Soldering to different Materials

[gogoman]

Hello all

I noticed that headers are available on the solder side as either gold or tin. Since solder is tin, is it preferred to chose tin over gold or does it matter.
If it does matter why is this the case?   

The headers are through hole and ~20mil diameter and mount to a PCB/FR4, standard thickness. 

thanks for your input

gogo

[Someone]

How fancy do you want to get?
https://www.circuitinsight.com/pdf/reliability_microstructure_lead_free_joints_ipc.pdf

Most DIY/hobbyists are happy once the joint flows and don't care about such details.

[gogoman]

Then it could be a problem, but I suspect for a header with 100mil spacing it may not be

[Siwastaja]

Doesn't really matter. I have used both. Contract manufacturers are fine with either and we don't see any reliability issues from these kind of choices.

[David Hess]

Are the headers tin or gold on both sides?

The gold plating is very thin and is intended to completely dissolve into the solder so that it will not present any problems.  Either tin or gold plating could be used, but often gold plating is used on the mating surface for compatibility with whatever the header will be connected to.  The mated surfaces of mechanical connections should be either gold or tin but not both for best reliability.

[tooki]

Hello all

I noticed that headers are available on the solder side as either gold or tin. Since solder is tin, is it preferred to chose tin over gold or does it matter.
If it does matter why is this the case?   

Gold plating on surfaces to be soldered is simply as a solderability preservative; gold won’t oxidize and tarnish like most other metals can, including tin.

However, it’s more expensive, and gold weakens solder, so in high-reliability applications like aerospace and military, gold plating must normally be removed before soldering. (Typically, by tinning with solder, then sucking off that solder, or in a solder bath.) For ordinary purposes, totally unnecessary.

Anyhow, where caution is warranted for the hobbyist is with cheap “gold plated” headers from eBay/aliexpress: some of them use something as fake gold plating, and whatever that something is, it’s not solderable!! I had to throw away a whole package of such headers because they were functionally useless, as it was impossible to get solder to stick to them, even with aggressive flux. I haven’t had any trouble with tin-plated eBay/AliExpress headers, but these days I just get them from DigiKey.

[David Hess]

However, it’s more expensive, and gold weakens solder, so in high-reliability applications like aerospace and military, gold plating must normally be removed before soldering. (Typically, by tinning with solder, then sucking off that solder, or in a solder bath.) For ordinary purposes, totally unnecessary.

That should not be required for flash gold plating like this because the small amount of gold will completely dissolve into the solder and never dilute it enough to form brittle intermetallics.

What you say applies to heavy gold plating and I have experienced this first hand.  In this case, the gold must be removed before finishing the solder joint.  If it is not, then the boundary between the gold and tin based solder will form brittle intermetallics as the gold and tin diffuse and eventually crack and separate.

This is also why connectors should have either gold plating or tin plating but not both.  Contacts with gold on one side and tin on the other are much more likely to fail over time.  This was a common problem with DRAM connectors where the SIMM or DIMM had tin plated fingers and the socket had gold plated contacts.

[helius]

This is also why connectors should have either gold plating or tin plating but not both.  Contacts with gold on one side and tin on the other are much more likely to fail over time.  This was a common problem with DRAM connectors where the SIMM or DIMM had tin plated fingers and the socket had gold plated contacts.

You expressed this point more clearly in Reply #4; it's common on many types of connectors for the mating surfaces to be gold plated and the solder tails to be tin or tin/lead plated. This is not a problem and is preferable to all-over gold or tin plating.

The problem is when connectors are mismatched, where a gold-plated female surface is mated to a tin-plated male or vice versa. The different rates of thermal expansion and different electronegativity lead to fretting corrosion in the presence of moisture or corrosive gas. If mismatched connectors must be used, it's important to protect them from fretting by using a lubricant containing polyphenylether.

[David Hess]

You expressed this point more clearly in Reply #4; it's common on many types of connectors for the mating surfaces to be gold plated and the solder tails to be tin or tin/lead plated. This is not a problem and is preferable to all-over gold or tin plating.

I have actually never seen that except for machined collet socket pins where the finger assembly inside the socket pin is a separate structure.

[tooki]

You expressed this point more clearly in Reply #4; it's common on many types of connectors for the mating surfaces to be gold plated and the solder tails to be tin or tin/lead plated. This is not a problem and is preferable to all-over gold or tin plating.

I have actually never seen that except for machined collet socket pins where the finger assembly inside the socket pin is a separate structure.

Really?? Selective gold plating is extremely common nowadays. (Many connector contacts that used to be gold all over have been transitioned to selective gold.)

[nctnico]

You expressed this point more clearly in Reply #4; it's common on many types of connectors for the mating surfaces to be gold plated and the solder tails to be tin or tin/lead plated. This is not a problem and is preferable to all-over gold or tin plating.

I have actually never seen that except for machined collet socket pins where the finger assembly inside the socket pin is a separate structure.

I have seen that plenty of times. I even have a whole bunch of brand new 0.1" headers in my stock which are tinned at the solder tails. Nothing special about these headers though.

[coppercone2]

I can see it causing problems if its not washed right after plating or tinning.


if there is a further problem IDK, it might be burried some where. I would love to hear details about that one.


I imagine if true, it has something to do with modified assembly line processes? I.e. fast assembly line plate related problems. might work for 1 bath but 2 baths it starts showing holes?

[tooki]

Given that probably billions of selectively-plated contacts are manufactured each year, and have been for ages, I suspect the processes are thoroughly understood and debugged.