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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Porosity in Good Plating

sam

#25069

Porosity in Good Plating | 4 July, 2003

I need some advice about a problem that happened in a SMT LED:

The failure symphoms of this SMT LEDs is that the LED body itself seemed to be functional, but when soldering on the products, it showed good wetting and soldering, LEDs are bright and light up as normal. However, after 48 hours of storage, some of the LEDs did not light up in the products, some of them became very dim or blinking when used. When we checked these problematic LEDs by multimeter, they were still bright as good LEDs.

we had done some experiment and examination and reaching a conclusion that the poor SMT surface was cause by Gold plating contaminated, and Nickel exposed to the plating surface.

My question is 1. whether this is named Pososity. 2. What lead to such contamination. 3. How it will affect the finished products after soldering 4. Can such contamination be cleaned? 5. If it can be cleaned, what chemical shall be used.

Thanks Sam

reply »

ramanandkini

#25072

Porosity in Good Plating | 4 July, 2003

I have a CEM-1 PCB board with 35 microns copper. This is PCB is finished with SMOBC+HALS. We do solder SMT LEDs, but till date we do not have problems that you have listed. The LEDs are good & bright. It even passed high & low temp cycle tests (-40 to +105 deg.c). It also passed high humidity tests (90%). Both the tests were performed with the circuit in powered condition.

I suspect the chip resistors that supply current to the LEDs. Check the wattage of these chip resistors. Check the forward current that goes in to the LEDs. You might have already checked for shorts in the circuit. Check if your board is bending during assembly. To avoid moisture condensing & creating shot/reistance,You need to have a conformal coating on the soldered leads.

Clean the Boards throughly in an ultrasonic cleaner and see the sodlered pads under a high zoom microscope to see if it is surrounded with solder balls. Presence of solder balls, can lead to interrupted electrical conncetion resulting in loose contact. Parallelly, you can manual solder some LEDs in one sample board and check after 48 hours. If this manual sodlered LED is OK, then you wave soldering/reflow process has to be improved.

Check the board in the powered condition. Is it heated up too much with all the LEDs? If, so try to reposition them on teh circuit board (change the layout, move the LEDs away from each other). Are you passing high forward current to brighten up the LEDs? Try high wattage Chip resistors. In my experience, just increasing the forward current to 50%, the resulted brightness is not all that claimable.

If these are OK, then, you can also suspect the presence of hair line crack in copper resulted from etching conditions, deep scratches or poor handling.

reply »

AlCapone

#25075

Porosity in Good Plating | 5 July, 2003

If the problem as you described as the gold plating contaminated, and the Nickel layer exposed to the solder pot. However, the purpose of the gold layer is to protect the nickel and copper layer not turn into corrosion and resist to other issue (humidity, temperature, etc..); if the gold layer is washed away, the tin itselft very active and can possibly eat up your nickle and copper layer just within a short period of time, make the circuit itself can't function properly. Before going to the ICT, I suggest you perform the ultrasonic wash as soon as the board come off the line, If this doesn't help, then call the bare board supplier to tackle this issue. I hope this will help you. good luck

This message was posted Add this forum to your site! Click to learn more. the Electronics Forum @

reply »

MA/NY DDave

#25077

Porosity in Good Plating | 6 July, 2003

Hi Sam,

Great answers so far.

Let me take a stab at a few of your questions.

Q1. whether this is named Pososity?

I can't tell from your description. Several companies and the IPC have porosity tests to see if the Gold has good coverage of the underlying metallurgy. Pores (Holes) and such are found sometime with reactive chemicals and dyes.

Q4. Can such contamination be cleaned? Q5. If it can be cleaned, what chemical shall be used.

This, in general, is a job for the PCB shop with some agressive chemicals. For cost and difficulty reasons they will often chose to build new PWBs rather than cleaning and then re-plating. --------

Can you run the LEDs with hand held tools. I wonder if the LEDs might not be compatible with your process or maybe something else is draining the voltage?

YiEngr, MA/NY DDave

This message was posted Add this forum to your site! Click to learn more. the Electronics Forum @

reply »

ramanandkini

#25078

Porosity in Good Plating | 6 July, 2003

I wish to inform you that our boards are reflow soldered with Infineon (OSRAM) make SMT LEDs.

reply »

sam

#25080

Porosity in Good Plating | 7 July, 2003

Thanks for all the answers.

I should clarify that the problem was found to be on the SMT LED gold plating, instead of PCB gold plating. We had confirmed this problem by running some DOE experiments and some reliability testings.

