality. * Higher creep resistance than Sn63, but that is only of importance if the temperature cycle dwells are short, as in accelerated testing. Most product experience dwells long enough that this difference is immaterial in reality.

Some equate surface tension & solidus temperature above to improved wetting characteristics.

You will also find two distinct sets of thermal fatigue life data when using Sn62. 1 No improvement. 2 A definite improvement.

It's possible that this thermal data is being influenced by the interaction of the assembly use environment and specific assembly design/construction issues.

The original use of the Sn62 alloy was to reduce the leaching/diffusion problems associated with ceramic chip capacitor end terminations. Using nickel as a barrier plating over copper on chip capacitors terminations has eliminated the need for Sn62 in those cases.

Finally, some believe that using Sn62 is preferred when soldering to gold containing surfaces. The reasoning is that silver reduces: * Rate at which gold is removed. * Amount of gold removed.

I would like to see some evidence showing this has been objectively tested and measured.

reply »