Printed Circuit Board Assembly & PCB Design SMT Electronics Assembly Manufacturing Forum

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


no-clean process

p. palshikar

#14136

no-clean process | 23 September, 1998

i am working on a no-clean process implementation project in a company. i want to change a board to no-clean which had dendrite problems underneath the small resistors on the bottom side(in smt).i want more information on the exact cause of dendrites and whether no-clean is the solution.

reply »

Dave F

#14137

Re: no-clean process / Got Dendritus Eh? | 24 September, 1998

| i am working on a no-clean process implementation project in a company. i want to change a board to no-clean which had dendrite problems underneath the small resistors on the bottom side(in smt).i want more information on the exact cause of dendrites and whether no-clean is the solution. | Dendrites are metallic fibers that grow (are plated) between component leads. They can cause short circuits between these leads. Here�s how it works. Component leads can have different direct current (DC) potentials due to routine operation of the circuit. This potential difference makes one lead the cathode and the other the anode � just like in a battery or plating tank at a PCB fab shop. The plating process is based on ionic movement. And ionic movement requires water (or someother solvent) to convey the ions on the cathode to the anode. The water can come from moisture (humidity) in the air or the environment where the board is used. Most cathode / anode connectors pairs either aren�t close enough together, don�t have the electro-potential, or don�t have enough moisture to grow dendrites. BUT if you throw-in some ionic contaminants from flux residues on the surface of the board, you really increase your chances to get some dendrite action. Try an experiment suggested by Howard Manko: 1 Select two closely spaced pads on a clean, bare copper board. 2 Solder a wire to each pad, leaving copper on the edges of the pads that are closest together. 3 Clean the board of all ionic contamination (good luck). 4 Connect the wires to a 10VDC source. 5 Put the board under a microscope, so that you can see the pads. 6 Put a drop of "pure" water between the pads. 7 Wait 10 minutes and look at the space between the pads � whala dendrites. Neat huh? In answer to your question, "will changing to a no-clean flux cure my dendritus?" � Good question. It depends on the cause(s) of the dendrites. If they were caused by: 1 Only poor cleaning of flux residues, then changing to a no-clean flux will probably help. 2 Only the layout of the pads, changing the flux will not help. 3 Only the moisture where the board is used, changing the flux will not help. 4 Multiple causes, changing the flux may or may not help. You have an opportunity to solve a process/design problem. Changing to no-clean is not a magic bullet. It will affect every solder joint in that process, not just those with the dendrites. It requires greater attention to process detail than water soluble or RMA. You may cure your dendritus and trash your yield due to poor wetting. If you are not prepared for the change in focus required when using a no-clean flux, you could be jumping from � CSL, Trace, and Robisan are among the better labs that do contamination studies. Good luck Dave F

reply »

Graham Naisbitt

#14138

Re: no-clean process / Got Dendritus Eh? | 26 September, 1998

As an addendum to Daves posting, please consider that: There is no such thing as no residue fluxing. A no-clean must herefore, have benign residues - in other words, they are not as efficient at removing surface oxides to enable good solder joints. So successful no-clean processes usually rely upon fresh substrates and components - if not, then you have problems. Now, imagine you have a paint for the ceiling with 15% solids and you are then given a similar paint with 2% solids for the same job. Will you have a mess? Probably but, they add wetting agents to help the product stick. Plus, they have all that extra liquid. So you need higher pre-heat and that makes the surface expand and become even more absorbant than usual so the extra liquid gets taken up throught the resist into the subsrate. PLUS: The surfactant additives (wetting agents) have very low surface energy, so they penetrate very easily through such an increasingly porous surface. Are these contaminants? Maybe. Try IPC-ANSI-J-STD-001B Appendix D testing using new SIR test methods, and you can define your process and resolve this issue. No-clean or OA? My guess, for high rel products, OA NOT no-clean. Regards, Graham Graham.Naisbitt@concoat.co.uk | | i am working on a no-clean process implementation project in a company. i want to change a board to no-clean which had dendrite problems underneath the small resistors on the bottom side(in smt).i want more information on the exact cause of dendrites and whether no-clean is the solution. | | | Dendrites are metallic fibers that grow (are plated) between component leads. They can cause short circuits between these leads. | Here�s how it works. Component leads can have different direct current (DC) potentials due to routine operation of the circuit. This potential difference makes one lead the cathode and the other the anode � just like in a battery or plating tank at a PCB fab shop. The plating process is based on ionic movement. And ionic movement requires water (or someother solvent) to convey the ions on the cathode to the anode. The water can come from moisture (humidity) in the air or the environment where the board is used. | Most cathode / anode connectors pairs either aren�t close enough together, don�t have the electro-potential, or don�t have enough moisture to grow dendrites. BUT if you throw-in some ionic contaminants from flux residues on the surface of the board, you really increase your chances to get some dendrite action. | Try an experiment suggested by Howard Manko: | 1 Select two closely spaced pads on a clean, bare copper board. | 2 Solder a wire to each pad, leaving copper on the edges of the pads that are closest together. | 3 Clean the board of all ionic contamination (good luck). | 4 Connect the wires to a 10VDC source. | 5 Put the board under a microscope, so that you can see the pads. | 6 Put a drop of "pure" water between the pads. | 7 Wait 10 minutes and look at the space between the pads � whala dendrites. Neat huh? | In answer to your question, "will changing to a no-clean flux cure my dendritus?" � Good question. It depends on the cause(s) of the dendrites. If they were caused by: | 1 Only poor cleaning of flux residues, then changing to a no-clean flux will probably help. | 2 Only the layout of the pads, changing the flux will not help. | 3 Only the moisture where the board is used, changing the flux will not help. | 4 Multiple causes, changing the flux may or may not help. | You have an opportunity to solve a process/design problem. Changing to no-clean is not a magic bullet. It will affect every solder joint in that process, not just those with the dendrites. It requires greater attention to process detail than water soluble or RMA. You may cure your dendritus and trash your yield due to poor wetting. If you are not prepared for the change in focus required when using a no-clean flux, you could be jumping from � | CSL, Trace, and Robisan are among the better labs that do contamination studies. | Good luck | Dave F

reply »

Void Free Reflow Soldering

SMT spare parts - Qinyi Electronics