We had decided to dis-continue the use of this problematic LEDs, but the failure sympthom was still a question that why the poor gold plating in LED could lead to a intermittent failure AFTER around 48 hours. The LEDs were 99.5% bright and gold immediately after assembly and functional testing. But failure rate were up to 30% after 48 hours and were still moving up.

Dear AlCapone, It seemed that the Tin-Nickel chemical result like what you stated might be the root cause of the problem. Can you tell me more about it? What will be the end-result of this chemical reaction? Is that leading to some insulated materials?

Sam Hung

reply »

MA/NY DDave

#25094

Porosity in Good Plating | 7 July, 2003

Hi Sam,

I am going to guess that a problem is being caused by processing to effect the inside of the LED package.

Send some failed LEDs back to the manufacturer to do a failure analysis. Also ask the manufacturer for info on their SMT processing qualification.

YiEngr, MA/NY DDave

This message was posted Add this forum to your site! Click to learn more. the Electronics Forum @

reply »

#25102

Porosity in Good Plating | 8 July, 2003

AlCapone: We have an understanding similar to yours, but slightly different. * At 230*C gold dissolves in to solder at the following rate: 63/37: 197 uinch/sec [5 um/sec]. So for a typical ENIG solderability protection, the gold is gone in about a second. * Soldering occures at the nickel surface. Generally, this plating is quite thick [150-250 uinches of nickel]. So, the nickel will not dissolve completely in the solder, but will act as a barrier between the solder and the copper pad.

reply »

#25103

Porosity in Good Plating | 8 July, 2003

Sam: If your gold overcoat was pourous, the nickel undercoat would corrode prior to your attempting to solder and you would not be able to solder without an aggressive flux. Since you see solder flow and wetting, it indicates the gold did an adequate job of protecting the solderability of the nickel and was NOT pourous.

reply »

sam

#25110

Porosity in Good Plating | 9 July, 2003

When we try another test by using less solder paste, we found under the microscope that the solder was actually stick on the LED gold surface, but not having good wetting. By this finding, we can basically conclude that it was poor solderability caused by the Nickel Oxidation on the plating surface.

It was really a danger situation that this type of poor soldering could not be normally observed by magnifying glass.

Thanks all for the advices Sam

reply »


RDR

#25111

Porosity in Good Plating | 9 July, 2003

Sam, You may want to be careful with your analisys, reducing paste volume also reduces flux volume. Flux is there to promote and aid in wetting. by reducing flux you may have initiated the poor wetting yourself with the experiment.

Just another thought on the situation. You may want to apply liquid flux to that LED that you soldered with less paste and reflow it again then check and see if you still have wetting problems.

Russ

reply »

sam

#25172

Porosity in Good Plating | 15 July, 2003

Thank you Russ for your comment.

This failure cause should have been confirmed as contaminated plating surface causing poor solderability and conductivity, according to all the testing and observations.

The only question remaining to me is that why the failure happened after 48 hours instead of immediate failure after soldering.

Sam

reply »

#25179

Porosity in Good Plating | 15 July, 2003

Sam says, �The only question remaining to me is that why the failure happened after 48 hours instead of immediate failure after soldering.�

Let�s recap: * Had good-looking solder flow. * Passed the initial electrical test. * Failed the later electrical test for poor solder connection with components that later tested OK. * Works fine with alternate parts.

If all of that was true, something must be incorrect. Consider: * Solder connection WAS good, but someone, like those guys [with purple, well blueberry hair] over in test, messed it up. * Solder connection looked good, but that was only an illusion [and as Joe Cocker sings, in the old Dave Mason song, �I�m not feelin� too good myself.�].

� but since the alternate parts work fine, it probably wasn�t the guy over in test. So, it must have been marginal solderability of the first set of parts and some flexing and handling of the boards between the two tests broke the parts loose. Further , this is supported by the intermittance of some the parts in the post storage testing. OK, here�s a bet. If you twist the parts from the board, parts from: * First supplier come off relatively easy and leave a smooth solder interface on the board. There'll be little solder on the component. * Second supplier come off relatively hard and leave a rough solder interface on the board. There'll be similar amounts of solder on the component and the board.

reply »

sam

#25195

Porosity in Good Plating | 16 July, 2003

Dave,

I believed you are really experienced in SMT soldering process.

What I meant good connection, should really be "Visual check by naked eye OK".

From your message, there are situation of "marginal solderability" that will lead to uncertainty and intermittent of the product, right?

Sam Hung

reply »

Plasma Prior to Conformal Coating

PCB, BGA Rework Services