The stain on the conductive traces appears to be an acidic or alkaline liquid that was dripped and left onto the traces.

Question, did you use selective soldering on this assembly? If so, were you using a water soluble flux?

This contamination appears to be due to particles of solder paste being present across the surface during reflow. These have melted and spread to cause these patches of solder.

Perhaps this has been a badly printed PEC which was not washed off properly.

Solder seems to have been trapped in the vias. During subsequent reflow, the solder inside the via will melt and can outgass causing it to spatter and fall on the gold pads.

Without more details, it is difficult to assess for certain what the issues are. It looks as though solder balls had reflowed onto the Au areas.

My first action would be to look to see if these boards were washed due to misprints. Misprint cleaning often leaves stray solder balls which could coalesce when reflowed later.

The second thing I would look at is the printing process and, most importantly, the under stencil cleaning process. The under stencil cleaning can sometimes smear solder paste across the bottom of the stencil allowing the solder to contaminate subsequent boards.

In both of these cases, the solder particles are very difficult to see prior to reflow.

After reflow, they will wet to the Au surface and become readily apparent. The third area to look at would be solder spatter during the reflow process.

A slow ramp in the preheat stage of the reflow process will help to slowly volatilize the flux and minimize the potential for splattering.

Unusual contamination?

The photos show what appear to be contamination below the soldermask from the etch process not properly rinsed. The pockets of contamination (typically sulfuric acid) rarely show up as an incoming board but run it over a wave or selective soldering process and these corrosion attacks of the base copper below the soldermask or occasionally through pin holes in the mask will appear.

I have attached a presentation on this issue. Please contact me and we can prove what it is and how to correct it at the fabricator.

It looks like a splatter of something on the underlying metal that kept the (what appears to be) gold plating from seeding and plating.

Note that there is no splatter on the substrate outside of the circuit so whatever it was washed or dissolved off subsequent to the plating process.

Perhaps an analysis of the material that appears in the center of the discolored areas might solve the mystery.

In my opinion the culprit appears to be lack of process detail just before gold plating

Unusual contamination indeed!

The most obvious characteristic of this contamination is its linearity. Next, it apparently stops and starts where the mask begins.

It looks to be corrosion of some kind ~ but that is a difficult call to make from an image. Under high magnification you may be able to better see what you have. Keep in mind, that gold is very resilient, and that chances are that “corrosion” is also present beneath the mask.

Linear contamination theoretically could come from anywhere on the production line. It could be a dripping action, however, as it appears to end and begin anew as the mask comes and goes it strikes me that the source is not likely to be in the "downstream" assembly process.

Do you have any other lots of these substrates to compare downstream processing versus as received? I suspect your root cause may be well upstream of assembly with your substrate suppler.

It is impossible to figure out what this is without more data. It looks like moldew or mold, or corrosion caused by splashes of something.

My suggestion is to get an analytical lab, equipped with an IR microscope to take a look at the spots on the board. The resolution can be very high. Also need to figure out if it's "organic" or inorganic. Just need more data.

Also, would be helpful to know stirage history, manufacturing process steps, etc. Sorry I can't be more helpful...but a photo is just not enough.

Again, I'll be happy to chat with the person, gratis to help solve the problem.