Nancy, you risk to drowned by information. Scientifically spoken it's a complex thing and best you can do is 1)buy a nice little book from SCI in Sweden called 'Corrosion of Electronics' bulletin No 102. Fax +46 8 16 72 70 2) buy a 5kg book about migration and corrosion from ASM, book n:o 13, from that book you will be corrosion professor.
Now, I would like to answer somewhat different from practical experience in our field, instead of trying the scientific way, other will do that.
1. What will cause Electrical Migration?
Think of spots where different metals are exposed to Voltage, Humidity and Contaminations. Sometimes without addition of 'outer' voltage. Simple example: a ceramic chip capacitor, tinned terminals, silverepoxy mounted on gold plated copper pads, dielectric is the FR4 board itself. After soldering process, acidous flux residues are left under the capacitor and the board is exposed to some humidity (e.g. condensation). In a few seconds silver atoms will creep between the capacitor's two terminals, especially with some DC volts difference. The atoms will build a nice 'tree', looking like snow crytals, and this electrical bridge under the capacitor can cause electrical problems.
How to avoid that kind of migration? The answer is, that there are no short ways. You have to get some understanding about the why's and how's. In my situation I would try some of following recipes:
a) MINIMIZE THE GALVANIC VOLTAGE DIFFERENCE (see these words), which means to combine metal systems that don't cause a too strong 'battery' effect. In my bad example we have a horrid scenario with noble and non-noble metals close to or mixed one with the other: Kopper, Tin, Lead, Silver, Gold etc.
b) GO FOR ABSOLUTELY BEST CLEANING after mounting (soldering), leave your boards for SIR-test(see these words) for get knowledge about your board status.
c) AVOID THE COMBINATION HIGH VOLTAGE/LITTLE SPACE, such countermeasures can eliminate migration totally.
d) AVOID CONDENSATION TO TAKE PART, small droplets which penetrate under components and mix themselves with flux residues (and other contaminations) are not easily removed! You can heat the board, blow with Nitrogen, do anything, and the little water amount is still there, hidden in the microstructure, seemingly harmless....until you switch your Highvoltage Power Supply on!
e) If the board is very (VERY) clean, and you have to make your product withstand wet atmosphere, there is still solutions, HUMISEAL, PARYLENE or other of the numerous coatings that you can spray on, or dip, or pot etc.
Nancy, that was one example. Migration can show up in many more forms. I have got MIL-spec'ed expensive, cadmium protected connectors in my hand, connectors changed to something that looks more like sand or dust, all that was left! All sorts exist. Best you can do is get a little basic about the phenomenon and get one experienced consulting, to solve your specific problem. Don't do anything without understanding! Can get you into real trouble!
Didn't I answer your question? Sorry, at least I tried.
Good Luck.
Ingemar Hernefjord
Ericsson Microwave Systems
PS. If no one in your area can help, we have people with decades of such experience, can give you ids, but shouldn't be necessary, there are with certainty hundreds of companies in the US that can help you.
--------------------------------------------------------------------------
Hi Every One.
I need information on Migration
Question
1. What will cause Electrical Migration
Reason :
We are seeing Dendritic Migration between components
As of now we are applying conformal coating trying to stop the migration.
Question:
1. Is there a rework to prevent this condition.
2. Any study or data out there on reliability as far as using this method
to stop the migration.
##############################################################
TechNet Mail List provided as a free service by IPC using LISTSERV 1.8c
##############################################################
To subscribe/unsubscribe, send a message to [log in to unmask] with following text in
the body:
To subscribe: SUBSCRIBE TECHNET <your full name>
To unsubscribe: SIGNOFF TECHNET
##############################################################
Please visit IPC web site (http://www.ipc.org/html/forum.htm) for additional
information.
If you need assistance - contact Gayatri Sardeshpande at [log in to unmask] or
847-509-9700 ext.5365
##############################################################
|
